Extinction of Conditioned Fear in the Developing Rat

G G

EXTINCTION OF CONDITIONED FEAR IN THE DEVELOPING RAT

Jee Hyun Kim Bachelor of Psychology (Hons)

March 2008

Doctor of Philosophy Thesis

School of Psychology The University of New South Wales PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Kim

First name: Jee Hyun Other name/s:

Abbreviation for degree as given in the University calendar: PhD

School: Psychology Faculty: Science

Title: Extinction of Conditioned Fear in the Developing Rat

Abstract 350 words maximum: (PLEASE TYPE)

The present thesis examined extinction of conditioned fear in the developing rat. In the adult rat, the hippocampus is thought to be important for the context-specificity of extinction. Because the hippocampus is a late-maturing structure, it was hypothesised that context-modulation of extinction may be different across development. The first series of experiments investigated reinstatement of extinguished fear in the developing rat (Chapter 2). The results showed that P24 rats exhibited context-specific reinstatement. On the other hand, P17 rats did not exhibit reinstatement of extinguished fear following a US reminder treatment. The failure to see reinstatement in P17 rats was not due to the reminder treatment being ineffective in these rats because the same treatment alleviated spontaneous forgetting in rat this age. The second series of experiments then examined the renewal effect and GABAergic involvement in extinction in P24 and P17 rats (Chapter 3). It was observed that P24 rats displayed renewal whereas P17 rats did not. Also, pre-test injection of FG7142 recovered extinguished fear in P24 rats but not in P17 rats, even across a range of doses. This failure to see any FG7142 effect on extinction in P17 rats was not due to the lack of responsiveness to this drug in these rats because FG7142 was found to be effective in alleviating spontaneous forgetting in rats this age. The third series of experiments then examined the effect of temporary inactivation of the amygdala on extinction and re-extinction in the developing rat (Chapter 4). It was observed that extinction retention is impaired in both P24 and P17 rats if the amygdala is inactivated during extinction training. Interestingly, when a CS that had been previously extinguished and then re-trained was re-extinguished, re-extinction was amygdala-independent if initial extinction occurred at 24 days of age but amygdala-dependent if initial extinction occurred at 17 days of age. That is, amygdala involvement in re-extinction was dissociated across development. Taken together, these experiments provide strong evidence for fundamental differences in mechanisms underlying fear extinction across development. The implications of the findings were discussed in light of the theoretical and neural models of extinction.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS G G

Abstract

The second series of experiments then examined the renewal effect and

GABAergic involvement in extinction in P24 and P17 rats (Chapter 3). It was observed that P24 rats displayed renewal whereas P17 rats did not. Also, pre-test injection of

FG7142 recovered extinguished fear in P24 rats but not in P17 rats, even across a range of doses. This failure to see any FG7142 effect on extinction in P17 rats was not due to the lack of responsiveness to this drug in these rats because FG7142 was found to be effective in alleviating spontaneous forgetting in rats this age.

The third series of experiments then examined the effect of temporary inactivation of the amygdala on extinction and re-extinction in the developing rat

(Chapter 4). It was observed that extinction retention is impaired in both P24 and P17 rats if the amygdala is inactivated during extinction training. Interestingly, when a CS that had been previously extinguished and then re-trained was re-extinguished, re- extinction was amygdala-independent if initial extinction occurred at 24 days of age but G G amygdala-dependent if initial extinction occurred at 17 days of age. That is, amygdala involvement in re-extinction was dissociated across development.

Taken together, these experiments provide strong evidence for fundamental differences in mechanisms underlying fear extinction across development. The implications of the findings were discussed in light of the theoretical and neural models of extinction.

G G

Table of Contents

Abstract ii

Certificate of Originality vii

Acknowledgements viii

Publications and Proceedings ix

Care and Use of Animals xii

List of Tables xiii

List of Figures xiv

Abbreviations xvii

Chapter 1. Introduction 1 1.1 Pavlovian Conditioned Fear 4 The neural model of conditioned fear: a summary 6 The ontogeny of conditioned fear 11 1.2 Extinction of Conditioned Fear 16 Reinstatement 18 Renewal 20 Spontaneous recovery 21 Pharmacological studies 22 1.3 Neurobiology of Extinction 27 The amygdala 28 The medial prefrontal cortex 33 The role of the hippocampus in extinction of conditioned fear 39 1.4 Contextual Learning in the Developing Rat 42 1.5 Experimental Rationale 43

Chapter 2. Reinstatement of extinguished fear in the developing rat 46 Reinstatement of forgotten fear following infantile amnesia 47 General Methods 52 Experiment 1 58 Experiment 2 62 Experiment 3.1 67 G G

Experiment 3.2 68 Experiment 3.3 70 Discussion 73

Chapter 3. Renewal and the effect of pre-test FG7142 injection on 78 extinguished fear in the developing rat GABAergic inhibition and fear 79 General Methods 83 Experiment 4 86 Experiment 5.1 91 Experiment 5.2 93 Experiment 5.3 96 Experiment 6.1 100 Experiment 6.2 102 Discussion 104

Chapter 4. Amygdala involvement in extinction and re-extinction in the 108 developing rat General Methods 111 Experiment 7.1 115 Experiment 7.2 119 Re-extinction of conditioned fear 122 Experiment 8.1 125 Experiment 8.2 128 Experiment 9 132 Discussion 140 Extinction in the 17-day-old rat and current models of extinction 143

Chapter 5. General Discussion 147 5.1 Extinction is Fundamentally Different Across Development – 150 Theoretical Accounts Extinction may be unlearning in 17-day-old rats 150 Extinction may be ‘deepened extinction’ in 17-day-old rats 153 5.2 Extinction is Fundamentally Different Across Development – Neural 159 Accounts Extinction memory may be stored in the amygdala in 17-day-old rats 167 G G

Extinction may be erasure in 17-day-old rats 168 5.3 Concluding Remarks 175

References 178

Appendices 197 A. Raw Data 197 Chapter 2 197 Chapter 3 204 Chapter 4 210 B. Summary Tables of Statistical Analyses 215 Chapter 2 215 Chapter 3 219 Chapter 4 224

G G

Certificate of Originality

ORIGINALITY STATEMENT ‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

Date ……………………………………………......

G G

Acknowledgements

I thank Jesus Christ, the lover of my soul. You love me despite how nothing in my life is hidden from You. That unconditional love is the source of my unrelenting confidence, courage, and hope. You carried me throughout this journey.

Rick Richardson, you are the most amazing and selfless supervisor in the world. I am deeply grateful of your trust and love. Your passion for research and life inspires me, and without your guidance I cannot be what I am today. My mentor and a dearest friend, I cannot think of a worthy repayment, except by becoming an extraordinary scientist like you are. Thank you for believing that I will be.

To dad, mum, and little brother. Ghab Sik, your life of integrity and excellence will forever be an example for mine. Ah Wha, I really think that you are a genius, and I will always aspire towards your wisdom, faith, and funk. Kyung Whan, what can I say, your hug is all I need to shake off the misery of bad data, zombie bugs, and rats that freeze at baseline. I love you all so much, not only because you are my family, but because you really deserve it.

To Lily Kwon, you are the most loyal and faithful friend in eternity, a gift from God in my life. I know that we will be bestfriends even after we pass from this life to the next one. Thank you, Betty Chang, my confidante, friend, and officemate anyone would be lucky to have. My fair warrior, what would I do without you? I also thank Lan Tran, Amanda Sharpley, Helen Jin, and Nicole Kim for their love and support.

To Julia Langton, my favourite collaborator and a loving friend. Thank you, my omniscient co-supervisor Gavan McNally, Marianne (Super)Weber, Adam Hamlin, Lucy Choi, Fred Westbrook, Glynis Bailey, and Vincent Laurent. I also thank John, Marty, Jonathan, Lynley, Linda, Peter, Helen, Raja, Nisha, Bronwyn, Jean, and Andreas. Also, UNSW coffee-cart boys (especially Tim, Ben, & Mark), I salute you.

To Sydney Full Gospel Church, especially Rose, Sang, Joon, Helline, Jinny, Sylvia, Olivia, Hae Sook, Sung Min, Myong, Chang Geun, Yoon Hwan, Seung Jin, Megan, Pstr Woo Sung Chung, Pstr Sam and his wife Mi-Yun, Elder Seo and his wife Shim, teacher Woo, and my outrageously good-looking students in Worship Cell 2007, especially Esther, James, and Andrew. Thank you all for your car-rides everywhere, food, prayers, and constant affection. G G

Publications

Kim. J. H., McNally, G., & Richardson, R. (2006). Recovery of fear memories in rats: Role of gamma-aminobutyric acid in infantile amnesia. Behavioral Neuroscience, 120, 40-48.

Kim, J. H. & Richardson, R. (2007a). A developmental dissociation of context and GABA effects on extinguished fear in rats. Behavioral Neuroscience, 121, 131-139. This publication forms the basis of Chapter 3 in the present thesis.

Kim, J. H. & Richardson, R. (2007b). A developmental dissociation in reinstatement of an extinguished fear response in rats. Neurobiology of Learning & Memory, 88, 48-57. This publication forms the basis of Chapter 2 in the present thesis.

Langton, J., Kim., J. H., Nicholas, J., & Richardson, R. (2007). The effect of the NMDA receptor antagonist MK-801 on the acquisition and extinction of learned fear in the developing rat. Learning & Memory, 14, 665-668.

Kim, J. H. & Richardson, R. (2007c). Immediate post-reminder injection of gamma- amino butyric acid (GABA) agonist midazolam attenuates reactivation of forgotten fear in the infant rat. Behavioral Neuroscience, 121, 1328-1332.

Kim, J. H. & Richardson, R. (2008). The effect of temporary amygdala inactivation on extinction and re-extinction of fear in the developing rat: Unlearning as a potential mechanism for extinction early in development. Journal of Neuroscience, 28, 1282- 1290. This publication forms the basis of Chapter 4 in the present thesis.

Kim, J. H. & Richardson, R. (in preparation). Expression of renewal is dependent on the extinction-test interval rather than the acquisition-extinction interval.

G G

Presentations

Kim, J. H., & Richardson, R. (2006). Extinction in preweanling rats: The absence of renewal, reinstatement, and FG7142 effects. International Society of Developmental Psychobiology (ISDP) Annual Conference, Atlanta.

Richardson, R. & Kim, J. H. (2007). Developmental differences in extinction of learned fear in rats. Second Special Interest Meeting on Fear and Learning, Le Lignely, Belgium.

Kim, J. H. & Richardson, R. (2007). To fear or not to fear: Erasure as a potential mechanism for extinction in the developing rat. ISDP Annual Conference, San Diego.

Posters

Kim, J. H., McNally, G., & Richardson, R. (2005). Repressed memories: A role for GABAergic inhibition? ISDP Annual Conference, Washington.

Kim, J. H., & Richardson, R. (2006). The loss and recovery of fear memories in the developing rat. The 4th International Conference on Memory, Sydney.

Richardson, R., & Kim, J. H. (2006). Renewal is dependent on extinction-test interval but not acquisition-extinction interval. ISDP Annual Conference, Atlanta.

Kim, J. H., & Richardson, R. (2006). The effect of context and FG7142 on extinguished fear memory in the developing rat. Annual Conference for Neuroscience, Atlanta.

Richardson, R., & Kim, J. H. (2006). Renewal is dependent on extinction-test interval but not acquisition-extinction interval. Annual Conference for Neuroscience, Atlanta.

G G

Kim, J. H. & Richardson, R. (2007). The role of the amygdala in extinction and re- extinction in the developing rat. ISDP Annual Conference, San Diego.

Kim, J. H. & Richardson, R. (2007). The role of the amygdala in extinction and re- extinction in the developing rat. Annual Conference for Neuroscience, San Diego.

Awards

2005 ISDP National Institute of Health (NIH) Travel Award 2006 ISDP NIH Travel Award 2007 ISDP Sandra G. Wiener Student Investigator Award 2007 ISDP John Wiley Best Poster Award

G G

Care and Use of Animals

The experiments presented in this thesis conformed to the guidelines on the ethical use of animals maintained by the Australian Code of Practice for the Care and Use of

Animals for Scientific Purposes (7th Edition), and all procedures were approved by the

Animal Care and Ethics Committee at the University of New South Wales. All efforts were made to minimise both suffering and the number of animals used.

G G

List of Tables

Table Page

2.1 Mean (± SEM) levels of baseline freezing at test for all groups across all 57 experiments in Chapter 2.

2.2 Mean baseline and CS-elicited freezing at test in P17 and P24 rats that 74 received training and the reminder treatment in Experiments 1 and 2.

2.3 Mean baseline and CS-elicited freezing at test in P17 rats that received 75 training and the reminder treatment in Experiments 1 and 2. P17 rats were divided into “high” and “low” groups based on their baseline levels of freezing.

2.4 Mean baseline and CS-elicited freezing at test in P24 rats that received 75 training and the reminder treatment in Experiments 1 and 2. P24 rats were divided into “high” and “low” groups based on their baseline levels of freezing.

3.1 Mean (± SEM) levels of baseline freezing at test for all groups across all 85 experiments in Chapter 3.

4.1 Mean (± SEM) levels of baseline freezing at test for all groups across all 114 experiments described in Chapter 4.

4.2 Experiment 9 design. 134

G G

List of Figures

Figure Page

1.1 The neurobiological model of conditioned fear. 8

1.2 Schematic diagram illustrating intracellular signalling pathways 9 involved in the long-term neuronal plasticity implicated in the formation of memory.

1.3 Summary of the approximate age at which conditioned freezing, heart- 11 rate changes, and fear-potentiated startle emerge for CSs of different sensory modalities.

1.4 Two theories on extinction of conditioned fear. 17

1.5 Reinstatement, renewal, and spontaneous recovery indicate that 19 extinction does not erase the original fear memory.

1.6 Simplified schematic diagrams of current models of extinction. 28

2.1 Two contexts used in the experiments presented in this thesis. 54

2.2 Mean CS-elicited freezing during extinction training in Experiment 1. 59

2.3 Mean (± SEM) CS-elicited freezing during test in Experiment 1. 60

2.4 Mean (± SEM) CS-elicited freezing during extinction in Experiment 2. 63

2.5 Mean (± SEM) CS-elicited freezing during test in Experiment 2. 64

2.6 The basic forgetting function in Experiment 3.1. 68

2.7 Mean (± SEM) CS-elicited freezing during test in Experiment 3.2. 69

2.8 Mean (± SEM) CS-elicited freezing during test in Experiment 3.3. 71

3.1 Mean (± SEM) percentage freezing to the CS during extinction training 87 in Experiment 1.

3.2 Mean (± SEM) percentage freezing by rats in response to the CS during 88 test in Experiment 4.

3.3 Mean (± SEM) percentage freezing by P24 rats to the CS during 91 extinction training in Experiment 5.1 G G

3.4 Mean (± SEM) freezing in response to the CS at test in Experiment 5.1. 92

3.5 Mean (± SEM) percentage CS-elicited freezing by P17 rats during 94 extinction training in Experiment 5.2.

3.6 Mean (± SEM) percentage CS-elicited freezing in Experiment 5.2. 95

3.7 Mean (± SEM) percentage CS-elicited freezing during extinction 96 training in Experiment 5.3.

3.8 Mean (± SEM) freezing by P17 rats in response to the CS during test in 97 Experiment 5.3.

3.9 Mean (± SEM) CS-elicited freezing during test in Experiment 6.1. 101

3.10 Mean (± SEM) CS-elicited freezing expressed by rats during test in 103 Experiment 6.2.

4.1 Histological reconstruction of coronal sections showing cannula 116 placements in the amygdala for Experiments 7.1 and 7.2.

4.2 Mean (± SEM) percentage freezing to the CS by P24 rats during 117 extinction training in Experiment 7.1.

4.3 Mean (± SEM) CS-elicited freezing by rats during test in Experiment 118 7.1.

4.4 Mean (± SEM) percentage freezing to the CS by P17 rats during 119 extinction training in Experiment 7.2.

4.5 Mean (± SEM) CS-elicited freezing by P17 rats during test in 120 Experiment 7.2.

4.6 Histological reconstruction of coronal sections showing cannula 124 placements in the amygdala for Experiments 8.1 and 8.2.

4.7 Mean (± SEM) percentage freezing to the CS by P24 rats during the 126 initial extinction training (drug-free) in Experiment 8.1.

4.8 Mean (± SEM) percentage freezing to the CS by P24 rats during re- 127 extinction in Experiment 8.1.

4.9 Mean (± SEM) percentage freezing to the CS by P24 rats during test in 128 Experiment 8.1. G G

4.10 Mean (± SEM) percentage freezing to the CS by P17 rats during the 129 initial extinction training (drug-free) in Experiment 8.2.

4.11 Mean (± SEM) percentage freezing to the CS by P17 rats during re- 130 extinction in Experiment 8.2.

4.12 Mean (± SEM) percentage freezing to the CS by P17 rats during test in 131 Experiment 8.2.

4.13 Histological reconstruction of coronal sections showing cannula 135 placements in the amygdala for Experiment 9.

4.14 Mean (± SEM) percentage freezing to the CS during the initial 136 extinction training (drug-free) in Experiment 9.

4.15 Mean (± SEM) percentage freezing to the CS during re-extinction in 137 Experiment 9.

4.16 Mean (± SEM) percentage freezing to the CS during test in Experiment 138 9.

5.1 Evidence for spontaneous recovery in P24 rats. 152

5.2 The lack of evidence for spontaneous recovery in P17 rats. 152

5.3 Mean freezing (+ or – SEM) by rats in response to the CS during the 162 extinction training in J. H. Kim and R. Richardson (unpublished observations).

5.4 A graph of total pMAPK stained neuron count in the IL and amygdala 162 in J. H. Kim and R. Richardson (unpublished observations).

5.5 Phosphorylated MAPK cell staining in the IL in J. H. Kim and R. 163 Richardson (unpublished observations).

5.6 Phosphorylated MAPK neuron staining in the amygdala in J. H. Kim 164 and R. Richardson (unpublished observations). The neurons were counted from the BLA, CeA and the STIA.

5.7 Possible neural structures involved in extinction of conditioned fear in 166 P24 and P17 rats.

G G

Abbreviations

ACTH adrenocorticotropic hormone

ANCOVA analysis of co-variance

ANOVA analysis of variance

AP5 D,L-2-amino-5-phosphonopentanoic acid

APV D,L-5-amino-phosphonovaleric acid

BA basal nucleus of the amygdala

BLA basolateral complex of the amygdala

Ca Calcium

CeA central nucleus of the amygdala

CPP D(-)-3-(2-carboxy-piperazine-4-yl)-propyl-1-phosphonic acid

CR conditioned response

CREB cyclic adenosine monophosphate response element binding protein

CS conditioned stimulus

DCS D-cycloserine

DH dorsal hippocampus dLA dorsal subnucleus of the lateral nucleus of the amygdala dmPFC dorsal medial prefrontal cortex

DSM diagnostic and statistical manual of mental disorders

ERK extracellular signal-regulated kinase

FG7142 -carboline-3-carboxylic acid N-methylamide

FPS fear-potentiated startle

GABA -aminobutyric acid

HSD honestly significant difference G G

IL infralimbic cortex

ITC intercalated (cells of the amygdala)

ITI inter-trial interval

LA lateral nuclei of the amygdala

LH lateral hypothalamus

LTD long-term depotentiation

LTP long-term potentiation

LVGCC L-type voltage-gated calcium channel

MAPK mitogen-activated protein kinase

MD mediodorsal thalamus

MO medial orbital cortex mPFC medial prefrontal cortex

MS mean sums of square

NMDA N-methyl-D-aspartate

P postnatal day

PAG periaqueductal gray

PKA cyclic adenosine monophosphate-dependent protein kinase

PL prelimbic cortex pMAPK phosphorylated MAPK

PnC pontine nucleus caudate

RNA ribonucleic acid

SE standard error

SEM standard error of the mean

SS sums of square G G

STD standard deviation

STIA stria terminalis intra-amygdaloid division

US unconditioned stimulus vmPFC ventromedial prefrontal cortex

Chapter One.

“Let the fear of danger be a spur to prevent it; he that fears not, gives advantage to the

danger.”

- Francis Quarles

"Anxiety is a thin stream of fear trickling through the mind. If encouraged, it cuts a channel

into which all other thoughts are drained."

- Robert Albert Bloch

The present thesis investigates extinction of conditioned fear in the developing rat. One reason for doing this is to gain a better understanding of the treatment of anxiety disorders in humans. At present, the most effective treatment for anxiety disorders is exposure therapies, which are based on the process of extinction. Anxiety disorders are characterised by overwhelming feelings of panic and fear accompanied with physical symptoms such as heart pounding, feeling sick to your stomach, and muscle tension [Diagnostic and Statistical

Manual of Mental Disorders (DSM-IV); American Psychiatric Association, 1994].

Normally, fear and anxiety are emotions that are necessary in the struggle for survival.

These emotions serve to protect one from danger by causing reactions such as hiding or avoidance. However, an anxiety disorder emerges when fear and anxiety become pervasive and interfere with normal functioning, or as put by Freud (1920, p. 341) ‘...if fear is too strong, it proves absolutely useless and paralyzes every action…’

The prevalence of anxiety disorders show that these adaptive emotions becomes

‘too strong’ for many people. In the recent United States National Comorbidity Survey 2

Replication study (Kessler, Chiu, Demler, & Walters, 2005), it was reported that anxiety disorders are the most common of mental disorders. In fact, more than 18% of adults met the 12-month DSM-IV criteria for an anxiety disorder, meaning that up to 54 million

Americans suffer from such a disorder. Further, the latest figures from the Australian

Bureau of Statistics (1997) show that 12% of women and 7.1% of men suffer from an anxiety disorder in Australia.

Unfortunately, the cause of anxiety disorders is yet largely unknown. One of the oldest ideas on the etiology of anxiety disorders posits that memories, especially those that were formed early in life, are critically involved. Specifically, Freud and Breuer (1893) suggested that ‘hysteria’ (i.e., an anxiety disorder) is a previously experienced traumatic event that remains in memory and triggers anxious thoughts and maladapative behaviours.

This general idea that previous traumatic experience, especially ones occurring early in life, triggering anxiety disorders is still is widely-held (e.g., Jacobs & Nadel, 1985; 1999;

Mineka & Zinbarg, 2006).

In fact, research indicates that anxiety disorders emerge quite early in life (e.g.,

Cartwright-Hatton, McNicol, & Doubleday, 2006; Kessler et al., 2005; Newman et al.,

1996). For example, in an unselected birth cohort of 21-year-olds in Dunedin, New Zealand, it was reported that nearly three fourths (73.8%) of the participants with a DSM-III-R diagnose of some type of anxiety disorder at age 21 had been previously diagnosed during childhood and adolescence (earliest age examined was 11; Newman et al., 1996). Further, that study showed that childhood/adolescence-onset cases of anxiety disorder express more severe symptoms than their disordered counterparts who had been diagnosed in adulthood.

The authors conclude that “treatment resources clearly need to focus on the younger sector of the population” (Newman et al., 1996, p. 561). 3

It is surprising then that not much is known about the treatment of anxiety disorders in children (Cartwright-Hatton, Roberts, Chitsabesan, Fothergill, & Harrington, 2004; Silva,

Gallagher, & Minami, 2006). In adults, the most utilised treatment for anxiety disorders is exposure therapies, in which the patient is typically exposed to a stimulus that represents the fear-triggering object (e.g., a picture of a spider for arachnophobics) in a safe environment and without any adverse consequences. As mentioned previously, exposure therapy relies on the concept of extinction, which refers to the decrease in fear responses expressed to a stimulus that was previously fear-eliciting. This decrease in fear responses is observed following the animal receiving multiple presentations of the stimulus without any aversive outcome. Hence, extinction serves as a laboratory model to study exposure therapy for the treatment of fear and anxiety.

Understanding the processes by which fear is diminished is critically important to the development of effective treatments for various anxiety disorders. It is also theoretically important to examine extinction, as extinction represents one of the simplest ways of modifying a previous learned behaviour. Due to these reasons, fear extinction has been extensively examined in adult animals over the past decade (see Davis & Myers, 2002, for review). However, we know very little about fear extinction across development.

Examining extinction in the developing animal is important because of the long-held belief that early learning experiences have a profound impact on later behaviour (e.g., Harlow

1959; Mineka & Zinbarg, 2006). Jacobs and Nadel (1985; 1999), for example, suggested that fear acquired early in development is particularly resistant to the effects of extinction, and forms the basis of anxiety disorders emerging later in life. However, there is very little, if any, empirical data to support this assertion.

Therefore, this thesis investigates behavioural and neural characteristics of fear extinction in the developing rat. Considering the importance of early experiences in the development of anxiety disorders, examining extinction in immature rats may have significant implications for the treatment of anxiety disorders in humans. Also, examining extinction in the developing rat will provide a unique way of assessing the processes involved in extinction. This is important because there is still little known about the mechanisms mediating extinction, especially in comparison to what we know about acquisition of learned fear.

I. PAVLOVIAN CONDITIONED FEAR

Among the early advocates of the idea that anxiety disorders emerge from traumatic memories were Watson and Rayner (1920) who proposed that humans are born with very few 'emotional reaction patterns', and the way in which stimuli come to evoke emotion throughout one’s life is through learning, or conditioning, processes. That is, animals learn to respond in distinct ways to different stimuli according to previous experiences with those stimuli. According to this notion, simple aversive learning and memory processes are critical for triggering anxiety disorders. To support this idea, in their famous ‘little Albert’ experiment Watson and Raynor (1920) conditioned a baby boy to fear a rat by sounding an aversive loud noise each time the baby touched the rat. Before this experience, the baby was keen to touch and play with the rat. After the experience, however, the baby displayed signs of fear and anxiety (e.g., crying and reaching for mum) when in the presence of the rat. The simple pairing of the rat and the aversive noise induced fear learning in little 5

Albert.

This associative learning is referred to as classical or Pavlovian conditioning because Pavlov was the first to demonstrate such associative learning in animals (see

Pavlov, 1927, for review). Typically, Pavlovian fear conditioning involves pairings of a discrete conditioned stimulus (CS), such as a tone, with an aversive unconditioned stimulus

(US), such as a footshock. Initially, the CS is a neutral stimulus that has little effect on animals. On the other hand, the US is a stimulus that already elicits fear responses via physiological, autonomic, and behavioural changes. These fear responses triggered by the

US are referred to as unconditioned responses. When the CS and the US are repeatedly paired, the CS starts to elicit such autonomic and behavioural fear responses on its own without the US, and these responses are referred to as conditioned responses (CRs).

Such fear conditioning is readily acquired in animals, and is not easily forgotten

(e.g., Gale et al., 2004). Because this form of learning can be acquired even in very immature animals (e.g., 12-day-old rats; Sullivan, Landers, Yeaman, & Wilson, 2000),

Pavlovian conditioning provides a model for investigating the neurobiological bases of anxiety disorders. That is, theories on the etiology of anxiety disorders suggest that aversive learning very early in life triggers anxiety disorders later on. Apart from Pavlovian conditioned fear, however, there are other laboratory models of anxiety disorders. For example, Grillon and his colleagues propose that anxiety disorders are caused by unpredictable aversive events that lead to hyperactive fear responses that are generalised across different situations and environments (Grillon, Baas, Lissek, Smith, & Milstein,

2004). Specifically, if an aversive US (e.g., shock) is not signalled reliably by a CS (e.g., coloured shape), subjects show more fear to the training context compared to subjects that received the US signalled reliably by a CS (Grillon et al., 2004). Grillon and his colleagues 6

concluded that the elevated fear to the context exhibited by the unsignalled group reflects generalised anxiety. Another laboratory model for anxiety disorders relies on the concept of

‘unlearned’ fear, which refers to the innate fear that animals have to various stimuli without any previous learning experiences with these stimuli (e.g., Boulis & Davis, 1989).

Unlearned fear is triggered by a diffuse cue rather than a specific cue, therefore, unlearned fear can serve as a good model for anxiety disorders that may not be triggered by a specific object or a situation (e.g., generalised anxiety disorder).

Nevertheless, Pavlovian fear conditioning is the most utilised neurobiological model for anxiety. This is partly because this preparation allows a variety of discrete stimuli to be used as the CS or the US, so that the role of each sensory modality and different neural structures in learning can be examined. Also, Pavlovian conditioning is easily acquired in many different animals across a range of different ages, unlike contextual conditioning in Grillon’s model for anxiety (Grillon et al., 2004). Further, considering that anxiety disorders may be normal fear gone astray (e.g., Rosen & Schulkin, 1998), using

Pavlovian conditioning as a model for fear learning provides a tool for understanding how the fear system works. It is no surprise then since Pavlov (1927), researchers investigated how the brain processes the CS and the US, how the CS and US are associated, and how the CS comes to elicit similar responses as the US as a result of conditioning.

The neural model of conditioned fear: a summary

The following brief summary of research on the neural bases of learned fear will focus on work done with the rat. It should be noted, however, that is has been suggested that similar neural processes are involved in learned fear across a range of species (Davis, 2000; 7

LeDoux, 2000; Price, 2003). These findings have pin-pointed the amygdala as the central structure for conditioned fear. That is, research has shown that the amygdala is essential for the acquisition, storage, and the expression of conditioned fear (Figure 1.1) (Davis, 2000;

LeDoux, 2000; Phelps & LeDoux, 2005; Ressler & Davis, 2003).

The amygdala is made up of many sub-nuclei such as the central (CeA), and basolateral (BLA) nuclei. The BLA is further divided into basal (BA) and lateral (LA) nuclei. During fear conditioning, perceptual (e.g., olfactory, visual, etc) and somatosensory

(e.g., pain) information regarding the CS and US is transmitted to the LA (e.g., LeDoux,

1993; LeDoux, Farb, & Ruggiero, 1990; McDonald, Mascagni, & Guo, 1996; Price 2003;

Shi & Davis, 2001). As the result of CS-US convergence, neuronal activity in the LA is modified (e.g., Quirk, Armony, & LeDoux, 1997; Quirk, Repa, & LeDoux, 1995; Repa et al., 2001). Once the CS and the US have been associated, the fear memory is stored in the

BLA via intracellular molecular changes (Figure 1.2). These intracellular changes include pre- synaptic release of glutamate, N-methyl-D-aspartate (NMDA)-mediated calcium entry, activation of protein kinases, activation of transcription factors in the nucleus, and the synthesis of new proteins (e.g., Bailey, Kim, Sun, Thompson, & Helmstetter, 1999;

Campeau et al., 1991; Farb, Aoki, Milner, Kaneko, & LeDoux, 1992; Lee & Kim, 1998;

Lin et al., 2001; Maren, Ferrario, Corcoran, Desmond, & Frey, 2003). Hence, fear conditioning results in an enhanced functional and structural connectivity between sensory pathways and the BLA, such that future presentation of the CS alone is sufficient to activate the BLA.

Figure 1.1 The neural mechanisms underlying Pavlovian conditioned fear (adapted from Phelps & Ledoux, 2005).

Figure 1.2 Schematic diagram illustrating intracellular signalling pathways involved in the long-term neuronal plasticity implicated in the formation of memory. NMDA receptor and LVGCC activation leads to intracellular events that leads to transcription of genes and protein synthesis (adapted from Ressler & Davis, 2003).

When the BLA is activated by the CS, the CeA also becomes activated by excitatory inputs from the BLA (e.g., Pitkanen, Savander, & LeDoux, 1997). The CeA then projects to a variety of neural structures responsible for mediating the expression of fear and anxiety (Davis, 2000; LeDoux, 2000). For example, the CeA may activate the midbrain periaqueductal gray (PAG) to elicit freezing in the rat (e.g., LeDoux, Iwata, Cicchetti, &

Reis, 1988). Freezing is characterised by an immobile, crouching position accompanied by hypervigilance, and is operationally defined as the absence of movement other than that required for respiration (Blanchard & Blanchard, 1972). Fear and anxiety can also be expressed as changes in heart rate and/or arterial pressure via activation of the lateral hypothalamus (LH) by the CeA (LeDoux et al., 1988). Fear-potentiated startle (FPS) is another commonly used index of fear. The startle response is evoked by an intense, unexpected stimulus (startle pulse; e.g., a brief but loud, sharp noise), and in studies with rats, the whole body jump is usually taken as the magnitude of startle (Davis, Falls,

Campeau, & Kim, 1993). FPS is the increase in startle response when the fear-associated stimulus (e.g., CS associated with shock) is present compared to when it is absent. The neural structure responsible for FPS is the pontine nucleus caudate (PnC), which is also an area that receives excitatory projections from the CeA (Rosen, Hitchcock, Sananes,

Miserendino, & Davis, 1991).

Research on Pavlovian conditioning using the developing rat has also provided evidence supporting this model of learned fear. Specifically, the expression of Pavlovian conditioned fear emerges in a cue- and response-specific sequence during the first several weeks of life (Figure 1.3). These findings illustrate that the fear system is made up of independent pathways from the sensory cortices into the amygdala, which in turn has separate projections to each independent neural structure that mediates different fear CRs. 11

Figure 1.3 Summary of the approximate age at which conditioned freezing, heart-rate changes, and fear-potentiated startle emerge for CSs of different sensory modalities (adapted from Hunt & Campbell, 1997).

The ontogeny of conditioned fear

Rats are altricial species, hence their central nervous system and motor abilities mature substantially after birth. Specifically, at birth, the rat brain is more immature than the newborn human brain, but within 30 days the rat brain becomes essentially adult-like (see

Spear & Rudy, 1991, for review). The findings on the ontogeny of conditioned fear in the 12

rat clearly illustrate how the growth of the rat’s brain affects learning and memory.

Specifically, fear conditioning to CSs of various different sensory modalities emerge in a similar sequence of basic sensory development. In basic sensory development, the functional onset of the tactile and chemical (olfactory and taste) senses precedes that of the auditory system in the rat (see Alberts & Gubernik, 1984, for review). Further, the functional onset of the auditory system precedes that of the visual system across a wide range of species (see Gottlieb & Klopfer, 1962, for review). For example, the ability of rats to discriminate between two different olfactory stimuli has been documented at 2 days of age (Rudy & Cheatle, 1977), whereas discrimination between different auditory stimuli was observed at around 10 days of age (Hyson & Rudy, 1984). Additionally, orienting responses to a visual stimulus has been reported in 14-day-old rats (Hunt, Hess, &

Campbell, 1998). In Pavlovian conditioning, rats express a learned fear response to an odour CS before they express the same learned response to an auditory CS (Campbell &

Ampuero, 1985; Kucharski & Spear, 1984), and express a learned fear response to an auditory CS before they express the same learned response to a visual CS (Figure 1.3) (see

Hunt & Campbell, 1997, for review).

Associative learning also emerges in a response-specific sequence (Figure 1.3).

That is, an auditory CS paired with an aversive US elicits freezing at 16 days of age, whereas it alters heart rate at 21 days of age; and it potentiates the startle response at 23 days of age (see Hunt & Campbell, 1997, for review). This specificity observed in the emergence of conditioned fear responses is not attributable to any apparent biological constraints preventing the expression of the responses. For example, Martin and Alberts

(1982) demonstrated that rats as young as 5-6 days of age show marked heart rate changes when presented with a novel odour. Such result shows that heart-rate changes as a 13

physiological response is functional by at least 5-6 days of age, although conditioned heart-rate changes are not observed until later. Further, 18-day-old rats show an elevated startle response when merely exposed to a bright light compared to when the light is absent, indicating that rats are capable of showing the startle response by 18 days of age (Weber,

Watts, & Richardson, 2003). However, FPS to an odour CS previously paired with shock is not observed until 23 days of age (Richardson, Paxinos, & Lee, 2000).

These studies on the ontogeny of conditioned fear provide converging evidence for the view that different neural structures mediate the various forms of learned fear. To re-iterate, freezing is thought to be mediated by the PAG, changes in heart rate by the LH, and FPS by the PnC (Figure 1.1). Given the multiple, independent projections from the amygdala to the neural structures responsible for different CRs, it is perhaps not surprising that some fear responses emerge at later ages than do others. The finding that these different fear responses emerge at different ages is usually interpreted as a result of the delayed development of the projections from the amygdala to the relevant downstream structures. This interpretation is based on the findings that show these responses can be evoked without conditioning, which suggests that the neural structures mediating these responses are functional. Without amygdala input, however, fear responses as a consequence of learning cannot be expressed because the amygdala is the site of CS-US convergence according to the neural model of conditioned fear.

Interestingly, the developmental analysis of fear learning also provides evidence against the current model of conditioned fear. An important idea in this model is that after conditioning, subsequent presentations of the CS activates a central state of fear in the amygdala. This central state of fear then can be expressed as any index of fear by the activation of downstream neural structures that mediate fear CRs (e.g., Davis, 1992). 14

However, there is recent evidence suggesting that some modifications are necessary to this model (e.g., Richardson & Fan, 2002; Weber & Richardson, 2004; Wilensky, Schafe,

Kristensen, & Ledoux, 2006). The most convincing evidence comes from a developmental study, which showed that behavioural expression of conditioned fear in rats is appropriate to their age at training, not their age at testing. This study utilised the fact that different CRs emerge at different ages in the rat. Specifically, Richardson and Fan (2002) conditioned rats to fear an odour CS by pairing it with shock. The rats were conditioned at an age at which they can show learned avoidance of the odour CS but not FPS (i.e., at 16 days of age). Interestingly, when rats were tested at 23 days of age (i.e., when FPS can be expressed), they showed odour-avoidance but failed to show conditioned odour-potentiated startle. This finding is contradicts the model of learned fear, which states that a CS previously paired with an aversive US elicits a central state of fear that can be expressed via any behavioural responses that the animal is capable of expressing. From this perspective, rats trained at 16 days of age but then tested at 23 days of age should show both odour avoidance and the conditioned odour-potentiated startle. However, that was not the case. The authors suggested that CS–US pairings produce neural plasticity not only in the amygdala, as is widely accepted, but also in structures downstream from the amygdala

(i.e., the various structures that mediate specific behavioural responses indicative of learned fear).

Taken together, the studies described above highlight the importance of the developmental approach to the analysis of conditioned fear. Specifically, the developmental approach provides a unique ‘natural lesion’ preparation to test neural basis of fear because of the immature brain in the developing rat (Fanselow & Rudy, 1998). Also, these developmental 15

studies indicate that immature animals readily acquire Pavlovian conditioning; showing that Pavlovian conditioned fear is a good model for investigating anxiety disorders. As mentioned earlier, theories on the development of anxiety disorders suggest that these disorders come from traumatic events experienced early in life that remain in memory and trigger fear and anxiety (e.g., Freud & Breuer, 1983; Jacobs & Nadel, 1985, 1999; Mineka

& Zinbarg, 2006). According to these theories, a good model of anxiety disorders must produce learning in immature animals that can persist in memory.

It is yet unclear whether immature animals do retain a learned fear memory across development. In the adult rat, however, we know that a fear memory, once acquired, is permanently stored in the amygdala. This memory is also extremely resistant to spontaneous forgetting. For example, one study showed that when adult rats were conditioned to fear an auditory CS by pairing it with shock, they expressed freezing to the

CS even when they were tested 18 months later (Gale et al., 2004).

However, Pavlovian conditioned fear can be easily reduced by post-conditioning treatments. One example is extinction. Because of its practical implications, significant progress has been made in our understanding of extinction in the adult rat. Nevertheless, we still know relatively little about extinction compared to acquisition of fear, and virtually nothing is known about fear extinction in the developing rat. Therefore, the present thesis examines extinction of conditioned fear in the developing rat.

II. EXTINCTION OF CONDITIONED FEAR

“Extinction is one of the constant facts of the physiology of conditioned reflexes”

- Ivan P. Pavlov

As with conditioning, Pavlov was the first person to document extinction in the laboratory

(see Pavlov, 1927, for review). He observed that animals show a robust CR after CS-US pairings; however, the CR dramatically decreased if the CS was presented repeatedly without the US. Fear extinction hence refers to the decrease in fear conditioned responding due to repeated presentation of the CS without any aversive outcome.

In a typical extinction session, the rat is exposed to non-reinforced presentations of a CS (e.g., tone) that had previously been paired with an aversive US (e.g., shock). Over the course of such extinction trials the CR elicited by the CS (e.g., freezing) is reduced in magnitude and frequency. Early theoretical models of Pavlovian conditioning suggested that this decreased response to the CS after extinction was due to the ‘unlearning’ or

‘erasure’ of the original CS-US association (e.g., Rescorla & Wagner, 1972). That is, the association between the CS and US is severed so that the CS becomes neutral again (Figure

1.4A). However, it is now widely believed that the CS-US association remains intact after extinction (Figure 1.4B). The primary evidence for this view comes from behavioural studies that show performance to an extinguished CS recovers without subsequent re-training of the CS-US association, namely in the cases of reinstatement, renewal, and spontaneous recovery (Figure 1.5). Further, pharmacological studies indicate that extinction shares similar neural mechanisms with fear acquisition, suggesting that 17

extinction involves new learning rather than unlearning. These findings on extinction have shaped the widely accepted view on extinction – that the decrease in the CR following extinction reflects new learning of a competing association that inhibits the expression of the original association (i.e., CS-no US vs. CS-US; see Bouton, 2002, for review).

Figure 1.4 Two theories on extinction of conditioned fear (A) The reduced conditioned response after extinction training reflects unlearning of the CS-US association. (B) Extinction training leads to learning of a second, competing association that inhibits the expression of the original CS-US association (adapted from Myers & Davis, 2007). 18

Reinstatement

Reinstatement is the recovery of an extinguished CR following a post-extinction (or pre-test) presentation of a ‘reminder’ cue (Figure 1.5A). In fear extinction, the reminder cue typically involves a re-exposure to the US alone (often of a slightly lower intensity than at training) or some other stressful experience. Rescorla and Heth (1975) demonstrated reinstatement of extinguished fear using conditioned suppression. In that study, food-deprived rats were first trained to bar-press for food, and then received tone-shock pairings. In extinction training, the rats initially suppressed bar-pressing behaviour during the tone presentations, but the level of conditioned supporession declined substantially by the last extinction session. Twenty-four hrs after the final extinction session, rats in the experimental group were presented with a single shock. When they were tested the next day, these rats showed a recovery in conditioned suppression behaviour compared to the control group that did not receive the reminder shock.

A series of follow-up studies by Bouton and his colleagues suggest that the underlying process of reinstatement relies on the context-reminder association that forms during the reminder episode. This view is supported by studies showing the reinstatement effect to be context-dependent; that is, post-extinction reinstatement is restricted to the context where the reminder had been presented (e.g., Bouton 1984; Bouton & Bolles 1979a;

Bouton & King 1983; Frohardt, Guarraci, & Bouton, 2000; Wilson, Brooks, & Bouton,

1995). However, Bouton and Bolles (1979a) showed that the reinstatement of the conditioned suppression behaviour was not a simple summation of the context-reminder association and the remaining CS-US association, because without the CS, rats showed no fear to the context where the reminder had been given. From this, it appears that a direct 19

Figure 1.5 Behavioural findings indicate that extinction does not erase the original fear memory. (A) Reinstatement of extinguished fear occurs when the US is given as a reminder between the final extinction training and a subsequent retention test. Reinstatement is context-specific, as it is only observed in the context where the reminder was given. (B) Renewal refers to the recovery of extinguished fear when the rat is tested outside the extinction context. (C) Spontaneous recovery of extinguished fear occurs with the simple passage of time following extinction. The magnitude of recovery increases with the length of the extinction-test interval (adapted from Myers & Davis, 2007). 20

association between the context and the US reminder disinhibits responding to the extinguished CS. In support of this idea, Bouton and Bolles (1979a) additionally showed that reinstatement is not observed if the rat receives extensive extinction to the context where the US reminder was given.

Although the exact mechanism underlying the reinstatement effect is still a debated issue (e.g., Richardson, Duffield, Bailey, & Westbrook, 1999), this effect shows that extinction does not cause an erasure of the CS-US association, even if the CR appears to be abolished completely. However, a recent study has pointed out that the reinstatement effect is not a ‘pure’ demonstration of how the CS-US association survives after extinction, as the reminder treatment may strengthen the remaining CS-US association by mediated conditioning via the context (e.g., Westbrook, Iordonova, McNally, Richardson, & Harris,

2002). Nevertheless, Westbrook et al. (2002) noted that the effect of mediated conditioning is small, therefore, the reinstatement effect is still an illustration of how the CS-US association remains intact after extinction.

Renewal

Perhaps the most robust demonstration of recovery in conditioned responding after extinction is the renewal effect, which refers to the return of an extinguished CR when the subject is tested in a different context to where extinction occurred (Figure 1.5B). The renewal effect is regarded as a good example of the new learning account of fear extinction because it requires no additional post-extinction treatment.

The first demonstration of renewal was provided by Bouton and Bolles (1979b). In that study, rats were conditioned in one distinctive context (A), and then received CS 21

extinction in another context (B) until the CR was completely lost. The following day, rats exhibited low levels of fear if tested in the extinction context (B) but higher levels of fear if tested in the original training context (A) (i.e., ABA renewal). ABA renewal is not due to a simple summation of the conditioning context’s excitatory properties and the extinguished

CS’s remaining associative strength because renewal can occur without evidence of conditioned responding in context A (e.g., Bouton & King, 1983; Bouton & Swartzentruber,

1989). Further, ABA renewal is not because of disinhibition (due to the context change between extinction and test) because renewal is still observed without evidence of inhibition in context B (e.g., Bouton & King, 1983; Bouton & Swartzentruber, 1989).

Interestingly, renewal can also occur if rats are tested in a novel context (ABC renewal; Bouton & Bolles, 1979b), and when rats are conditioned and extinguished in the same context but then tested in a second context (AAB renewal; Bouton & Ricker, 1994).

These results indicate that putative excitation in the original training context is not necessary for renewal to occur. Instead, the context appears to modulate or “set the occasion”, so that either the CS-US or CS-no US association is retrieved in the appropriate setting (see Bouton, 2002, for review). The renewal effect strongly suggest that extinction leads to the formation of a new memory so that the CS has two available meanings, and the retrieval of any one meaning is critically determined by the testing context.

Spontaneous recovery

Spontaneous recovery also illustrates how the CS-US memory remains intact after extinction. Spontaneous recovery is the return of responding to an extinguished CS when some interval of time elapses after extinction. This effect is often viewed as a form of 22

renewal, except that the source of context change is not physical but temporal (see Bouton,

Westbrook, Corcoran, & Maren, 2006, for review).

Pavlov (using an appetitive paradigm) was the first to document spontaneous recovery after extinction (Pavlov, 1927). Spontaneous recovery is also observed after fear extinction. In one such study, Quirk (2002) gave rats auditory CS-shock US pairings followed by 20 non-reinforced CS presentations (i.e., extinction), and tested for spontaneous recovery after a delay of 0, 1, 2, 4, 6, 10, or 14 days. Conditioned freezing to the tone gradually recovered with time, and by 10 days after extinction, CS-elicited freezing reached pre-extinction levels. Quirk (2002) noted that 100% recovery of freezing observed at test after a delay suggests that the CS-US memory is completely preserved after extinction. This study also examined whether the extinction memory also persists after spontaneous recovery by giving rats additional extinction sessions after observing 100 % spontaneous recovery in the extinguished response. Interestingly, rats showed savings in their rate of extinction after spontaneous recovery, indicating that even after the original fear memory has been fully recovered, extinction memory still persists. These data suggest that conditioning and extinction of fear are learned and maintained independently.

Pharmacological Studies

In addition to the behavioural phenomena described above, various pharmacological studies support the notion that extinction reflects a new inhibitory learning rather than erasure of the fear memory. For example, Richardson, Riccio, and Devine (1984) showed that extinction cannot be due to erasure because extinguished avoidance behaviour can be recovered by pre-test systemic injections of adrenocorticotropic hormone (ACTH). Specifically, rats were 23

trained on a passive avoidance procedure, in which footshocks were given in the black compartment of a two-compartment shuttle box; whereas nothing happened (no shocks were given) in the other, white compartment. During subsequent extinction training sessions, rats initially displayed avoidance of the black compartment, indicated by a longer latency to enter the normally preferred black side. By the end of extinction training, rats were readily entering the black side. At test, injections of ACTH restored the original avoidance behaviour, an effect that lasted up to 24 hrs. This finding appeared to be due to memory processes rather than an alteration in motor behaviour because ACTH had no effects on the behaviour of rats that had been shocked in another experimental chamber (i.e., not trained on the passive avoidance task). Further, pre-test administration of ACTH was also effective in alleviating performance deficits after extinction of an active avoidance task, in which shorter latency indicates fear memory recovery (i.e., opposite to passive avoidance).

As ACTH alleviated extinction using memory tasks that use opposite behavioural measures, it cannot be said that the effects of ACTH is on motor behaviour of the rat. Overall,

Richardson et al (1984) illustrates that extinction does not erase memory.

Since Richardson et al. (1984), recent advances in our understanding of the molecular processes involved in the formation of memory has renewed interest in the pharmacological studies of extinction. We now know that formation of long-term memory is linked to a molecular cascade involving pre-synaptic glutamate release, NMDA- mediated calcium entry, activation of protein kinases, gene expression and transcription, and protein synthesis (Figure 1.2). As mentioned earlier, fear learning is dependent on this molecular cascade. Hence, involvement of this cascade in extinction would provide support for the idea that extinction is new learning. Indeed, blocking NMDA receptors systemically prevents the formation of long-term memory for extinction (e.g., Baker & Azorlosa, 1996; 24

Santini, Muller, & Quirk, 2001). In one study using CS-elicited freezing and conditioned suppression as indices of fear, it was found that systemic injection of the NMDA antagonist

D(-)-3-(2-carboxy-piperazine-4-yl)-propyl-1-phosphonic acid (CPP) prior to extinction impaired extinction expression at test the next day, although there were no apparent effects during the acquisition of extinction (Santini et al., 2001). As CPP had no effects during extinction training (i.e., rats injected with CPP showed comparable rate of extinction as the control rats), it appears that NMDA neurotransmission is critical for the consolidation/storage rather than acquisition of extinction.

However, a recent study by Sotres-Bayon, Bush, and LeDoux, (2007) reported that a subset of NMDA receptors (those that include the NR2B subunit) is involved specifically with the acquisition of extinction. In that study, selective NR2B blockade with systemic injections of ifenprodil, given before extinction training impaired within-session extinction learning. Importantly, the magnitude of spontaneous recovery at test 24 hrs later (drug-free) was the same in both vehicle and ifenprodil groups, suggesting that the consolidation of extinction was unaffected by NR2B blockade. In contrast to Santini et al (2001), these results suggest that NMDA neurotransmission (involving the NR2B subunit) is critical for acquisition of extinction. Although more studies are necessary to elucidate the role of different NMDA receptor subunits in extinction of conditioned fear, the overall involvement of NMDA receptors in various stages of extinction indicates that extinction may be new learning.

On the other hand, there are studies that show extinction encompasses mechanisms that are distinct from fear acquisition. Cain, Blouin, and Barad (2002) demonstrated that pre-training injection of L-type voltage-gated calcium channel (LVGCC) inhibitors at doses that completely block extinction fail to prevent acquisition of fear in mice. Any long-term 25

effect of pre-training injection of nifedipine or nimodipine (LVGCC inhibitors) on acquisition of fear was observed to be state-dependent. State-dependent effect refers to a deficit in memory retrieval due to certain pharmacological agents causing a distinct internal state during training so that the internal context at test is different from training.

State-dependent effect can be eliminated by matching the training-test contexts by administration of the drug at training and at test. Cain et al. (2002) showed that injections of nimodipine before conditioning and testing eliminated any effect of nimodipine on conditioned fear that was observed previously. Therefore, it appears that LVGCCs are not necessary for fear acquisition but are necessary for extinction. In contrast to Cain et al.

(2002), Bauer, Schafe, and LeDoux (2002) showed that pre-training intra-LA infusion of

LVGCC blocker verapamil blocks consolidation of conditioned fear. However, in Bauer et al. (2002), the state-dependent account was not addressed. As Cain et al. (2002) demonstrated that systemic injection of LVGCC blockers leads to state-dependent effects, it may be the case that in Bauer et al. (2002), the observed effects of verapamil on acquisition of conditioned fear was due to state-dependent effects. Therefore, it appears likely that LVGCC neurotransmission uniquely involved in extinction but not in acquisition conditioned fear.

Extinction appears to also involve increased -aminobutyric acid (GABA) inhibitory neurotransmission, a mechanism not involved in fear acquisition. GABA is the dominant inhibitory neurotransmitter in the mammalian central nervous system, both with respect to the number of synapses and to functional relevance (Haefely, 1990; Olsen, 2002).

As extinction leads to a decrease in fear, it is believed that extinction is likely to involve inhibitory mechanisms. Hence, GABAergic neurotransmission has been studied as the 26

candidate underlying inhibition of fear following extinction. Indeed, pre-test injections of the GABA receptor inverse agonist -carboline-3-carboxylic acid N-methylamide (FG7142) have been shown to attenuate extinction (Harris & Westbrook, 1998b). Harris and

Westbrook (1998b) concluded that FG7142 reduced the inhibitory activity of GABA, which led to the retrieval of the original fear memory. This result shows that extinction relies on inhibitory mechanisms that are distinct from acquisition of fear.

These pharmacological and behavioural findings suggest the following three characteristics of extinction. First, extinction largely reflects a formation of a new and separate memory from the initial fear acquisition memory. Second, extinction memory appears to be inhibitory. Third, extinction expression is contextually-bound. Taken together, fear extinction appears to be a context-specific inhibitory learning that competes with the original fear memory expression.

Viewing extinction as new inhibitory learning is actually an old idea. Upon observing spontaneous recovery and reinstatement after extinction, Pavlov (1932, p. 99) commented that “extinction is actually inhibition”. Following Pavlov, Konorski (1948) further proposed that extinction is a new learning, hence it may share similar neural mechanisms as conditioning. For example, both initial learning and extinction may cause the formation and multiplication of synaptic connections. Further, Konorski (1948) also posited that extinction must involve neurobiological mechanisms that are distinct from those initially engaged by conditioning, hypothesising that conditioning likely involved excitatory synaptic connections, whereas extinction likely involved inhibitory synaptic connections. Guided by these ideas, researchers have begun to delineate the neural 27

substrates of extinction across different levels of analysis, from systems to cellular to molecular.

III. THE NEUROBIOLOGY OF FEAR EXTINCTION

In recent years, significant advances have been made in the neurobiological study of fear extinction. Local infusion studies, lesion studies, in vivo electrophysiology studies, and immunohistochemical studies of chemical expression in specific neural structures strongly suggest that extinction of conditioned fear involves a circuit including the amygdala and the medial prefrontal cortex (mPFC) (Figure 1.6A) (Garcia, 2002; Herry & Garcia, 2002;

Milad & Quirk, 2002; Morgan & LeDoux, 1995; Morgan, Romanski, & LeDoux, 1993;

Quirk, Russo, Barron, & Lebron, 2000). Additionally, the hippocampus has been incorporated into this neural circuitry, to account for the context effects on extinction

(Figure 1.6B) (see Bouton et al., 2006; Myers & Davis, 2007; Sotres-Bayon, Bush, &

LeDoux, 2004; Sotres-Bayon, Cain, & LeDoux, 2006, for reviews). Although other areas such as the mediodorsal thalamus (MD) and the PAG are also involved in extinction of conditioned fear (e.g., Garcia, Chang, & Maren, 2006; McNally, Lee, Chiem, & Choi,

2005), it appears that the most critical structures for extinction are the amygdala, mPFC, and the hippocampus. Hence, the following review on the neurobiology of extinction will focus on these three structures.

Figure 1.6 Simplified schematic diagrams of current models of extinction (A) During extinction expression, IL excitatory projections activate inhibitory ITC projection neurons, which in turn inhibit CeA output activity. This pathway competes with excitatory projections to the CeA from the BLA (adapted from Sotres-Bayon et al., 2004). (B) Context-dependence of extinction is mediated by the hippocampus. In the extinction context, fear is decreased because the IL mPFC inhibits amygdala output. However, amygdala output is uninhibited outside the extinction context (adapted from Quirk & Mueller, 2008).

The amygdala

It is intuitive to hypothesise that fear extinction requires the amygdala as fear memories are stored in the amygdala (e.g., Gale et al., 2004). Indeed, substantial evidence indicates that 29

the BLA (especially the LA) plays a crucial role in extinction learning. The first study showing the involvement of BLA in extinction demonstrated that bilateral infusion of the

NMDA antagonist D,L-2-amino-5-phosphonopentanoic acid (AP5) into the BLA disrupted long-term extinction (Falls, Miserendino, & Davis, 1992). In that study, rats were initially conditioned with 10 light CS-shock US pairings, and then received 30 CS-alone presentations per day for 2 days. Prior to each extinction session, rats were infused either with vehicle or different doses of AP5 into the BLA. One day after the last extinction session, rats were tested for FPS, drug-free. Rats that received the vehicle infusions showed low potentiated startle to the light, confirming that the learned fear had been extinguished.

On the other hand, intra-BLA infusions of AP5 dose-dependently disrupted extinction. This result suggests that long-term extinction is dependent on NMDA receptors in the amygdala.

Consistent with Falls et al. (1992), it has been demonstrated that intra-BLA infusions of the NMDA partial agonist D-cycloserine (DCS) facilitates extinction (Walker,

Ressler, Lu, & Davis, 2002). Although these studies clearly show intra-BLA NMDA receptor involvement in fear extinction, it is difficult to determine from these studies whether the disrupted extinction is due to the NMDA-mediated effects on extinction acquisition and/or consolidation because within-session extinction conditioned responding was not measured. Follow-up studies clarified that the drugs acting on NMDA receptors are most likely to be affecting the consolidation of extinction (Ledgerwood, Richardson, &

Cranney, 2003; Santini et al., 2001). Using freezing as an index of fear, Ledgerwood et al.

(2003) demonstrated that intra-BLA infusions of DCS had no effect on within-session extinction of conditioned freezing compared to vehicle, whilst facilitating extinction at test.

Form this, it appears that intra-BLA NMDA receptors are necessary for the consolidation of the long-term extinction memory, but not its initial acquisition. 30

However, it was observed recently that NMDA receptors containing the NR2B subunit in the LA is critical for the acquisition of extinction (Sotres-Bayon, Bush, &

LeDoux, 2007). In that study, rats were conditioned to fear a light CS by pairing it with shock. Rats then received extinction training either 2 hrs or 24 hrs afterwards. As measured by freezing, it was found that both systemic and intra-LA NR2B blockade with ifenprodil interfered with extinction acquisition, regardless of the conditioning-extinction interval.

This result, combined with other findings with different drugs acting on NMDA receptors, suggests that the role of different intra-BLA NMDA receptor subunits is dissociated at various stages of extinction.

Related to these findings on NMDA neurotransmission is immunohistochemical data showing post-extinction mitogen-activated protein kinase/extracellular signal- regulated kinase (MAPK/ERK) activation in the BLA. As mentioned previously, these molecules are a part of the intracellular events triggered by NMDA neurotransmission.

These intracellular events are important for the formation of long-term memory, and they appear to be also involved in fear extinction. For example, Davis and his colleagues showed that intra-BLA infusion of the MAPK inhibitor PD98059 before extinction blocked long-term extinction (Lu, Walker, & Davis, 2001). In that study, however, within-session extinction data was not collected, and intra-BLA MAPK/ERK activation was not measured.

Nevertheless, a recent study confirmed that there is MAPK/ERK activation in the BLA 1 hr post-extinction, and additionally reported that pre-extinction infusions of the MAPK inhibitor U0126 into the BLA completely blocked extinction acquisition in mice (Herry,

Trifilieff, Micheau, Luthi, & Mons, 2006).

The studies described so far reveal fear extinction shares similarities with fear acquisition, as both are disrupted by similar pharmacological manipulations. These 31

manipulations include NMDA receptor antagonism (Falls et al., 1992; Miserendino et al.,

1990) and inhibition of MAPK (Lin, Yeh, Lu, & Gean, 2003c; Lu et al., 2001) in the amygdala. Additionally, both acquisition and extinction of fear appear to require protein synthesis in the amygdala (Lin et al., 2003c; Schafe, Nader, Blair, & LeDoux, 2001). The similarities between acquisition and extinction of fear strongly indicate that fear extinction is new learning that requires cellular plasticity not unlike fear acquisition.

However, extinction must require neural processes that are distinct from fear acquisition because these two forms of learning have opposite behavioural consequences.

As mentioned earlier, Harris and Westbrook (1998b) showed that increased GABAergic inhibitory neurotransmission is critical for extinction, a finding that differentiates extinction from acquisition of fear. That is, increased GABAergic activity is not critical for acquisition of fear. In fact, increased GABA disrupts fear conditioning (e.g., Harris &

Westbrook, 1994). Further evidence of GABAergic involvement in extinction comes from a study that showed the transcriptional gene for the GABAA receptor clustering protein gephyrin was significantly increased 2 hrs post-extinction, as was the surface expression of

GABAA receptors in the BLA (Chhatwal, Myers, Ressler, & Davis, 2005). Hence, alterations in inhibitory GABAergic synapses appear to occur in the BLA as a consequence of fear extinction. Consistent with this idea, post-extinction infusion of the GABAA agonist muscimol into the BLA has been shown to facilitate extinction (Akirav, Raizel, & Maroun,

2006). These findings on GABAergic neurotransmission demonstrate how extinction is different from acquisition of conditioned fear.

Electrophysiological studies also show that extinction has a distinctive physiological marker compared to acquisition of fear (e.g., Quirk et al., 1995; Quirk et al.,

1997; Repa, et al., 2001). In Quirk et al. (1995)’s study, short latency (< 20 msec after tone 32

onset) cell spiking in LA in response to an auditory CS was increased by fear conditioning and returned to pre-conditioning levels after extinction, suggesting that extinction resets some aspects of LA physiology to the preconditioning state. However, changes in the synchrony of spontaneous firing between LA cell pairs after conditioning did not revert to the original state following extinction trials, a result that strongly suggests that LA cells participate in the permanent storage of conditioned fear. Another study identified that there are two types of cells in the dorsal subnucleus of the LA (dLA) that respond differently to extinction (Repa et al., 2001). One type of cells, exclusively belonging to the dorsal region of LAd, showed enhanced CS-evoked activity during early extinction trials, which fell back to baseline levels by the end of extinction. However, the other type of cells (mostly in the ventral region of dLA) continued to display elevated CS-evoked firing rates in early and late extinction trials. Collectively, these findings demonstrate extinction-related plasticity in the LA.

It is worth noting that although these electrophysiological studies demonstrate that

LA cell activity consistently changes due to extinction, these studies do not localise where these changes in cell activity are initiated. That is, the source of these changes could be occurring within LA or could be triggered anywhere upstream of LA. The amygdala is not only connected to the sensory cortices and brainstem areas that are responsible for fear acquisition and expression, but it is also densely connected with prefrontal areas

(McDonald et al., 1996; Price, 2003). In fact, the amygdala is one of the major neural structures that receive prefrontal projections. Indeed, recent findings indicate that along with the amygdala, prefrontal areas are critically involved in extinction of conditioned fear.

Amongst all the areas in the prefrontal cortex, research indicates that the mPFC is a particularly important neural structure for extinction. 33

The medial prefrontal cortex

Damage to the prefrontal cortex had long been known to produce failure in the ability of animals to change or inhibit responses in certain cognitive tasks (e.g., Frysztak & Neafsey,

1994; Kolb, 1984). For example, Kolb (1984) reported that rats with ventral lesions of the frontal cortex continued to make a previously reinforced response (e.g., bar presses for food) long after the response was no longer reinforced. That is, lesions of the ventral prefrontal cortex attenuated the extinction of bar-pressing behaviour. Kolb (1984) suggested that it may be the case that ventral prefrontal lesions do not affect the acquisition of the task but significantly interfere with the readjustment in responding to the CS when it no longer signals food.

Current interest in the role of the mPFC in fear extinction began when LeDoux and his colleagues demonstrated that rats with permanent lesions of the mPFC showed increased resistance to extinction (Morgan et al., 1993). In that study, rats were given either electrolytic or sham lesions of the mPFC before receiving tone CS-shock US pairings. Then the rats were tested daily with CS only trials (i.e., extinction) until their CS-elicited freezing levels were very low. During fear conditioning, there were no observed differences between sham- and mPFC-lesioned rats, suggesting that mPFC lesions had no effects on fear acquisition. On the other hand, it was observed that rats with mPFC lesions took significantly longer to reach the extinction criterion compared to the sham-lesioned group

(14 vs 8 days, respectively). Interpreting these findings in the light of previous research on the involvement of mPFC in cognition, Morgan et al. (1993) proposed that the resistance to extinction following mPFC damage reflected perseverative tendencies in emotional learning. A follow-up study by Morgan and LeDoux (1995) compared the effects of 34

permanent damage to the dorsal mPFC (dmPFC) and ventral mPFC (vmPFC). They found that pre-training vmPFC lesions had no effect on the amount of CS-elicited freezing during extinction but greatly increased the number of CS-alone trials required to extinguish the CS, whereas damage to the dmPFC led to an increased resistance to extinction as well as an increased amount of freezing expressed on each trial.

Because damage to the vmPFC selectively impaired extinction over days without increasing the expression of fear within extinction sessions, subsequent research on the role of prefrontal areas in extinction mainly investigated the vmPFC rather than the dmPFC. For example, Quirk and his colleagues showed that pre-training electrolytic lesions of the infralimbic cortex (IL) of the vmPFC had no effects on extinction acquisition but selectively impaired the retrieval of extinction learning the next day (Quirk et al., 2000).

Lesions of vmPFC that excluded the major parts of the IL, on the other hand, did not appear to have any effects. This result suggests a specific role for the IL in the consolidation and/or retrieval of extinction, rather than extinction acquisition.

Parallel to these findings using permanent lesions of the mPFC on extinction are thse using localised infusion of various drugs into the mPFC. As mentioned earlier, formation of long-term memory entails NMDA-dependent molecular cascade that ultimately leads to protein synthesis (Figure 1.2) (Kandel, 2001). This molecular cascade in the BLA is also important for extinction. Recent studies suggest a similar molecular cascade also subserves extinction in the mPFC. For example, infusion of the NMDA antagonist CPP into the vmPFC immediately before or after extinction training impaired extinction recall at test the next day (Burgos-Robles, Vidal-Gonzalez, Santini, & Quirk,

2007). CS-elicited freezing during extinction training was unaffected even when the drug was given before the extinction session, which strongly suggests that antagonising NMDA 35

activity in the vmPFC disrupted the consolidation of extinction memory, rather than extinction acquisition. Interestingly, Burgos-Robles et al. (2007) also showed that CPP infusion into the vmPFC prior to fear conditioning had no effects on the acquisition of fear, suggesting a selective role of NMDA receptors in the vmPFC in fear extinction.

Extinction is also disrupted by MAPK/ERK signalling pathway blockade in the mPFC (Hugues, Chessel, Lena, Marsault, & Garcia, 2006; Hugues, Deschaux, & Garcia,

2004). In Hugues et al. (2004), rats were conditioned to fear an auditory CS, and this fear was extinguished across 3 extinction sessions (in the same day). After the last extinction session, rats were either infused with the selective MAPK inhibitor PD098059 or vehicle into the mPFC. When the rats were tested the next day, it was found that immediate post-extinction infusion of PD098059 disrupted extinction. However, Hugues et al. (2004) did not actually measure MAPK/ERK immunoreactivity in the mPFC after extinction. An indirect evidence for MAPK activation in the mPFC in extinction comes from a study that showed extinction training stimulates the immediate early gene c-Fos in the mPFC, and microinfusion of the protein synthesis inhibitor anisomycin into the mPFC blocks extinction (Santini, Ge, Ren, De Ortiz, & Quirk, 2004). Taken together, these results indicate that extinction requires molecular processes in the mPFC, not unlike the processes that occur in the BLA for acquisition of fear. However, it is clear that more studies are needed to delineate the role of mPFC in fear extinction, as some studies failed to replicate the above findings (e.g., Garcia et al., 2006; Gewirtz, Falls, & Davis, 1997).

Especially convincing evidence for mPFC involvement in fear extinction comes from a series of electrophysiological studies by Quirk and his colleagues (see Quirk, Garcia,

& Gonzalez-Lima, 2006, for review). This research strongly suggests that the vmPFC, especially the IL, is necessary for long-term extinction of conditioned fear. A seminal study 36

by Milad and Quirk (2002) investigated the role of mPFC in extinction by recording neuronal activity from the mPFC across Pavlovian conditioning, extinction, and test. The target cells were located in the IL, prelimbic cortex (PL), or medial orbital cortex (MO).

Consistent with the previous findings on mPFC lesions not affecting fear acquisition (e.g.,

Quirk et al., 2000), CS-elicited activity was not found during the conditioning session in the IL (and other brain areas measured). Interestingly, CS-elicited activity was still not found in IL neurons during extinction training. By the next day, however, robust

CS-elicited activity in the IL was observed at test. This increase in the neuronal tone response was inversely correlated with the amount of freezing at test, thereby showing that the stronger the extinction expression at test (i.e., less freezing), the stronger the CS-elicited neuronal activity in the IL. To examine whether the neuronal activity in the IL underlies extinction expression at test, Milad and Quirk (2002) then electrically stimulated the IL during CS-alone presentations in rats that had previously been conditioned. Control groups received CS-alone presentations either with unpaired IL stimulation, or no stimulation. It was observed that pairing the tone CS with brief IL stimulation reduced freezing in rats during extinction training, demonstrating that IL stimulation paired with the CS is sufficient to simulate extinction. Remarkably, this facilitated extinction persisted until the test next day.

The finding that IL neurons are responsive to the fear-eliciting CS after extinction, but not during extinction itself, indicates IL-dependent mechanisms are responsible for long-term but not short-term extinction memory. Similarly, Garcia and colleagues observed that induction of long-term potentiation (LTP) by high frequency stimulation of mediodorsal thalamic inputs to mPFC is associated with the maintenance of extinction but not its acquisition (Garcia, 2002; Herry & Garcia, 2002; 2003; Herry, Vouimba, & Garcia, 37

1999). LTP is a marker for activity-dependent synaptic plasticity in the mammalian brain, and is triggered by NMDA receptor and/or LVGCC activation (Figure 1.2) (Bliss &

Collingridge, 1993). Further, when extinction is accompanied by a decrease in mPFC activity, a full recovery of conditioned fear responses is seen at test (Herry & Garcia, 2002,

2003). Thus, findings from both the Quirk and Garcia groups suggest that the mPFC is involved in the retrieval (or consolidation) of extinction.

To summarise, it appears that extinction consolidation involves NMDA-dependent plasticity in the vmPFC and the BLA. However, it is yet unclear how extinction memory is acquired. It is hypothesised that extinction is acquired during CS-alone presentations via the BLA activating the IL. Indeed, BLA sends direct excitatory projections to the IL

(Conde, Maire-Leporvre, Audinat, & Crepel, 1995; McDonald et al., 1996; Price, 2003;

Sesack, Deutch, Roth, & Bunney, 1989). The IL in turn sends a strong excitatory input to the LA and GABAergic intercalated (ITC) cells (Millhouse, 1986; Price, 2003). The IL also receives heavy projection from the PL, which is densely connected to the BA (McDonald et al. 1996; Sah, Faber, De Armentia, & Power, 2003). Thus, the amygdala can send information to the vmPFC, which in turn can modulate the amygdala as vmPFC is densely connected to its different subregions.

Compared to what we know about acquisition of extinction, how extinction, once acquired, may be expressed is better known. As mentioned earlier, the CeA modulates the expression of conditioned fear responses by means of projections to midbrain and hypothalamic sites (Figure 1.1). Given that most mPFC projections to the amygdala are excitatory, it is believed that mPFC inhibition of amygdala output involves the activation of 38

inhibitory neurons within the amygdala to inhibit the CeA (see Quirk & Mueller, 2008, for review). Interestingly, prefrontal inputs to the amygdala terminate almost entirely on dendritic spines (Smith, Pare, & Pare, 2000), suggesting that the main targets of mPFC axons are projection neurons, such as the ITC neurons. Hence, the current dominant model of extinction maintains that extinction expression relies on feedforward inhibition of the

CeA via activation of GABAergic ITC cells (Figure 1.6A) (e.g., Pare, Quirk, & LeDoux,

2004; Quirk, Likhtik, Pelletier, & Pare, 2003). In support of this model, it has been shown that stimulating the IL area decreased the excitability of neurons of the CeA, thereby decreasing the expression of conditioned fear (Milad, Vidal-Gonzalez, & Quirk, 2004;

Quirk et al., 2003). Further, it was demonstrated that disinhibition of IL neurons by the

GABAA receptor antagonist picrotoxin increased c-Fos expression in ITC cells (Berretta,

Pantazopoulos, Caldera, Pantazopouls, & Pare, 2005).

The role of the hippocampus in extinction of conditioned fear

A major characteristic of fear extinction is its context-dependence, as shown by behavioural phenomena such as reinstatement, renewal, and spontaneous recovery. The neurobiological models of extinction explains the context-dependence of extinction by suggesting that the hippocampus modulates the mPFC, which in turn modulates amygdala output (e.g., Bouton et al., 2006; Quirk & Mueller, 2008; Sotres-Bayon et al., 2004).

It is a widely-held belief that the hippocampus is critical for contextual learning

(e.g., Anagnostaras, Gale, & Fanselow, 2001; Good & Honey, 1991; Hirsh, 1980). A number of studies have demonstrated that hippocampal lesions disrupt freezing to a context that was previously paired with shock (Anagnostaras, Maren, & Fanselow, 1999; Kim & 39

Fanselow, 1992; Maren & Holt, 2000; Phillips & LeDoux, 1992). Typically, in these studies rats receive tone CS-shock US pairings in a distinctive context. Subsequently, conditioned rats display freezing to both the CS and the context. Rats that receive lesions of the dorsal hippocampus (DH) prior to or soon after conditioning, however, exhibit dramatically reduced freezing to the context although they continue to show CS-elicited freezing. The results cannot be due to hippocampal lesions causing general hyperactivity

(that interferes with freezing) as CS-elicited freezing is unaffected. Further, hippocampal lesions leave ‘remote’ context-elicited fear intact. That is, if the rat receives hippocampal lesion 52 days after conditioning to a context, freezing elicited by that context is unaffected by the lesion (Anagnostaras et al., 1999).

The hippocampus also appears to be necessary for utilising contextual representations to guide memory retrieval (Good & Honey, 1991; Holt & Maren, 1999;

Honey & Good, 1993; Maren & Holt, 2000). Honey and Good (1993) used a latent inhibition procedure to demonstrate this. Latent inhibition refers to how conditioning is attenuated if the animal receives extensive CS pre-exposure before conditioning. Like extinction, latent inhibition is context specific, as it has been shown that if the CS is pre-exposed in one context and conditioning occurs in another, then latent inhibition is disrupted (see Bouton, 1991, for review). Honey and Good (1993) showed that excitotoxic lesions of the hippocampus (using ibotenic acid) disrupted only the context-specific aspect of latent inhibition. That is, hippocampal-lesioned rats displayed low levels of CR (i.e., latently inhibited fear) regardless of the contexts they were pre-exposed and conditioned in.

This finding supports the suggestion that the hippocampus is involved in various aspects of contextual learning (e.g., Hirsh, 1980).

To examine whether the hippocampus is also involved in the context-specificity of 40

extinction, Bouton and his colleagues electrolytically lesioned the hippocampal system (i.e., fimbria-fornix and DH) before training and extinguishing rats on a conditioned suppression task (Wilson et al., 1995). They reported that rats failed to show reinstatement following lesions of the hippocampus, whereas ABA renewal remained intact. This finding was replicated with excitotoxic lesions of the hippocampus (Frohardt et al., 2000). Because renewal was still observed after hippocampal lesions, Bouton and his colleagues concluded that the hippocampal system may be important in the formation of the context-US association important for the reinstatement effect, but not in other types of learning about context. This conclusion is different to the one made by Honey and Good (1993).

However, there are other findings showing that the hippocampus may be involved in the renewal effect. Specifically, Maren and his colleagues used reversible inactivation of the DH to investigate the role of the hippocampus in renewal of extinguished fear

(Corcoran, Desmond, Frey, & Maren, 2005; Corcoran & Maren, 2001, 2004). These studies demonstrated that after conditioning and extinction training, pre-test inactivation of DH

(using high doses of the GABAA receptor agonist muscimol) reliably disrupted AAB and

ABC renewal but not ABA renewal. These results are not inconsistent with Wilson et al.

(1995) in that ABA renewal remained unaffected by DH lesion. Nevertheless, there is one study that has shown disrupted ABA renewal following hippocampal lesions; Ji and Maren

(2005) demonstrated that pre-training electrolytic lesions of DH disrupted ABA renewal.

Considering that ABA renewal is the most robust type of renewal than the others, it may be the case that ABA renewal is much stronger than other types of renewal, so is relatively difficult to eliminate. Although the evidence is far from conclusive, finding that hippocampal lesions affect ABC and AAB renewal does suggest that the hippocampus is important for context-dependent extinction. Hence, the current models of extinction state 41

that hippocampus mediates context-specific effects observed in extinction.

Interestingly, there is evidence suggesting that the hippocampus may also be necessary for extinction learning in and of itself. For example, in Corcoran et al. (2005), rats received either vehicle or muscimol infusion into the DH before extinction and were tested the next day (drug-free) in either in the same or different context to where extinction was given. It was observed that pre-extinction inactivation of DH blocked extinction because when rats were tested in the same context as extinction, group that received muscimol froze significantly higher than did the vehicle group. More evidence for the hippocampus’s role in extinction acquisition comes from a recent study analysing synaptic plasticity changes in the projection from the hippocampus to the mPFC (Farinelli,

Deschaux, Hugues, Garcia, 2006). That study reported that the hippocampus–mPFC pathway displays LTP-like changes following extinction training. Further, suppression of extinction-related hippocampus–mPFC LTP by hippocampal low-frequency stimulation, applied immediately after extinction training, led to disrupted long-term extinction. As extinction training appears to normally involve learning about the context where extinction occurs, it may be the case that the hippocampus is necessary for extinction acquisition in certain circumstances.

The importance of the hippocampus in extinction of conditioned fear in the adult rat leads to some interesting hypotheses regarding extinction in the developing rat. Specifically, the hippocampus is a late-maturing structure (Wilson, 1984), and it has been observed that young rats are impaired in learning about the context-US association (e.g., Pugh & Rudy,

1996; Rudy, 1993; Rudy & Morledge, 1994). Hence, it can be said that immature rats have

‘natural lesion’ of the hippocampus (Fanselow & Rudy, 1998). Using the developing rat 42

therefore provides a unique approach in the study of extinction of conditioned fear.

Specifically, young rats may fail to learn extinction, considering that the hippocampus may be necessary for extinction acquisition in the adult rat. Alternatively, extinction may be more robust in young rats because young rats may not show context-specificity of extinction (e.g., renewal), due to their immature hippocampus.

IV. CONTEXTUAL LEARNING IN THE DEVELOPING RAT

The hippocampus is a late-maturing neural structure that undergoes major anatomical and physiological changes after birth. For example, almost all the granule cells in the rat dentate gyrus are generated postnatally (Schlessinger, Cowan, & Gottlieb, 1975). In addition, synaptogenesis, hormone and neurotransmitter receptor site development, and myelination continue throughout the hippocampal formation into the fourth postnatal week (i.e., 22 days of age; Baudry, Arst, Oliver, & Lynch, 1981; Crain, Cotman, Taylor, & Lynch, 1973;

Jacobson, 1963).

Consistent with these findings on hippocampus immaturity, it has been observed that young rats are impaired in hippocampus-dependent tasks such as contextual conditioning (Pugh & Rudy, 1995; Rudy, 1993; Rudy & Morledge, 1994). Rudy (1993) was the first to examine conditioned fear to both a context and an auditory CS in 18- and

23-day-old rats, using a single CS-footshock US pairing. Twenty-four hrs later, rats were tested for their context- and CS-elicited freezing. It was observed that both 23- and

18-day-old rats exhibited robust freezing to the clicker CS, however, only the 23-day-old rats displayed context-elicited freezing. That is, 18-day-old rats failed to show any 43

conditioning to the context. In a follow-up study, Rudy and Morledge (1994) demonstrated that 18-day-old rats could show freezing to the conditioned context immediately after the conditioning episode but failed to show any freezing to the conditioned context when tested after a 24 hrs retention interval. They concluded that 18-day-old rats may learn about the context but cannot maintain a long-term contextual representation.

Based on these findings on contextual learning in the developing rat, one would predict that young rats would not exhibit context-specificity of extinction. For example, rats younger than 23 days of age should not display either the renewal or the reinstatement of an extinguished CR. Also, considering that the hippocampus may also be critical in extinction acquisition in the adult rat, it is possible that acquisition of extinction may be fundamentally different in young rats due to their hippocampus immaturity.

IV. EXPERIMENTAL RATIONALE

The aim of this thesis is to explore the mechanisms underlying fear extinction in the developing rat. Investigating extinction in the developing animal not only provides a unique way of examining the processes involved in extinction, but is also important because of the long- and widely-held belief that early learning experiences have a profound impact on later behaviour (e.g., Harlow, 1959; Mineka & Zinbarg, 2006). Jacobs and Nadel

(1985; 1999), for example, suggested that fear acquired early in development is particularly resistant to extinction, and forms the basis of anxiety disorders emerging later in life.

However, there is very little, if any, empirical data to support this assertion. In fact, in one of the few systematic studies on extinction across age, Campbell and Campbell (1962) 44

trained 25-, 50-, and 100-day-old rats on an avoidance procedure. When tested across 4 days, rats at all three ages exhibited comparable rates of extinction. This result was replicated in 23-day-old rats and adult rats using conditioned FPS (Richardson, Tronson,

Bailey, & Parnas, 2002). However, neither of those studies examined extinction-related phenomena like renewal or reinstatement. Furthermore, nothing is yet known about extinction of conditioned fear in rats younger than 23 days of age.

Thus, the present thesis systematically examined extinction and its related phenomena in the developing rat. Pavlovian conditioning consisting of pairings of an auditory CS and footshock US was used. Rats were either 16 or 23 days of age at the beginning of each experiment, and extinction was given at either 17 or 24 days of age.

These ages were chosen based on previous behavioural and neural findings that indicate hippocampus maturity at 24 days of age and immaturity at 17 days of age (e.g., Jacobson,

1968). Freezing was used as a measure of fear learning throughout, as rats at these ages reliably exhibit conditioned freezing to an auditory CS. The first series of experiments in

(Chapter 2) examined whether there is a developmental difference in susceptibility to reinstatement following extinction of conditioned fear. The second series of experiments

(Chapter 3) further examined whether there are developmental differences in extinction with behavioural and pharmacological manipulations. Specifically, after conditioning and extinction, rats were either tested in the same or a different context to extinction to investigate developmental differences in renewal of extinguished fear. Additionally, rats were injected with the GABAA receptor inverse agonist FG7142 before test because

FG7142 has been shown to recover an extinguished fear response in the adult rat (Harris &

Westbrook, 1998b). The last series of experiments (Chapter 4) then investigated whether the amygdala was involved in extinction and re-extinction in the developing rat. In these 45

experiments, the amygdala was temporarily inactivated using localised infusion of the sodium-channel blocker bupivacaine. The results from these experiments are of interest because no-one had yet examined whether the amygdala was involved in extinction of conditioned fear in rats these ages. In other words, nothing is known about the neural structures involved in extinction of conditioned fear in the developing rat. Additionally, amygdala involvement in re-extinction of a previously extinguished CS was examined.

This is because recent studies in the adult rat showed that re-extinction is a useful of way in elucidating the mechanisms behind extinction (e.g., Morgan, Schulkin, & LeDoux, 2003).

Considering the differences in hippocampus maturity in 17- and 24-day old1 rats, I predicted that rats these ages would exhibit interesting differences in fear extinction. Hence, the results of this thesis would provide valuable information on how extinction works across development, and the findings could have substantial theoretical and practical implications. The importance of developmental research into extinction is neatly captured by a quote in Jacobs and Nadel (1999, pp. 99-100) – “The immature nature of the infant’s nervous system sets certain constraints on the inner world of the child. Perception of everyday scenes, understanding of parts, wholes, relationships, and properties, and our ability to think about others and ourselves as actors in an intricately structured world all seem to depend in certain ways upon regions of the brain that may not be fully functional in early life… This fact has cardinal implications for understanding cognitive development…

It also has, we suggest, implications for an understanding of phobic and panic reactions.”

1 The age labels regarding the experiments in the present thesis refer to the rats’ age when the extinction training was given, unless specified otherwise. 46

Chapter Two.

REINSTATEMENT OF EXTINGUISHED FEAR IN THE DEVELOPING RAT

In the adult rat, it is widely accepted that the decrease in the CR following extinction reflects new learning of a second, competing association that inhibits the expression of the original association (see Bouton et al., 2006, for review). The primary evidence for this view comes from behavioural studies where performance to an extinguished CS recovers without subsequent re-training of the CS-US association. Reinstatement, for instance, is the recovery of an extinguished CR following a post-extinction presentation of a ‘reminder’ cue (Bouton, 1984; Delamater, 1996; Pavlov, 1927; Rescorla & Heth,

1975). In extinction of conditioned fear, the reminder cue typically involves a re- exposure to the US alone. The recovery of extinguished fear following this reminder treatment indicate that the decrease in the CR after extinction training does not reflect an erasure of the original association, but is instead due to the suppression or inhibition of the original association by a competing association acquired during the extinction session.

In the reinstatement effect, the context-US association that forms during the reminder episode appears to disambiguate which memory (CS-US vs CS-no US) is to be retrieved at test. That is, in the context where the reminder US was given, the CS-US memory is retrieved; but in a context where the reminder was not given, the CS-noUS memory is retrieved. This view is supported by studies showing the reinstatement effect to be context-dependent; that is, post-extinction reinstatement is restricted to the context where the reminder had been presented (e.g., Bouton, 1984; Bouton & Bolles, 1979a;

Bouton & King, 1983; Frohardt et al., 2000; Wilson et al., 1995). Additionally, these studies show that the reinstatement effect is not a simple summation of context-elicited 47

freezing and the remaining CS-elicited freezing after extinction. For example, in Bouton and Bolles (1979a), initial exposure to the context where the reminder was presented did not trigger any fear responses. When the extinguished CS was introduced in this context, however, rats showed robust fear responses. From this, it appears that a direct association between the context and the US reminder disinhibits responding to the extinguished CS. Consistent with this view, Bouton and his colleagues reported that rats failed to show reinstatement following permanent lesions of the hippocampus (Frohardt et al., 2000; Wilson et al., 1995), a structure critically involved in contextual learning

(see Anagnostaras et al., 2001, for review).

The apparent importance of the hippocampus for the reinstatement effect in the adult rat leads to an interesting hypothesis regarding extinction in the developing rat.

The hippocampus is a late-maturing structure (Wilson, 1984), and as a consequence, young rats are impaired in learning about context (Pugh & Rudy, 1995; Rudy, 1993;

Rudy & Morledge, 1994). Based on these findings, one would predict that young rats would not exhibit reinstatement of an extinguished CR. However, previous research on spontaneous forgetting in the developing rat would suggest that these young rats may still show the reinstatement effect (e.g., Campbell & Jaynes, 1966). Although no study to date has systematically explored reinstatement after extinction of conditioned fear in the developing rat, reinstatement of spontaneously forgotten fear in the developing rat is a well-established finding.

Reinstatement of forgotten fear following infantile amnesia

Infant animals display very rapid rate of spontaneous forgetting. This reduced ability to remember experiences that happened early in development is referred to as

‘infantile amnesia’. For example, a 20-year-old cannot remember events that occurred 48

when he/she was 1, whereas a 40-year-old can remember many of the events that occurred when he/she was 21. Freud (1920) was the first to use the term infantile amnesia to describe this observation in humans, but the phenomenon had received attention from scientists as early as 1895 (Henri & Henri, 1895). Interestingly, infantile amnesia is not a uniquely human phenomenon – many altricial animals demonstrate poor long-term retention early in development (Campbell & Spear, 1972). Altricial animals are animals in which development occurs mostly after birth (e.g., humans), as opposed to precocial species in which development is virtually complete before birth

(e.g., guinea pigs).

Among the many altricial animals that have been shown to exhibit infantile amnesia the rat is the most widely studied. The rat was first used to study infantile amnesia in 1962, when B. A. Campbell and E. H. Campbell demonstrated that infant rats forget considerably faster than adult rats. They trained 5 groups of rats on a passive avoidance procedure, in which rats were shocked while being confined in the black compartment of a black/white shuttle box. These 5 groups of rats were either 18, 23, 38,

54, or 100 days of age. Rats were then tested for retention of fear either immediately, or after a delay of 7, 21, or 42 days. Retention was measured by the passive avoidance of the black compartment. There were no differences between the groups when tested immediately after training, with all age groups avoiding the black compartment. This finding indicates that animals of all ages were able to learn and express fear. However, when tested after a delay, retention increased dramatically with age. Eighteen-day-olds showed only moderate retention when tested after 7 days, which declined even more after 42 days. In contrast, 100-day-old rats showed nearly 100% avoidance even at the

42-day interval. 49

It is important to note that in Campbell and Campbell’s (1962) study, rats of different ages displayed no differences in avoidance when tested immediately after training, indicating that the degree of original learning was equivalent across all ages.

The degree of original learning can also be assessed by examining the rate of acquisition (e.g., the number of trials to learn avoidance). Replicating Campbell and

Campbell’s (1962) general finding, Feigley and Spear (1970) found that although the rate of acquisition of avoidance learning was similar across groups, infant rats showed a significant retention deficit compared to adult rats. Furthermore, Schulenburg, Riccio, and Stikes (1971) deliberately overtrained infant rats, and observed that infant rats’ memory was still markedly inferior to that of adult rats. Schulenburg et al. (1971) concluded that the accelerated rate of forgetting by infant rats compared to adult rats is not simply a consequence of differences in their degree of original learning. It would be of little significance to investigate infantile amnesia if the events of adulthood are remembered better than those of infancy merely because they were learnt better in the first place. The interest lies in the differential retention that occurs despite equivalent learning.

Interestingly, like extinction in the adult rat, infantile amnesia is not due to an erasure of the original conditioned fear. Campbell and Jaynes (1966), in their now classic ‘reinstatement’ study, provided the first demonstration that infantile amnesia could be alleviated by reminder episodes occurring during the retention interval. In their experiment, 3 groups of 25-day-old rats were used. Two groups were trained to fear the black side of the black-white shuttle-box, whilst the remaining group was merely exposed to the apparatus (this group served as a non-conditioned control group). One of the trained groups and the control group then received an abbreviated form of the training procedure (reinstatement) 7, 14, and 21 days after the original training session. 50

The abbreviated form consisted of receiving a single shock in the black side. One week after the last reinstatement episode, all rats were tested for the passive avoidance of the black compartment. At test, it was observed that the group that was conditioned and received reinstatement treatments displayed significantly greater avoidance than the group that was conditioned but did not receive any reinstatement treatments. Moreover, the group that received the reinstatement treatments but not the initial training session did not show any avoidance behaviour. This latter finding shows that the reinstatement procedure itself did not produce the avoidance behaviour.

Reinstatement treatments used to alleviate spontaneous forgetting in the infant rat typically involves multiple, periodic re-exposures to the cues (e.g., grid floor and shock) that were present at the time of training. This procedure is different from that involved in reinstatement of extinguished fear in adult rats. Reinstatement of extinguished fear typically involves a single reminder episode that occurs 1 day before test. Hence, reinstatement of extinguished fear observed in adult rats is closer in resemblance to the ‘reactivation’ effect in the infant rat. Reactivation of forgotten fear in infant rats involves only a single re-exposure to the US that was present at the time of training. Spear and Parsons (1976) first demonstrated that a single reminder episode administered 24 hours prior to testing is as effective as multiple reminder treatments in alleviating infantile amnesia in 16-day-old rats. As in the case of Campbell and Jaynes

(1966) study, the group that received the reminder treatment without being trained did not show any conditioned response, thereby confirming that the reminder treatment itself did not produce any learning.

Taken together, alleviation of infantile amnesia by US reminder demonstrates that forgetting due to the simple passage of time in the infant rat is a retrieval failure, rather than an erasure of the original memory. Further, these findings illustrate that the 51

decrease in performance due to spontaneous forgetting in infant rats is susceptible to reminder treatments. Interestingly, reactivation of forgotten memory in the infant rat is context-independent. That is, the forgotten memory is recovered even if the US reminder is given in a different context to where the memory is tested (Spear & Parsons,

1976). Also, the fact that the reactivation effect is observed in 16-day-old rats illustrates that this effect is not dependent on contextual learning. According to these findings, the decrease in performance after extinction in infant rats may also be reinstated following a reminder treatment.

The present chapter examines whether there is a developmental difference in susceptibility to reinstatement following extinction of learned fear. Specifically, I examined whether postnatal day (P)172 rats were less likely to exhibit a return of fear following a post-extinction US reminder than were P24 rats. The current neurobiological models of fear extinction state that the hippocampus is necessary for reinstatement to occur in the adult rat (Figure 1.5B). According to these models, it may be the case that P17 rats are unable to show the reinstatement effect due to their immature hippocampus, whereas P24 rats do. On the other hand, rats both ages may display the reinstatement effect, considering the previous findings on spontaneous forgetting in the infant rat.

2 The age labels regarding the experiments on extinction described in this and other chapters in the present thesis refer to the rats’ age when the extinction training was given, unless specified otherwise. 52

General Methods

Subjects

All experiments used experimentally naive Sprague-Dawley derived rats, bred and housed in the School of Psychology, University of New South Wales. Rats were either

P16 or P23 at the start of an experiment. All rats were male, and no more than one rat per litter was used per group. Rats were housed with their littermates and mother in plastic boxes (24.5cm long x 37cm wide x 27cm high) covered by a wire lid. Animals were maintained on a 12 hours light/dark cycle (lights on at 6am) with food and water available ad libitum. All animals were treated according to the principles of animal use outlined in The Australian Code of Practice for the Care and Use of Animals for

Scientific Purposes (7th Edition), and all procedures were approved by the Animal Care and Ethics Committee at the University of New South Wales.

Apparatus

Two distinctive types of experimental chambers were used to provide different contexts.

One type (Figure 2.1A) was a set of two identical chambers that were rectangular (13.5 cm long x 9 cm wide x 9 cm high), with the front wall, rear wall, and ceiling constructed of clear Plexiglas. The floor and side walls consisted of 3 mm stainless steel rods set 1 cm apart. Two high-frequency speakers were located 8 cm from either side of the chamber. A custom built constant-current shock generator could deliver electric shock to the floor of each chamber as required. A tray of bedding was placed 10 cm below the grid floor. Each chamber was housed within a separate wood cabinet. A red light-emitting diode (LED) located on the cabinet door was the sole source of illumination in these chambers. A low, constant background noise (50d, measured by

Bruel Kjaer sound level meter, type 2235, A scale) was produced by ventilation fans 53

located within the cabinets. These chambers were wiped with tap water and the bedding was replaced after each experimental session.

The second type (Figure 2.1B) was a set of two identical chambers that were rectangular (30 cm long x 30 cm wide x 23 cm high) and wholly constructed of

Plexiglas, with the exception of the grid floor (consisting of 3 mm stainless steel rods set 1 cm apart). Instead of bedding, a clear Plexiglas sheet (35 cm x 35 cm) was placed beneath the grid floor. All the walls were transparent, except for the two side walls that consisted of vertical black and white stripes (each 5 cm wide). Two high-frequency speakers were mounted on the ceiling of each of these chambers. Each chamber was housed within a separate wood cabinet. A white LED and a red LED located on the cabinet door were the sole sources of illumination in these chambers. A low, constant background noise (48d) was produced by ventilation fans located within the cabinets.

Thus, these two sets of contexts differed primarily in terms of their size and in their visual features.

The CS was a white-noise; noise level in the chambers was increased by 8dB when the CS was presented. A computer controlled all presentations of the CS and the footshock US. The software and hardware used were developed at the University of

New South Wales.

In Experiments 1 and 2, the use of these two sets of experimental chambers were counter-balanced. In Experiments 3.1, 3.2, and 3, all training occurred in the first type of chamber (Figure 2.1A) and testing always occurred in the second type (Figure 2.1B).

Figure 2.1 (A) Context type 1 (illuminated by a red LED). (B) Context type 2 (illuminated by a red LED & a white LED). 55

Procedures

Training. Rats were placed in the conditioning chambers, and after a two-minute adaptation period, the CS was presented for 10 s and co-terminated with the shock US

(0.6 mA, 1 s). The inter-trial interval (ITI) ranged from 85s to 135s with a mean of 110s.

In Experiments 1 and 2, 16-day-old rats received 6 CS-US pairings and 23-day-old rats received 3 CS-US pairings, as pilot experiments indicated that these parameters produce roughly similar levels of performance and extinction at the two ages. In Experiment 3,

P16 rats received 2 pairings in an attempt to increase spontaneous forgetting. The ITI was 110 s.

Extinction. Extinction training consisted of 30 CS presentations in the absence of shock. Rats were placed in the experimental chambers, and after a two-minute adaptation period, the 10 s CS was presented (30 times) with a 10 s ITI.

Reinstatement. The reminder treatment consisted of a two-minute adaptation period, followed by a single shock US (0.4 mA, 1 s).

Test. All rats were tested for CS-elicited fear. Initially, their baseline level of freezing in the absence of the CS was recorded for 1 minute. The CS was then presented for 2 minutes.

Scoring and Statistics

Each animal was scored for freezing during extinction training and test. Freezing was scored by a time sampling procedure whereby each rat was scored every three seconds as freezing or not freezing. Freezing was defined as the absence of all movement other than those required for respiration (Blanchard & Blanchard, 1969). A percentage score was calculated for each animal to indicate the proportion of total observations scored as freezing. A second scorer unaware of the experimental condition of each rat scored a 56

random sample (40%) of all rats tested in the experiments presented this chapter. The inter-rater reliability was very high (r = .93).

CS-elicited freezing is difficult to detect if rats display high levels of baseline freezing. Therefore, if a rat showed more than 50% freezing during the baseline period of the test session, then it was removed from the test chamber without the CS being presented and returned to its home cage. After 5-10 minutes, the rat was returned to the test chamber for a second test of baseline freezing. This was repeated until the baseline level of freezing was less than 50%, or until three baseline freezing tests had been conducted. One P16 rat from group Reinstated in Experiment 1 and 1 rat from each group in Experiment 2 did not meet the baseline criteria after three baseline tests, so were not tested. There were no other rats re-tested for baseline in this chapter (i.e., all met criterion on the first test).

No significant differences in baseline levels freezing at test were detected in any

Experiment except in Experiment 3.3, t (18) = 2.15, p = .046 (see Table 1 for baseline means at test for all groups in each experiment described in this chapter). Because of the difference in baseline levels of freezing in Experiment 3.3, CS-elicited freezing data during the test were analysed by analysis of co-variance (ANCOVA) for all experiments using the baseline scores as the co-variate. The same results were obtained, however, whether the data were analysed with ANOVA or ANCOVA in all experiments. A p value < .05 was considered statistically significant.

Experiment Groups % Freezing

1 P17-No Reminder 28 (± 7) P17-Reminder 28 (± 8) P17-Not Trained-Reminder 14 (± 6) P24-No Reminder 7 (± 2) P24-Reminder 17 (± 7) P24-Not Trained-Reminder 22 (± 8)

2 P17-Same 19 (± 7) P17-Different 18 (± 6) P24-Same 22 (± 5) P24-Different 9 (± 3)

3.1 Immediate 1 (± 1) 48 Hrs 5 (± 3)

3.2 No Reminder 1 (± 1) Reminder 11 (± 5) Not Trained-Reminder 8 (± 7)

3.3 No Reminder 11 (± 5) Reminder* 26 (± 5)

Table 2.1 Mean (± SEM) levels of baseline freezing at test for all groups across all experiments in Chapter 2. * Indicates a significant difference to the other group/s in baseline levels of freezing. 58

Experiment 1. Reinstatement of extinguished fear in P17 and P24 rats

This experiment investigated reinstatement of extinguished fear in P17 and P24 rats. I predicted that P24 rats, like adults, would exhibit the reinstatement effect following a post-extinction presentation of the US. The question of interest was whether P17 rats would also exhibit reinstatement of an extinguished fear response.

Method

A 2 x 3 factorial design was used, in which the first factor was Age at extinction (P17 or

P24) and the second was Group (No Reminder, Reminder, or Not Trained-Reminder).

All groups were run in an ABBB condition with each letter indicating the context of conditioning, extinction, reinstatement, and testing, respectively.

On Day 1, 16- and 23-day-old rats in groups No Reminder and Reminder were trained. Rats in the Not Trained-Reminder group at each age were exposed to the experimental chamber for the same amount of time as the conditioned groups, but were not presented with either the CS or the US. On Day 2, rats in the trained groups were given extinction training. Rats in the Not Trained-Reminder groups at both ages were given an equivalent amount of exposure to the experimental chamber, but were not given any presentations of the CS. On Day 3, rats in groups Reminder and Not Trained-

Reminder received the reinstatement treatment. Rats in the No Reminder group at each age were given an equivalent amount of exposure to the experimental chamber, but were not exposed to the US. All groups were tested for CS-elicited fear on Day 4. 59

Results and Discussion

Trained rats at both ages exhibited substantial levels of freezing on the first block (3 CS presentations) of extinction (Figure 2.2) but substantially less freezing by the last block of extinction. A mixed-design ANOVA of the extinction data, grouped into 10 blocks of

3 CS presentations, yielded a significant main effect of Block, F (9, 243) = 18.40, p

<.0001, but no main effects of Age, F < 1, or Group, F < 1. There were no significant interactions. Importantly, these results indicate that the initial levels of performance, rate of extinction, and the terminal levels of performance did not differ between any of the groups. As mentioned previously, equating levels of performance as close as

Figure 2.2 Mean freezing by rats in response to the auditory conditioned stimulus during extinction training in Experiment 1. Each block represents the average freezing over 3 CS presentations. No significant differences were found between any groups at any extinction block.

possible is essential in developmental studies. If the different age groups vary substantially either in initial or terminal levels of performance, then drawing any meaningful conclusions about potential developmental differences in post-extinction reinstatement would be difficult, if not impossible. In the present experiment, equal levels of performance were observed at all points of the extinction session.

Figure 2.3 Mean (± SEM) CS-elicited freezing during test in Experiment 1. Rats of two ages (postnatal day 17 and 24) were in three conditions; P17-No Reminder (n = 8), P17-Reminder (n = 7), P17-Not Trained-Reminder (n = 8), P24-No Reminder (n = 8), P24-Reminder (n = 8), and P24-Not Trained-Reminder (n = 8). * Indicates a significant difference to the other groups of the same age.

The mean, and standard error of the mean (SEM), levels of CS-elicited freezing at test are shown in Figure 2.3. P24 rats that had been trained, extinguished, and then given a post-extinction reminder exhibited higher levels of freezing at test than did those trained and extinguished but not given the US reminder. The level of freezing in this latter group was comparable to that observed in the Not Trained-Reminder group.

Taken together, these results show that rats this age exhibited post-extinction reinstatement. In contrast, all P17 groups showed low levels of freezing at test; that is, rats this age did not exhibit post-extinction reinstatement. ANCOVA confirmed this description of the data; there was a significant main effect of Group [F (2, 40) = 6.83, p

< .005], but not Age [F (1, 40) = 3.25, p = .079]. Importantly, there was an Age x Group interaction, F (2, 40) = 3.39, p < .05. To understand the significant Age x Group interaction, separate analyses were done at each age. These tests yielded a significant group effect for the P24 rats [F (2, 21) = 15.21, p < .0001] but not for the P17 rats (F <

1). Pairwise comparisons, with Tukey’s honestly significant difference (HSD) tests, showed that the P24 rats in the Reminder group exhibited significantly more freezing than did rats in either of the other two groups at 24 days of age (p < .0001). Taken together, these analyses show that a post-extinction reminder US alleviates extinction- induced performance decrements in P24 rats but not in P17 rats.

Experiment 2. Context-specificity of the reinstatement effect

Reinstatement of an extinguished fear response was observed following a post- extinction US in P24 rats but not in P17 rats in Experiment 1. Experiment 2 attempted to replicate this finding, and also examined the context specificity of the reinstatement effect. Previous research with adult rats has generally shown that reinstatement of an extinguished fear CR is dependent on subjects being tested in the context where the post-extinction US was given.

In this experiment, some rats at each age were given the reinstatement treatment in the same context in which they would be subsequently tested, while other rats were given the reinstatement treatment in a different context from where they would be subsequently tested. Based on the results of Experiment 1, I predicted that P24 rats would exhibit reinstatement of an extinguished fear response following a post- extinction US, but that P17 rats would not. Furthermore, if reinstatement of extinguished fear observed in P24 rats in Experiment 1 is due to these rats learning the context-reminder association as in adult rats, then context-specific reinstatement should also be observed in P24 rats.

Method

A 2 x 2 factorial design was used, in which the first factor was Age at extinction (P17 or

P24) and the second factor was Reminder Context (Same or Different to the testing context). Specifically, rats in the Same condition were in an ABBB condition (same as

Experiment 1), in which the reminder US was given in the same context as where the test occurred. Rats in the Different condition were given the post-extinction US in a context different from where test occurred (i.e., ABAB condition). Note that rats in both conditions were tested in the same context as the extinction context (i.e., context B), to 63

prevent renewal. All rats were trained, extinguished, reinstated, and tested as described in the General Methods.

Results

Rats of both ages exhibited substantial levels of freezing on the first block (a block of 3 trials) of extinction, which decreased substantially by the last (i.e., 10th) block (Figure

2.4). These freezing levels are very similar to that observed in Experiment 1. A mixed- design ANOVA of the freezing performance during the first and last minute of extinction yielded a significant main effect of Block, F (1, 35) = 67.67, p < .0001, but no main effect of either Age (F < 1), or Group [F (1, 35) = 1.48, p = .232]. Further, there were no significant interactions.

Figure 2.4 Mean (± SEM) CS-elicited freezing during extinction in Experiment 2. There were no group differences. 64

Figure 2.5 Mean (± SEM) CS-elicited freezing during test in Experiment 2. Rats were tested in either the same or different context to where they were reinstated; P17-Same (n = 10), P17- Different (n = 10), P24-Same (n = 10), and P24-Different (n = 10). * Indicates a significant difference to the other group of the same age.

The mean and SEM levels of freezing to the CS on test are shown in Figure 2.5.

P24 rats exhibited high levels of freezing in the Same condition but low levels of freezing in the Different condition. P17 rats, on the other hand, showed low levels of freezing in both conditions. ANCOVA of CS-elicited freezing levels, with baseline freezing as a co-variate, yielded a significant main effect of Group, F (1, 35) = 4.13, p

< .05, and a significant Age x Group interaction, F (1, 35) = 7.69, p < .01. There was no main effect of Age (F < 1). To understand the significant Age x Group interaction, separate analyses were done at each age. In P24 rats, group Same exhibited significantly higher levels of freezing than group Different [t (18) = 5.76, p < .0001]; this difference shows that the reinstatement of extinguished fear was contextually-modulated in P23 65

rats. In contrast, there was no group difference in 17-day-old rats [t (18) = .359, p = .72].

As the mean level of freezing performance in the two groups of reinstated rats at this age were comparable to the mean level of performance of the No Reminder rats this age in Experiment 1, it appears that reinstatement of extinguished fear once again failed to occur in P17 rats. 66

Experiments 3.1, 3.2, & 3.3. Reinstatement of forgotten fear

Experiments 1 and 2 show that P24 rats exhibit reinstatement of an extinguished fear response and that this effect is dependent on the rats being tested in the same context where the post-extinction US was given. These experiments also show that P17 rats do not exhibit reinstatement of an extinguished fear response. These results suggest that extinction in P24 rats is context-dependent, whereas extinction in P17 rats is context- independent. However, it is possible that the present results were obtained because of the parameters selected. That is, the parameters of the reinstatement treatment (e.g., intensity, duration, etc.) were effective for the older rats but not for the rats conditioned, extinguished, reinstated, and tested at 16, 17, 18, and 19 days of age, respectively. One way to assess this hypothesis would be to systematically vary the reminder US following extinction. This approach, however, is not pragmatic as increasing the number and/or intensity of the reminder US would most likely lead to high baseline levels of freezing.

Therefore, the present series of experiments examined the effectiveness of the reinstatement treatment used in Experiments 1 and 2 in alleviating spontaneous forgetting, rather than extinction-induced performance decrements, in 16-day-old rats.

Spear and Parsons (1976) demonstrated that forgetting in young rats can be alleviated by a pre-test US reminder. Thus, the following experiments examined whether the same reinstatement treatment as was used in Experiments 1 and 2 would be effective in alleviating spontaneous forgetting when rats are trained at P16, reinstated at P18 and tested at P19 (these ages are match the ages at training, reinstatement, and test of P17 rats in Experiments 1 & 2). Experiment 3.1 first determined the forgetting function in

P16 rats when only 2 CS-US pairings were given at training. P16 rats should show substantial CS-elicited freezing immediately after conditioning that declines by P18. 67

Experiment 3.2 then assessed whether the same reminder used in Experiments 1 and 2 would be effective in alleviating spontaneous forgetting. Experiment 3.3 then re- examined reinstatement of spontaneous forgetting in P16 rats using slightly different parameters to Experiment 3.2, in order to bring the levels of CS-elicited freezing closer to the freezing levels observed in Experiments 1 and 2.

Experiment 3.1

Method

Two groups of P16 rats received CS-US conditioning. All training parameters were the same as in Experiment 1 except for the number of CS-US pairings (i.e., 2 instead of 6).

One group was tested immediately after conditioning and the other group was tested 48 hrs after conditioning (i.e., at P18). The testing procedure was identical to Experiment 1.

Training occurred in the first type of chamber (Figure 2.1A) and testing occurred in the second type (Figure 2.1B).

Results and Discussion

The mean and SEM levels of freezing on test are shown in Figure 2.6. Rats displayed substantial freezing to the CS immediately after training, but 48 hrs later the level of freezing was significantly reduced (i.e., infantile amnesia was observed). ANCOVA of

CS-elicited freezing yielded a significant main effect of Training-Test Interval, F (1, 13)

= 20.60, p < .001. 68

Figure 2.6 The basic forgetting function in Experiment 3.1; Immediate (n = 8) and 48 Hrs (n = 8). * Indicates a significant difference to the other group.

Experiment 3.2

Method

Experiment 3.2 consisted of 3 groups of P16 rats. The groups were: No Reminder,

Reminder, and Not Trained-Reminder. On Day 1, rats in groups No Reminder and

Reminder received training (same as in Experiment 3.1). Rats in the Not Trained-

Reminder group were given an equivalent amount of exposure to the training chamber, but not presented with the CS or the US. Two days later, rats in groups Reminder and

Not Trained-Reminder received a reminder US as in Experiment 1. Group No Reminder was given an equivalent amount of exposure to the experimental chamber, but not presented with the US. Twenty-four hrs later, all groups were tested as in Experiment 1. 69

Training occurred in the first type of chamber (Figure 2.1A); and reminder treatment and testing occurred in the second type (Figure 2.1B).

Results and Discussion

Figure 2.7 Mean (± SEM) CS-elicited freezing during test in Experiment 3.2; No Reminder (n = 8), Reminder (n = 8), and Not Trained-Reminder (n = 8). * Indicates a significant difference to the other groups.

The mean and SEM levels of freezing to the CS on test are shown in Figure 2.7. Rats trained at P16 and reinstated at P18 showed recovery from spontaneous forgetting as they exhibited high levels of freezing compared to the No Reminder and Not Trained-

Reminder groups. ANCOVA, with baseline freezing as the co-variate, yielded a significant group difference, F (2, 20) = 6.03, p < .01. This difference was due to rats in group Reminder exhibiting significantly higher levels of freezing at test than did rats in 70

the other two groups (p < .01; Tukey’s HSD test); performance in the No Reminder and

Not Trained-Reminder groups did not differ. These results show that the reinstatement treatment given in Experiments 1 and 2 was effective at alleviating a performance decrement in P16 rats. This finding suggests that the developmental difference observed in post-extinction reinstatement in Experiments 1 and 2 reflects a fundamental difference in the processes involved in extinction at these two ages rather than the reinstatement treatment being generally ineffective in the younger rats.

Experiment 3.3

Experiment 3.2 shows that spontaneous forgetting in P16 rats is alleviated by the same reinstatement treatment that was ineffective at reversing extinction-induced performance deficits in P17 rats in Experiments 1 and 2. This result suggests that the observed developmental difference in post-extinction reinstatement in Experiments 1 and 2 reflects a genuine developmental difference in extinction processes, rather than a failure of the reminder to be effective in P17 rats. However, it should be noted that the mean level of freezing observed at test in rats in the No Reminder condition in

Experiment 3.2 was lower than that observed in P16 rats in the No Reminder condition in Experiment 1. That leads to a possibility that the US reminder was more effective in

Experiment 3.2 compared to Experiments 1 and 2 because the rats in Experiment 3.2 had a greater loss of performance. This possibility was examined in Experiment 3.3; in order to increase the mean level of freezing in the No Reminder group, the number of

CS-US pairings was increased to 3 in Experiment 3.3. Because the Not Trained-

Reminder group failed to exhibit substantial freezing in any of the experiments reported 71

in this chapter (i.e., rats merely given the pre-test US did not exhibit much freezing at test), this group was not included in Experiment 3.3.

Method

Experiment 3.3 consisted of 2 groups of P16 rats: No Reminder and Reminder. All rats received 3 pairings of the CS and US on day 1. Two days later, rats in group Reminder received a reminder US as in Experiment 1. Rats in group No Reminder was exposed to the chamber only. Twenty-four hrs later, both groups were tested as in Experiment 1.

Training occurred in the first type of chamber (Figure 2.1A); and reminder treatment and testing occurred in the second type (Figure 2.1B).

Figure 2.8 Mean (± SEM) CS-elicited freezing during test in Experiment 3.3; No Reminder (n = 10), and Reminder (n = 10). * Indicates a significant difference to the other group/s.

Results and Discussion

At test, group Reminder displayed higher levels of freezing than did group No

Reminder (Figure 2.8). ANCOVA, with baseline freezing as the co-variate, yielded a significant group difference, F (1, 17) = 5.13, p < .05. The rats in the No Reminder condition exhibited levels of freezing that were comparable to those observed in the extinguished P17 rats in Experiments 1 and 2 (36% compared to 38% and 43%, respectively) whereas the rats in the Reminder group exhibited substantially higher levels of freezing. These results show that the effectiveness of the reminder US in alleviating spontaneous forgetting in Experiment 3.2 was not due to the low levels of freezing observed in that experiment. Even when levels of freezing were more comparable to that observed following extinction, the reminder US was still effective at alleviating spontaneous forgetting in P16 rats.

Discussion

The present series of experiments examined whether there are developmental differences in the reinstatement of an extinguished fear response following a post- extinction administration of the US. The results showed that a US reminder treatment led to the recovery of extinguished fear in P24 rats, an effect that was context- dependent. In contrast, P17 rats did not exhibit reinstatement of extinguished fear following a US reminder treatment. The failure to see post-extinction reinstatement in

P17 rats was not due to the US reminder being ineffective in rats this age because the same treatment given at the same age (i.e., P18) was effective at alleviating spontaneous forgetting (Experiments 3). These experiments demonstrate for the first time that P16 rats are impaired in showing the post-extinction reinstatement effect, and suggest that different processes may mediate extinction of conditioned fear early in development compared to those involved later in development.

It is worthwhile to point out that although there appear to be systematic differences in baseline freezing (with those groups showing reinstatement having higher levels than those who didn’t; Table 2.1), a summation account of these results is unlikely for several reasons. For example, these systematic differences in baseline freezing levels are extremely small and were not statistically significant in 3 of the 4 experiments that examined reinstatement. In addition, ANCOVAs were used in all experiments to take any differences in baseline freezing into account. Nevertheless, to more completely explore the possibility that the observed pattern of results were simply due to the summation of baseline freezing levels with CS-elicited freezing, I conducted a number of post-hoc analyses of the data from the two experiments where the extinction procedure was used (i.e., Experiments 1 & 2). To be more specific, the data from the Reminder groups in Experiment 1 and the data from the Same groups in 74

Experiment 2 were pooled for these analyses. These groups were chosen because these groups received both the training and the reminder treatment, and were tested in the same context as the reminder episode. With this much larger data set, I was able to (1) explicitly compare the different age groups on baseline freezing levels, and (2) explicitly compare whether the observed post-extinction reinstatement effect was due to different levels of baseline freezing in each age group. It was found with the compiled data set that P17 rats displayed low CS-elicited freezing although the baseline levels of freezing was comparable to P24 rats (Table 2.2). Further, when each age group was divided into ‘high’s and ‘low’s based on each rat’s baseline level of freezing at test

(high: 25% and above; low: below 25%), it was found that rats exhibited similar levels of CS-elicited freezing regardless of their baseline levels of freezing (Tables 2.3 & 2.4).

The results of all of these analyses conclusively demonstrate that observing (or not observing) the reinstatement effect is not determined by baseline freezing levels.

Age P17 (n = 17) P24 (n = 18)

Baseline CS Baseline CS

Mean (± SEM) 23 (± 5) 40 (± 7) 20 (± 4) 70 (± 4) Freezing

Table 2.2 Mean baseline and CS-elicited freezing at test in P17 and P24 rats that received training and the reminder treatment in Experiments 1 and 2.

The developmental dissociation in recovery of an extinguished fear response found in the present chapter is especially interesting from the perspective of current

P17 Baseline High (n = 7) Low (n = 10)

Baseline CS Baseline CS

Mean (± SEM) 45 (± 3) 39 (± 8) 7 (± 2) 42 (± 11) Freezing

Table 2.3 Mean baseline and CS-elicited freezing at test in P17 rats that received training and the reminder treatment in Experiments 1 and 2. P17 rats were divided into “high” and “low” groups based on their baseline levels of freezing. At test, rats this age displayed levels of CS- elicited freezing comparable to those freezing levels at the end of extinction regardless of their baseline freezing levels at test (i.e., no evidence for reinstatement).

P24 Baseline High (n = 10) Low (n = 8)

Baseline CS Baseline CS

Mean (± SEM) 32 (± 3) 76 (± 4) 4 (± 2) 63 (± 5) Freezing

Table 2.4 Mean baseline and CS-elicited freezing at test in P24 rats that received training and the reminder treatment in Experiments 1 and 2. P24 rats were divided into “high” and “low” groups based on their baseline levels of freezing. At test, rats this age displayed substantially high levels of CS-elicited freezing compared to the freezing levels at the end of extinction regardless of their baseline freezing levels at test (i.e., evidence for reinstatement).

neurobiological models of extinction. These models maintain that fear extinction, at least in adult rats, involves a connection between the amygdala and mPFC (Hobin,

Goosens, & Maren, 2003; Quirk et al., 2000; Sotres-Bayon et al., 2006). The basic idea is that following extinction the mPFC inhibits activity in the amygdala. This idea is supported by studies showing either impaired or modulated extinction performance in rats with mPFC lesions (Morgan et al., 1993; Santini et al., 2004). Further, a growing 76

body of neurophysiological evidence suggests that the mPFC inhibits the amygdala by activating inhibitory GABAergic neurons during the expression of extinction (see

Hobin et. al., 2003, for review). The hippocampus is suggested as a critical part of this neural circuit, responsible for the context-modulation of extinction. Indeed, there are both inhibitory and excitatory hippocampal projections to the mPFC (Carr & Sesack,

1996; Ishikawa & Nakamura, 2003). Thus, current neurobiological models of extinction maintain that the hippocampus inhibits or excites the mPFC and thereby modulates the expression of learned fear or extinction. According to these models, reinstatement may be the result of the context-reminder association triggering the hippocampal inhibition of the mPFC, preventing the inhibitory input normally applied on the amygdala following extinction. As a consequence, the learned fear response is recovered. From this perspective, the context-dependent reinstatement effect (Experiments 1 & 2) observed in the present chapter suggests that the hippocampus-mPFC-amygdala network functions in an adult-like manner by 24 days of age in rats.

In contrast to P24 rats, P17 rats failed to exhibit reinstatement of extinguished fear following a post-extinction US reminder (Experiments 1 & 2). The lack of a reinstatement effect in P17 rats may be due to a failure to establish a context-US association during the reinstatement episode because of hippocampus immaturity.

Given that Bouton and his colleagues have demonstrated that the strength of the reinstatement effect can be predicted by the strength of contextual conditioning resulting from the US reminder treatment (Bouton & King, 1983), the impaired contextual learning abilities of rats early in development would be expected to lead to impaired reinstatement. The importance of contextual conditioning in post-extinction reinstatement is highlighted by the finding that this effect is often restricted to the 77

context where the US reminder was given (e.g., Bouton & Bolles, 1979); an effect that was found in P24 rats in Experiment 2 of the present chapter.

The present findings also have important theoretical implications because the persistence of extinction displayed in P17 rats may be indicative of ‘unlearning’. The early unlearning accounts of extinction (e.g., Rescorla & Wagner, 1972) were rejected based on the findings of spontaneous recovery, reinstatement, and renewal. The failure to observe the reinstatement effect in P17 rats raises the possibility of unlearning as the underlying mechanism of extinction in rats early in development. However, further examination into the behavioural and neural differences in the modulation of extinction in the developing rat is necessary in order to investigate whether extinction is mediated by different mechanisms across development.

Chapter Three.

RENEWAL AND THE EFFECT OF PRE-TEST FG7142 INJECTION ON

EXTINGUISHED FEAR IN THE DEVELOPING RAT

In the previous chapter, it was demonstrated that there is a developmental dissociation in reinstatement of extinguished conditioned fear. Specifically, P24 rats exhibited a return of extinguished CS-elicited freezing following a US reminder treatment, whereas

P17 rats did not. This result suggests that there may be fundamentally different processes underlying fear extinction across development. The developmental difference could be due to a couple of reasons: first, P17 rats may simply fail to learn the context-

US reminder association due to the developmental lag in hippocampus maturation; and/or second, the neural basis of fear extinction may be different in P17 rats, so that extinction at this age may rely on different processes (e.g., unlearning). Both of these reasons can account for the failure to see a return of extinguished fear after the reminder treatment in P17 rats. To further investigate the developmental difference found in extinction, this chapter investigates other extinction-related phenomena such as renewal, in the developing rat.

Renewal refers to the return of a previously extinguished conditioned response when rats are tested in a context different from where extinction occurred. Similar to reinstatement, renewal also appears to be modulated by the hippocampus (Corcoran &

Maren, 2001; 2004; Corcoran et al., 2005; Ji & Maren, 2005; but see Frohardt et al.,

2000; Wilson et al., 1995). The present chapter aims to extend the findings of Chapter 2 by examining whether there also is a developmental difference in renewal of extinguished conditioned fear. Additionally, this chapter examines another manipulation that has been shown to affect extinction in adult rats. Specifically, the effect of pre-test 79

alterations in GABAergic neurotransmission on extinction in the developing rat will be examined.

GABAergic inhibition and fear

GABA is the dominant inhibitory neurotransmitter in the mammalian central nervous system, both with respect to the number of synapses and to functional relevance

(Olsen, 2002). GABAergic neurons fall into two major categories. One includes projecting neurons with long axons that innervate structures located at various distances from the cell body. The other includes interneurons with short axons that do not leave the region where their cell bodies are located. Most GABAergic neurons fall into the latter category. These GABAergic interneurons play an important part in the local modulation of information processing and in the prevention of excessive neuronal activity (Haefely, 1990).

GABAergic neurons exert their inhibitory effect when the released GABA binds to GABA receptors. Among three major types of GABA receptors that are presently known, the GABAA receptor is the most important from both the physiological and pharmacological points of view (Haefely, 1990). Also, the GABAA receptor is the only type of GABA receptor that has benzodiazepine receptor sites (Olsen, 2002). Therefore, only GABAA receptor neurotranmission will be mentioned in detail, as previous research has primarily used benzodiazepines to assess the role of GABAergic neurotransmission in memory.

Benzodiazepines are a class of agonists that bind to GABAA receptors to facilitate the inhibitory action of GABA (Haefely, 1990). Benzodiazepines do not mimic GABA itself, as they are active only in the presence of GABA (Costa, Guidotti,

& Mao, 1975). GABAergic neurotransmission is facilitated by benzodiazepines because 80

benzodiazepines increase the binding of GABA to GABAA receptors (Haefely, Kyburz,

Gerecke, & Mohler, 1985). Behaviourally, benzodiazepines are anxiolytic drugs, reducing the expression of fear and anxiety responses elicited by unlearned (i.e., innate) or learned sources of danger. For example, systemic injection of these drugs reduces

FPS (Davis, 1979). Also, it has been found that these drugs reduce the freezing, analgesic, and passive avoidance response provoked by a chamber associated with shock or a heated floor (Harris & Westbrook, 1994; 1996).

The decrease in fear expression induced by benzodiazepines indicates that the neural mechanisms underlying fear are inhibited by GABAergic neurotransmission.

This conclusion is supported by the observation that GABAA receptors are particularly abundant in the amygdala (McDonald, 1985), a structure critical for learned and unlearned fear (Davis & Whalen, 2001). Further, it has been demonstrated that microinfusion of benzodiazepine into the amygdala reduces conditioned fear responses

(Harris & Westbrook, 1995; 1998a). In one study, Harris and Westbrook (1995) first trained rats on step-down passive avoidance procedure. Specifically, rats were exposed to a hot-plate that was 54 °C for conditioned rats and 23 °C for non-conditioned rats.

Twenty-four hrs later, memory retention was measured as the latency of rats to step- down onto the hot-plate from a wooden platform. Prior to the test, rats were microinfused with either saline or midazolam (benzodiazepine) into the BLA. At test, conditioned rats that received saline exhibited significantly longer latency to step-down than non-conditioned rats. Infusion of midazolam, however, reduced the step-down latency (i.e., less memory) in conditioned rats. This result shows that benzodiazepines inhibit conditioned fear responses.

In contrast, benzodiazepine receptor inverse agonists (e.g., -carbolines) reduce the inhibitory effects of GABA by interfering with GABAergic neurotransmission. 81

Benzodiazepine receptor inverse agonists are purported to be anxiogenic (i.e., facilitates the expression of fear and anxiety), having effects that are opposite to benzodiazepines.

Thus, one might expect these compounds to increase, produce, or potentiate fear responses. Indeed, systemic administrations of high doses of -carbolines have been reported to provoke general analgesia (Fanselow & Kim, 1992), changes in heart rate

(Sanders & Shekhar, 1991), and potentiate startle responses (Hijzen & Slangen, 1989).

Intuitively, these effects of -carbolines (benzodiazepine receptor inverse agonist) are in fact opposite to the effects of benzodiazepines (benzodiazepine receptor agonist).

If GABAergic neurotransmission inhibits fear, then -carbolines could also potentiate fear responses under circumstances when fear is inhibited (e.g., after extinction). As mentioned previously, extinction in adult rats does not destroy the original memory. Rather, extinction appears to be a new, inhibitory learning that competes with the retrieval of the original memory. Then, -carbolines may attenuate extinction by reversing the inhibitory action of GABAergic neurotransmission.

Additionally, the current neural models of extinction state that the reduced responding observed following extinction is due to the inhibition of amygdala activity by the mPFC

(e.g., Quirk et al., 2000), and this inhibition is thought to be due to the activation of

GABAergic interneurons and/or projecting neurons during the expression of extinction

(see Hobin et al., 2003, for review). According to these models, -carbolines should suppress GABAergic inhibitory activity and recover extinguished fear responses.

Harris and Westbrook (1998b) provided the first demonstration of this idea by showing that pre-test systemic injections of FG7142 (a -carboline) restored extinguished fear responses in adult rats. Moreover, FG7142 did not elicit freezing in rats that had not previously been shocked, indicating that FG7142 itself does not induce freezing. A follow-up experiment ruled out state-dependent account of the findings by 82

showing that FG7142 failed to reverse latently inhibited fear. This result indicated that

FG7142 does not simply cause anxiety in the animal, thereby aiding retrieval of conditioned freezing response. From this, GABA appears to be involved in inhibition of conditioned fear following extinction.

The present chapter examines extinction in the developing rat by exploring the renewal and FG7142 effects in P17 and P24 rats. Firstly, I examined whether 17- and 24-day-old rats exhibit a return of fear if tested in a different context to the extinction context.

Based on the results of Chapter 2, I expected to observe renewal in 24-day-old rats.

Whether 17-day-old rats would show renewal was the first question of interest in this chapter. Secondly, I asked whether there was a developmental dissociation in the recovery of extinguished fear following a pre-test injection of FG7142. As GABAergic neurotransmission appears to be the general inhibitory mechanism underlying extinction in the adult rat, both P24 and P17 rats may be affected by a pre-test injection of FG7142.

This hypothesis is based on the finding that FG7142 reverses extinction independent of contextual influences (Harris & Westbrook, 1998b). However, if there are fundamental differences in the mechanisms mediating extinction at different stages of development,

FG7142 may recover an extinguished fear response in P24 rats but not in P17 rats. 83

General Methods

Subjects

All experiments used experimentally naive Sprague-Dawley derived rats, bred and housed in the School of Psychology, University of New South Wales, as described in

Chapter 2. Rats were either P16 or P23 at the start of an experiment.

Drug Injections

Injections, of either FG7142 or vehicle, were given subcutaneously (nape of the neck).

Rats were weighed before injection, and all injections were given at a volume of 2ml/kg, in a separate room. FG7142 was dissolved in 0.9% w/v sterile saline with one drop of

Tween 80 added per 5 ml of saline. Saline, with Tween 80 added, was used for vehicle injections. In all experiments reported in this chapter, injections occurred 10 minutes prior to testing. FG7142 was injected at a dose of 10mg/kg in all experiments except

Experiments 5.3 and 6.2 (dose-response experiments).

Apparatus

Two distinctive types of experimental chambers described in Chapter 2 were used to provide different contexts. In Experiments 4, 5.1, 5.2, and 5.3, the use of these two sets of experimental chambers was counterbalanced. In Experiments 6.1, and 6.2, training always occurred in the first type of chamber (Figure 2.1A) and testing always occurred in the second type (Figure 2.1B). 84

Procedures

Training. In Experiments 4, 5.1, 5.2, and 5.3, 16-day-old rats received 6 CS-US pairings and 23-day-old rats received 3 CS-US pairings as described in Chapter 2. In

Experiments 6.1, and 6.2, P16 rats received 2 pairings as described in Chapter 2.

Extinction. Extinction training consisted of 30 CS presentations in the absence of shock as described in Chapter 2.

Test. All rats were tested for baseline (1 min) and CS-elicited fear (2 mins) as described in Chapter 2.

Scoring and Statistics

Each animal was scored for freezing during extinction training and test. Freezing was scored by a time sampling procedure whereby each rat was scored every three seconds as freezing or not freezing, as described in Chapter 2. A percentage score was calculated for each animal to indicate the proportion of total observations scored as freezing. A second scorer unaware of the experimental condition of each rat scored a random sample (40%) of all rats tested in the experiments described in this chapter. The inter- rater reliability was very high (r = .95).

Two rats from the Vehicle-Same group in Experiment 5.1, one rat from the

Vehicle-Different group in Experiment 5.2, one rat from the 1mg/kg group in

Experiment 2.3, and one rat from the Vehicle group in Experiment 6.1 did not meet the baseline criteria after three baseline tests, and so were not tested. There were no other rats re-tested for baseline in this chapter (i.e., all met criterion on the first test). One- way ANOVA revealed significant group differences in baseline freezing levels at test in

Experiment 4, F (3, 30) = 5.29, p < .005. No other significant baseline differences were found in the other experiments (Table 3.1 displays the baseline freezing levels in each 85

experiment described in this chapter). Statistical analyses of performance at test were done with ANCOVA (with baseline scores as the co-variate) because of the baseline difference in Experiment 4. The same results were obtained, however, whether the data were analysed with ANOVA or ANCOVA in all experiments.

Experiment Groups % Freezing

4 P17-Same 5 (± 3) P17-Different 11 (± 5) P24-Same 19 (± 7) P24-Different* 31 (± 6)

5.1 Vehicle-Same 16 (± 4) Vehicle-Different 33 (± 4) FG7142-Same 27 (± 5) FG7142-Different 31 (± 6)

5.2 Vehicle-Same 7 (± 4) Vehicle-Different 6 (± 4) FG7142-Same 8 (± 4) FG7142-Different 9 (± 5)

5.3 Control 8 (± 5) 1mg/kg 9 (± 4) 5mg/kg 4 (± 3) 10mg/kg 5 (± 3)

6.1 Vehicle 4 (± 3) FG7142 5 (± 3)

6.2 Vehicle 6 (± 2) 1mg/kg 2 (± 2) 5mg/kg 3 (± 2) 10mg/kg 2 (± 2)

Table 3.1 Mean (± SEM) levels of baseline freezing at test for all groups across all experiments in Chapter 3. * Indicates a significant difference to the other groups in baseline levels of freezing. 86

Experiment 4. Renewal of extinguished fear in P17 and P24 rats

Experiment 4 investigated renewal of extinguished fear in P17 and P24 rats. Based on the findings of the previous chapter, P24 rats should exhibit renewal of extinguished fear when tested in a context different from where extinction training occurred. The question of primary interest was whether P17 rats also exhibit renewal of extinguished fear, or express extinction (i.e., low levels of freezing) regardless of the test context.

Method

A 2 x 2 factorial design was used: the first factor was Age at extinction (P17 or P24); and the second was Test Context (Same or Different from the extinction context). The rats in the Same groups were in an AAA or ABB condition. That is, half the rats the

Same group were trained, extinguished, and tested in context A, while the other half were trained in context A, and then extinguished and tested in context B (there were no significant differences between the two conditions for both ages, t’s < 1). The Different groups were in an ABA condition (each letter indicates the context of conditioning, extinction, and testing, respectively).

On Day 1, all rats were given pairings of the auditory CS and the shock US. On

Day 2, the learned fear of the CS was extinguished by giving repeated non-reinforced presentations of the CS. On Day 3 the rats were tested in either the Same or Different context to where they had received extinction trials.

Results and Discussion

One rat from the P17-Different group was excluded from the statistical analysis because it was an outlier at test [4.5 standard deviations (STD) away from the group mean]. Rats 87

at both ages exhibited substantial levels of freezing on the first minute (a block of 3 trials) of extinction (Figure 3.1). By the last block of extinction, substantially less freezing was seen in both age groups. A mixed-design ANOVA of these data yielded a significant main effect of Block, F (1, 30) = 60.66, p < .0001, but no main effect of Age,

F (1, 30) = 1.39, p = .248, or Testing Context, F < 1. Further, there were no significant interactions.

Figure 3.1 Mean (± SEM) percentage freezing to the CS during extinction training in Experiment 4. No significant differences were found between any groups.

The mean and SEM levels of freezing on test are shown in Figure 3.2. P24 rats tested in the same context as where extinction was given exhibited low levels of freezing while those tested in a different context exhibited substantially higher levels of 88

freezing. That is, rats this age exhibited renewal. In fact, the performance of P24 rats in the Different group was comparable to that seen on the first block of extinction trials

(i.e., renewal of fear was complete). In contrast, both P17 groups showed very little freezing at test; that is, rats this age did not exhibit renewal. ANCOVA confirmed this description of the data; there was a significant main effect of Age, F (1, 29) = 17.47, p

< .0001, Testing Context, F (1, 29) = 4.89, p < .05, and an Age x Testing Context interaction, F (1, 29) = 10.75, p < .005. Subsequent post-hoc comparisons, with Tukey’s

HSD test, showed that rats in the P24-Different group froze significantly more than all the other groups (p < .0001); no other group differences were detected.

Figure 3.2 Mean (± SEM) percentage freezing by rats in response to the CS during test in Experiment 4. Rats of two ages were tested in either the same or different context to where they were extinguished: P17-Same (n = 8), P17-Different (n = 7), P24-Same (n = 9), and P24- Different (n = 10). * Indicates a significant difference to the other groups. 89

These data show that rats trained at P24 exhibit renewal of an extinguished fear response while rats trained at P17 do not. This result is consistent with the findings of

Chapter 2 that showed P24 rats display reinstatement of extinguished fear where P17 rats do not. Interestingly, P24 rats tested in the same context as extinction exhibited slightly elevated levels of freezing compared to their freezing levels at the end of extinction suggestive of short-term spontaneous recovery. This elevation was absent in

P17 rats. 90

Experiments 5.1, 5.2, & 5.3. Renewal and FG7142-induced recovery of extinguished fear in P24 and P17 rats

The results of Experiment 4 suggest that there is a developmental difference in the contextual modulation of extinction. Specifically, P24 rats exhibit a pattern of performance similar to that observed in adults (i.e., renewal), while P17 rats exhibit context-independent extinction (i.e., no renewal). The failure to see renewal in P17 rats in Experiment 4 and the failure to see reinstatement in P17 rats in Experiments 1 and 2 from Chapter 2 suggest that extinction processes are fundamentally distinct across development in rats. Alternatively, P17 rats may simply fail to utilise the contextual information to modulate the expression of extinction memory, due to hippocampus immaturity. Therefore, in Experiments 5.1, 5.2, and 5.3 I differentiated these two accounts by asking whether the developmental difference observed in extinction was restricted to the context-specificity of extinction (i.e., renewal and reinstatement). More specifically, FG7142-induced recovery of extinguished fear was examined in the developing rat because it has been previously observed that adult rats given a pre-test injection of FG7142 exhibit a recovery from extinction regardless of the context of test

(Harris & Westbrook, 1998b).

In Experiments 5.1 and 5.2, the effects of a pre-test systemic injection of

FG7142 in P24 and P17 rats were examined (respectively). These experiments also aimed to replicate Experiment 4 by testing the rats in either the same or different context to where extinction occurred. In Experiment 5.3 different doses of FG7142 on extinction were tested in P17 rats. Experiment 5.3 also had a non-extinguished control group, which tested whether the initial conditioned fear memory remains intact without spontaneous forgetting over the conditioning-test interval in P17 rats. 91

Experiment 5.1

Method

A 2 x 2 factorial design was used, in which the first factor was Drug (Vehicle or

FG7142) and the second was Testing Context (Same or Different from the extinction context). All rats were P23 on Day 1 of this experiment. The procedures were the same as in Experiment 4, except that rats were given an injection of Vehicle or FG7142 prior to test.

Results and Discussion

Figure 3.3 Mean (± SEM) percentage freezing by P24 rats to the CS during extinction training in Experiment 5.1. No significant differences were found between any groups. 92

All groups exhibited substantial levels of freezing on the first block of extinction, which significantly decreased by the last block (Figure 3.3). A mixed-design ANOVA of these data yielded a significant main effect of Block [F (1, 38) = 154.41, p < .0001], but no main effect of Drug [F (1, 38) = 3.08 p = .087] or Testing Context (F < 1); there were no significant interactions of these factors.

Figure 3.4 Mean (± SEM) freezing in response to the CS at test in Experiment 5.1. P24 rats were tested in either the same or different context to where they were extinguished and were injected with either FG7142 or vehicle before test: Vehicle-Same (n = 10), Vehicle-Different (n = 11), FG7142-Same (n = 10) and FG7142-Different (n = 11). * Indicates a significant difference to the other groups. 93

CS-elicited levels of freezing on test are shown in Figure 3.4. Rats tested in a context different from the extinction context or those injected with FG7142 prior to test exhibited high levels of freezing. Statistical analyses confirmed these descriptions of the results. ANCOVA revealed a significant main effect of Testing Context, F (1, 37) =

5.37, p < .05, and a Testing Context x Drug interaction, F (1, 37) = 4.01, p < .05. The main effect of Drug was not significant, F (1, 37) = 2.65, p = .112. Subsequent post-hoc comparisons, with Tukey’s HSD procedure, showed that rats in the Vehicle-Same group exhibited significantly less freezing than did rats in all the other groups (p < .005), and that the other groups did not differ significantly.

These results replicate the findings reported on P24 rats in Experiment 4, as P24 rats exhibited complete renewal of extinguished fear. Similar to Experiment 4, short-term spontaneous recovery was again observed in these rats. Furthermore, this experiment demonstrates that P24 rats exhibit a return of extinguished fear if given a pre-test injection of FG7142. Finally, these results show that adding these two treatments (i.e., context change and pre-test FG7142) does not lead to further increases in recovered fear in P24 rats.

Experiment 5.2

Method

A 2 x 2 factorial design was used, in which the first factor was Drug (Vehicle or

FG7142) and the second was Testing Context (Same or Different from the extinction context). On Day 1, P16 rats received 6 CS-US pairings as described in Chapter 2. All 94

rats received extinction at P17 and were tested at P18. The procedures were the same as in Experiment 5.1.

Results and Discussion

One rat from the FG7142-Different group was excluded from the analysis because it was a statistical outlier at test (3 STD away from the group mean). All groups exhibited substantial levels of freezing on the first block of extinction, which significantly decreased by the last block (Figure 3.5). A mixed-design ANOVA of these data yielded a significant main effect of Block [F (1, 34) = 66.97, p < .0001], but no main effect of

Drug (F < 1) or Testing Context [F (1, 34) = 1.28, p = .265], and there were no significant interactions of these factors.

Figure 3.5 Mean (± SEM) percentage CS-elicited freezing by P17 rats during extinction training in Experiment 5.2. No significant differences were found between any groups. 95

At test, P17 rats in all groups exhibited low levels of CS-elicited freezing (i.e., extinction). There was no effect of test context or of drug injection at this age (Figure

3.6). Statistical analyses confirmed these descriptions of the results. ANCOVA showed no effect of Testing Context [F (1, 33) = 1.23, p = .277], Drug (F < 1), or Testing

Context x Drug interaction (F < 1). These results replicate those reported for this age group in Experiment 4, as P17 rats did not show renewal. In addition, these results show that rats this age do not exhibit a return of fear when given a pre-test injection of

FG7142.

Figure 3.6 Mean (± SEM) percentage freezing in response to the CS during test in Experiment 5.2. P17 rats were tested in either the same or different context to where they were extinguished and were injected with either FG7142 or vehicle before test: Vehicle-Same (n = 9), Vehicle- Different (n = 9), FG7142-Same (n = 9), and FG7142-Different (n = 11). 96

Experiment 5.3

Method

On Day 1, 4 groups of P16 rats received 6 CS-US pairings as described in Chapter 2.

Three groups then received extinction training on Day 2, whilst the remaining group received context exposure only to serve as a non-extinguished Control. On Day 3, all groups were tested in the same context as Day 2. Ten minutes before testing, rats in the non-extinguished Control group received an injection of Vehicle and the rats in the three extinguished groups received an injection of FG7142 at one of the following doses: 1, 5, or 10mg/kg.

Results and Discussion

Figure 3.7 Mean (± SEM) percentage CS-elicited freezing during extinction training in Experiment 5.3. No significant differences were found between any groups. 97

All extinguished groups exhibited substantial levels of freezing on the first block of extinction that significantly decreased by the last block (Figure 3.7). A mixed-design

ANOVA of the extinction data yielded a significant main effect of Block, F (1, 20) =

56.60, p < .0001, but no effect of Group, F < 1, and no significant interaction of these factors.

Figure 3.8. Mean (± SEM) freezing by P17 rats in response to the CS during test in Experiment 2.3. Groups of rats that had been extinguished were injected with various doses of FG7142 prior to test while rats in the non-extinguished control group were injected with vehicle: Control- Vehicle (n = 8), 1mg/kg-FG7142 (n = 7), 5mg/kg-FG7142 (n = 8), and 10mg/kg-FG7142 (n = 8). * Indicates a significant difference to the other groups.

Rats in the non-extinguished Control group displayed substantial levels of freezing to the CS at test, showing that rats conditioned at P16 retain the CS-US 98

association over the 2 day retention interval (Figure 3.8). In contrast to the performance of the Control group, rats in the three extinguished groups exhibited much lower levels of freezing, regardless of the dose of FG7142 injected prior to test. Subsequent statistical analysis confirmed these descriptions of the data. ANCOVA of CS-elicited freezing yielded a significant main effect of group, F (3, 26) = 7.49, p < .001, and post- hoc comparisons, with Tukey’s HSD procedure, showed that the Control group exhibited significantly more freezing than did the other three groups (p < .005), which did not differ from one another.

Taken together, the results of Experiments 5.1 and 5.2 replicate those of Experiment 4, as it was observed that P17 rats do not show renewal while P24 rats do. Further, the results of Experiments 5.1, 5.2, and 5.3 demonstrate for the first time that pre-test injection of FG7142 at a range of doses has no effects on extinction in rats extinguished at P17, whereas it restores the extinguished fear response in rats extinguished at P24.

From this latter finding, it would appear that developmental differences in extinction are not restricted to the contextual modulation of extinction. These results suggest that extinction in P17 rats is mediated by different neural mechanisms compared to extinction in P24 and adult rats. 99

Experiments 6.1 & 6.2. Recovery of forgotten fear by pre-test injection of FG7142

Experiments 4, 5.1, and 5.2 demonstrated that P17 rats display context-independent extinction whereas P24 rats show context-dependent extinction. Furthermore, pre-test injections of FG7142 recovered extinguished fear memory in P24 rats, which is consistent with previous findings with adult rats (Harris & Westbrook, 1998b). From this it would appear that P24 rats behave just as adult rats do in regards to extinction – pre-test FG7142 injection and a change in context allows for retrieval of the original

CS-US association. On the other hand, FG7142 had no effects on extinction in P17 rats.

Taken together, P17 rats appear to be different, not showing renewal or the FG7142 effect, even across a range of doses of FG7142 (Experiment 5.3).

There are at least two potential explanations for the failure to observe a return of fear in P17 rats given a pre-test injection of FG7142. The first is that the underlying mechanisms involved in extinction are different in P17 rats compared to older rats.

Further, this difference is not restricted to the contextual modulation of extinction but may be more fundamental, as evidenced by the failure of FG7142 to have any effects.

The second is that infant rats may not be affected by FG7142 in any situation.

Experiments 6.1 and 6.2 tested this latter possibility by examining whether pre-test injection of FG7142 alleviates spontaneous forgetting in rats that are conditioned at P16 and tested at P18 (these ages match those for conditioning and test in Experiments 5.2

& 5.3). Recently, it has been demonstrated that pre-test administration of FG7142 attenuates infantile amnesia (Kim, McNally, & Richardson, 2006). Specifically, Kim et al. (2006) paired an auditory CS with footshock US in P18 rats. In their first experiment, these rats displayed robust freezing to the CS 24 hrs after the conditioning episode but exhibited considerable forgetting over a 10-day retention interval. Subsequent 100

experiments then showed that rats given a pre-test injection of FG7142 exhibited significantly more freezing to the CS compared to rats given vehicle at the 10-day retention interval. That is, FG7142 restored the forgotten fear in rats conditioned at P18 and tested at P28. As observed in Harris and Westbrook (1998b), FG7142 did not elicit freezing in rats that had not previously been shocked, indicating that FG7142 itself does not induce freezing. However, rats in Kim et al. (2006) were given FG7142 at 28 days of age. Therefore, it is possible that the rats tested with FG7142 at P18 in Experiments

5.2 and 5.3 simply were not affected by the drug. Therefore, in Experiments 6.1 and 6.2 rats were given an abbreviated training session at P16 and then tested at P18. In Chapter

2, it was reported that these training parameters produce CS-elicited freezing that is rapidly reduced over 2 days in rats this age (Experiment 3.1). Therefore, Experiment 6.1 examined whether an injection of FG7142 prior to test would alleviate this forgetting. In

Experiment 6.2, the dose-response function of this alleviation of spontaneous forgetting was characterised.

Experiment 6.1

Method

Two groups of P16 rats received 2 pairings of the CS and the US as described in

Chapter 2. At P18, rats received an injection of either vehicle or FG7142, and then were tested 10 mins later. 101

Results and Discussion

Figure 3.9 (A) Mean (± SEM) CS-elicited freezing during test in Experiment 6.1. Two days after conditioning, rats were injected with vehicle or FG7142 before test: Vehicle (n = 7) and FG7142 (n = 8). * Indicates a significant difference to the Vehicle group. (B) The basic forgetting function obtained in Experiment 3.1 (Chapter 2).

The mean and SEM levels of freezing on test are shown in Figure 3.9. ANCOVA revealed that group FG7142 froze significantly more after the 48 hr retention interval than did group Vehicle, F (1, 12) = 11.38, p < .01. This result shows that a pre-test injection of FG7142 significantly reduced spontaneous forgetting in rats tested at 18 102

days of age (i.e., the same age where pre-test FG7142 injections were found to be ineffective in reversing extinction in Experiments 5.2 and 5.3). As there was no significant group difference in baseline levels of freezing, t (13) = 0.37, p = .72 (see

Table 3.1 for more detail), the effect of pre-test FG7142 injection cannot be attributed to the drug simply causing freezing behaviour. Further evidence that pre-test FG7142 does not produce sensitisation of freezing nor state-dependent retrieval is provided by the latent inhibition experiment in Kim et al. (2006).

Experiment 6.2

Method

Four groups of P16 rats were conditioned and tested as in Experiment 6.1. Prior to test, they received a pre-test injection of either Vehicle or FG7142, at a dose of 1, 5, or

10mg/kg.

Results and Discussion

One rat from group Vehicle was excluded from the statistical analysis because it was a statistical outlier at test (4.5 STD away from the group mean). The mean and SEM levels of freezing on test are shown in Figure 3.10. Pre-test administration of FG7142 dose-dependently alleviated spontaneous forgetting. ANCOVA yielded a significant linear trend, F (3, 28) = 8.44, p < .01. Subsequent post-hoc comparisons, with Tukey’s

HSD procedure, showed that rats given 10mg/kg FG7142 exhibited significantly higher levels of freezing than did rats given a pre-test injection of Vehicle (p < .05); there were no other significant group differences. 103

Figure 3.10 Mean (± SEM) CS-elicited freezing expressed by rats during test in Experiment 6.2. Pre-test injection of FG7142 dose–dependently alleviated spontaneous forgetting in P16 rats: Vehicle (n = 7), 1mg/kg-FG7142 (n = 8), 5mg/kg-FG7142 (n = 8), and 10mg/kg-FG7142 (n = 9). * Indicates a significant difference to the Vehicle group. 104

Discussion

The experiments reported in the present chapter examined the effects of context manipulation and a pre-test injection of FG7142 on extinction in the developing rat.

Experiment 4 showed that P24 rats displayed a return of extinguished fear when tested in a different context to where they had received extinction training whereas P17 rats expressed extinction regardless of the test context. In other words, the older rats exhibited renewal of an extinguished fear response whereas P17 rats did not. This finding is consistent with the developmental difference found in reinstatement of extinguished fear in Chapter 2. Experiments 5.1 and 5.2 replicated Experiment 4, and also showed that pre-test injections of FG7142 led to the recovery of extinguished fear in P24 rats but not in P17 rats, even across a range of doses (Experiment 5.3). The failure to observe any FG7142 effect in P17 rats in extinction cannot be attributed to a general lack of responsiveness to this drug in these rats, as FG7142 was found to be effective in alleviating spontaneous forgetting in rats this age (Experiments 6.1 & 6.2).

Taken together, the results presented in this chapter show that extinction in P24 rats is susceptible to context manipulations and pre-test alterations in GABAergic neurotransmission while extinction in P17 rats are not influenced by either. These findings suggest that the processes mediating extinction may be distinct across development.

The present investigation so far show that three different treatments that attenuate extinction in adult rats (change in context, post-extinction US reminder, and pre-test FG7142) also do so in P24 rats, whereas all three fail to attenuate extinction in

P17 rats. This is extremely suggestive evidence that different processes underlie extinction of learned fear in P17 rats compared to older rats. The developmental dissociations reported in this thesis have important theoretical and practical implications 105

because the loss in conditioned responding in P17 rats may be indicative of ‘unlearning’ of the original CS-US association. The early unlearning accounts of fear extinction (e.g.,

Rescorla & Wagner, 1972) were rejected based on the findings of spontaneous recovery, reinstatement, and renewal. Additionally, FG7142-induced recovery of an extinguished fear response in adult rats was taken as strong evidence for extinction being new learning that inhibits the original fear response (Harris & Westbrook, 1998b). The failure to observe FG7142 effects, renewal, or reinstatement, in P17 rats in the present study raises unlearning as a potential underlying mechanism of extinction in rats early in development. No one has yet systematically examined spontaneous recovery in the developing rat, and finding a developmental dissociation in such phenomenon would further support the idea of unlearning as the possible candidate for extinction in P17 rats.

In fact, many recent reviews on extinction now suggest both ‘new learning’ and

‘unlearning’ as mechanisms for extinction in the adult rat (Delamater, 2004; Barad,

Gean, & Lutz, 2006; Lattal, Radulovic, & Lukowiak, 2006; Myers & Davis, 2007). It may be the case that when extinction occurs early in development, the balance between unlearning and new learning mechanism of extinction is simply shifted, in that extinction relies more on unlearning rather than new learning.

The present findings are also interesting in light of the current neurobiological models of extinction. As noted previously, these models maintain that fear extinction, at least in adult rats, involves a circuit including the hippocampus, amygdala, and mPFC

(Hobin et. al., 2003; Quirk et. al., 2000; Sotres-Bayon et. al., 2006). More specifically, the current neurobiological models of extinction state that the hippocampus inhibits or excites the mPFC and thereby modulates the expression of learned fear depending on the test context. For example, in renewal, a shift in context from where extinction occurred may activate the inhibitory projections from the hippocampus to the mPFC, 106

preventing the inhibitory input normally applied on the amygdala following extinction.

Alternatively, after extinction training, the hippocampus may excite the mPFC only in the extinction context. As a result, renewal is observed when the rat is tested in a different context to extinction because the hippocampus no longer activates the mPFC, which leads to disinhibition of the amygdala. In either case, the reliable occurrence of renewal (and reinstatement; Chapter 2) in P24 rats in the present thesis strongly indicate that the hippocampus-mPFC-amygdala network is able to function in an adult-like manner by the age of 24 days in rats. Furthermore, using FG7142, I have demonstrated that extinction in P24 rats is mediated by GABA, which is the inhibitory neurotransmitter in the neural models of extinction. Specifically, FG7142 may have diminished GABAergic inhibitory activity directly in the amygdala or through mPFC projections to the amygdala and thereby permitted the retrieval of the original fear memory in P24 rats. Overall, it appears that extinction in P24 rats is essentially adult- like, as it is context-dependent and GABA-mediated. However, the exact age at which the hippocampus-mPFC-amygdala network begins to function in an adult-like manner remains to be determined.

In contrast to P24 rats, neither a change in context nor a pre-test injection of

FG7142 produced a recovery of the extinguished freezing response in P17 rats

(Experiments 4, 5.2, & 5.3). These findings suggest that the neural circuit mediating extinction in P17 rats may be functionally and/or structurally different from that involved in older rats. If the same hippocampus-mPFC-amygdala circuitry is involved in extinction in these rats, then the present series of experiments suggests that the hippocampus may not exert control over the mPFC at this age, as evidenced by the failure to observe renewal in P17 rats. This failure could be due to the delayed maturation of the hippocampus (Baudry et al., 1981; Jacobson, 1963; Schlessinger et al., 107

1975; Wilson, 1984), or to the delayed development of projections between the hippocampus and mPFC.

However, according to current models of extinction in the adult rat, the absence of the information on the extinction context should render the extinction memory inaccessible (i.e., disrupt extinction; Delamater, 2004). This was clearly not the case in

P17 rats in the experiments described in this chapter. Also, the failure to observe a recovery from extinction in P17 rats following pre-test injections of FG7142 would not seem to be amenable to an explanation that focuses on the delayed maturation of the hippocampus. It was shown that reducing GABAergic inhibition by pre-test injection of

FG7142 failed to recover an extinguished fear response in P17 rats (Experiments 5.2 &

5.3). Such a finding suggests that GABAergic inhibition may not be involved in the processes mediating extinction in rats early in development. Additionally, the lack of

FG7142 effect found in the present study questions the involvement of the mPFC in extinction in P17 rats because the current neural models of extinction maintain that

GABAergic neurotransmission in the amygdala is mediated by the mPFC (see Hobin et. al., 2003, for review). In fact, no one has yet examined the role of the mPFC or the amygdala in extinction of conditioned fear in the developing rat. Therefore, a possible alternative account for the present results is that a different neural circuitry mediates extinction in P17 rats. This idea that a different neural circuitry may mediate extinction in younger compared to older rats is indirectly supported by previous research. For example, the amygdala does not participate in odour-shock associative learning early in development (i.e., prior to P12, Sullivan et al., 2000). Hence, it is possible that the role of the amygdala in extinction emerges even later in development. In any case, a closer investigation on the neural mechanisms of extinction in the developing rat appears necessary. 108

Chapter Four.

AMYGDALA INVOLVEMENT IN EXTINCTION AND RE-EXTINCTION

IN THE DEVELOPING RAT

Chapters 2 and 3 provide strong evidence for a developmental dissociation in mechanisms underlying fear extinction: when extinction occurs at 17 days of age rats fail to show renewal or reinstatement of extinguished fear; whereas when extinction occurs at 24 days of age rats do show these phenomena. Also, reducing GABAergic inhibitory activity by pre-test injection of the GABAA receptor inverse agonist FG7142 reverses extinction if extinction is given at P24 but not if given at P17.

In all these experiments, the levels of CS-elicited freezing expressed at the beginning of extinction and the rate of extinction did not differ between P17 and P24 rats. Hence, the developmental dissociations in extinction observed in Chapters 2 and 3 are not due to potential quantitative differences in fear learning or the rate of extinction in these ages. Rather, it appears that qualitatively different mechanisms underlie extinction across development.

To date, no-one has explicitly examined whether the neural structures implicated in extinction in adult rats are also critical for extinction of conditioned fear in the developing rat. In adult rats, the amygdala is an unequivocally important neural structure for extinction of conditioned fear (see Figure 1.6). Specifically, studies have shown that blocking various neurotransmitter systems in the amygdala during extinction training disrupts long-term extinction in the adult rat (see Barad et al., 2006; Quirk &

Mueller, 2007, for reviews). For example, intra-BLA infusions of the NMDA receptor antagonist AP5 prior to extinction training disrupt extinction of FPS to a light CS (Falls et al., 1992). Conversely, intra-BLA infusions of the NMDA partial agonist DCS facilitate extinction (Ledgerwood et al., 2003; Walker et al., 2002). These studies show 109

that the NMDA neurotransmitter system in the amygdala is critically involved in for extinction of learned fear in adult rats.

Another neurotransmitter system in the amygdala that has been implicated in extinction in adult rats is GABA. Chhatwal et al. (2005) first reported that mRNA for the GABAA receptor clustering protein gephyrin was significantly increased in the BLA

2 hrs post-extinction, as was the surface expression of GABAA receptors. Hence, alterations in inhibitory GABAergic synapses appear to occur in the BLA during extinction. Consistent with this idea, post-extinction microinfusion of the GABAA receptor agonist muscimol (at 1/10th of the usual dose used for inactivation) into the

BLA has been shown to facilitate extinction (Akirav et al., 2006). These results, combined with studies on the role of NMDA in extinction, indicate that extinction relies on at least two types of neurotransmitter systems in the amygdala in the adult rat.

Therefore, the present chapter first aimed to investigate the role of the amygdala in extinction of conditioned fear in the developing rat. Instead of microinfusing a drug that targets a specific neurotransmitter system, however, the local anaesthetic bupivacaine was used to temporarily inactivate the amygdala during extinction training.

Bupivacaine, like lidocaine, is an amide-linked local anaesthetic that reduces sodium conduction by blocking voltage-gated sodium channels. However, it is more lipophilic than lidocaine, and hence, has a longer duration of action (Catterall & Mackie, 1996).

Bupivacaine, and other sodium-channel modulators such as tetrodotoxin, has previously been used to reversibly inactivate brain areas such as the nucleus accumbens, the amygdala, and the vmPFC during fear-conditioning tasks in rats (e.g., Calandreau,

Desmedt, Decorte, & Jaffard, 2005; Corcoran & Quirk, 2007; Haralambous &

Westbrook, 1999; Weber & Richardson, 2004). 110

I chose not to manipulate a specific neurotransmitter system in the amygdala because it has been shown that systemic injections of GABAergic and NMDA antagonists have no effects on extinction in P17 rats. Specifically, it was shown in

Chapter 3 that a pre-test injection of FG7142 did not recover an extinguished freezing response in P17 rats (Experiments 5.2 & 5.3). Further, it has recently been demonstrated that a pre-extinction injection of the NMDA antagonist MK-801 had no effect on long- term extinction in P17 rats, whereas long-term extinction in P24 rats was impaired

(Langton, Kim, Nicholas, & Richardson, 2007). As there is no evidence so far indicating that GABA or NMDA is involved in extinction at 17 days of age, it appeared unwise to target either of these neurotransmitter systems in the amygdala. Hence, the local anaesthetic bupivacaine was used to bilaterally inactivate the amygdala before extinction training in P17 and P24 rats in the current series of experiments. It is also worth noting that the infusion needles were aimed between the BLA and CeA (i.e., the drug was not aimed at a particular nucleus in the amygdale) because virtually nothing is known about the role of the amygdala in extinction of conditioned fear in the developing rat.

111

General Method

Subjects

All experiments used experimentally naive Sprague-Dawley derived rats, bred and housed in the School of Psychology, University of New South Wales, as described in

Chapter 2. Rats were either P16 or P23 at the start of an experiment.

Surgery

Rats were anaesthetised with isoflurane (Laser Animal Health, Queensland, Australia) mixed with oxygen and nitrous oxide. The gas was delivered via a single tube attached to a nosepiece (Stoelting, Wood Dale, IL) that fit onto a stereotaxic frame (David Kopf

Instruments, Tujunga, CA). Twenty-six gauge guide cannulas (Bioscientific, Sydney,

Australia) were bilaterally implanted between the central and the basolateral nucleus of the amygdala. Slightly different stereotaxic coordinates were used in each experiment because rats were operated on at different ages in the experiments reported in this chapter. The coordinates used for Experiments 7.1, 8.1, and 9 (surgery at P22, P25, &

P25, respectively) were 2.1 mm posterior, ± 4.3 mm lateral, 7.4 mm ventral to bregma.

For Experiment 7.2 the coordinates were 1.9 mm posterior, ± 4.1 mm lateral, 7.0 mm ventral to bregma (surgery at P15). Finally, for Experiment 8.2 the coordinates were 2.0 mm posterior, ± 4.3 mm lateral, 7.4 mm ventral to bregma (surgery at P18). Pilot experiments have indicated that these coordinates produce similar cannula placements across different ages. The guide cannulas were fixed directly onto the skull with cyanoacrylate glue and were surrounded with dental cement. Cannula stylets were inserted to prevent clogging of the guide cannulas. Following surgery, rats were injected 112

subcutaneously with 0.03 ml of a 100 mg/ml solution of cephazolin and of a 300 mg/ml solution of benacillin to prevent infection.

While the rats were undergoing surgery, the dam was removed from the home cage. After surgery, the pups were placed back into the home cage and given approximately 30 min for temperature and skin odours to return to normal before the dam was returned (see Hofer, 1991, for more detail). It has previously been shown that rats these ages recover from surgery well within 24 hrs (e.g., Weber & Richardson,

2001).

Rats were decapitated less than 1 hr after the final test, and tissue sections were collected and stained for Nissl substance for verification of cannula placement (Figure

4.1 for Experiments 7.1 & 7.2; Figure 4.6 for Experiments 8.1 & 8.2; and Figure 4.13 for Experiment 9) (Paxinos & Watson, 1998). A total of 130 rats were used for the experiments reported in this chapter. Data from 24 rats were excluded from analysis either for sustaining amygdala damage or misplacement of cannula/s.

Drug Infusion

For infusion of bupivacaine hydrochloride (5% wt/vol; Delta West, Perth, Australia) or saline, the cannula stylets were removed and 33-gauge infusion cannulas, extending 1 mm below the tip of the guide cannula, were inserted. The infusion cannulas were connected to microsyringes driven by a microinfusion pump (Syringe pump SP101IZ,

World Precision Instruments) via polyethylene 50 tubing. Rats were placed in individual plastic buckets and solution (0.5 l per hemisphere) was delivered over 2 min.

The injection cannulas were left in position for an additional 2 min before withdrawal to minimise dragging of the solution along the injection track.

113

Apparatus

A single set of identical chambers were used for all sessions in all experiments described in this chapter. As described in Chapter 2 (Figure 2.1B), the identical chambers were rectangular and wholly constructed of Plexiglas, with the exception of the grid floor.

Scoring and Statistics

Each animal was scored for freezing during extinction training and test. Freezing was scored by a time sampling procedure whereby each rat was scored every three seconds as freezing or not freezing, as described in Chapter 2. A percentage score was calculated for each animal to indicate the proportion of total observations scored as freezing. A second scorer unaware of the experimental condition of each rat scored a random sample (25%) of all rats tested in this chapter. The inter-rater reliability was very high (r

= .93).

One rat from the Saline group in Experiment 7.2, one rat from the Bupivacaine group in Experiment 8.1, and one rat from each of the P17-Saline, P24-Saline, and P24-

Bupivacaine groups in Experiment 9 did not meet the baseline criteria after three baseline tests, and so were not tested. There were no other rats re-tested for baseline in this chapter (i.e., all met criterion on the first test). No significant baseline differences were detected at test in any experiment (Table 4.1 displays the baseline freezing levels in each experiment described in this chapter). However, statistical analyses of performance at test were done with ANCOVA3 (with baseline scores as the co-variate)

3 In the publication describing these experiments - “Kim, J. H. & Richardson, R. (2008). The effect of temporary amygdala inactivation on extinction and re-extinction of fear in the developing rat: Unlearning as a potential mechanism for extinction early in development. Journal of Neuroscience. 28, 1282-1290.” - ANOVA F-values, not ones of ANCOVA, were reported 114

to be consistent with the analyses reported in Chapters 2 and 3. The same results were obtained, however, whether the data were analysed with ANOVA or ANCOVA in all experiments in this chapter.

Experiment Groups % Freezing

7.1 Saline 15 (± 5) Bupivacaine 28 (± 7)

7.2 Saline 6 (± 4) Bupivacaine 19 (± 6)

8.1 Control 36 (± 6) Saline 24 (± 6) Bupivacaine 24 (± 9)

8.2 Control 19 (± 11) Saline 19 (± 5) Bupivacaine 30 (± 5)

9 17-Saline 32 (± 7) 17-Bupivacaine 32 (± 7) 24-Saline 30 (± 7) 24-Bupivacaine 30 (± 8)

Table 4.1 Mean (± SEM) levels of baseline freezing at test for all groups across all experiments described in Chapter 4. No significant differences were found between any groups in any experiment.

115

Experiments 7.1 & 7.2. Temporary inactivation of the amygdala during extinction in the developing rat

Research on extinction using adult rats has consistently found that fear extinction is dependent on the amygdala (e.g., Falls et al., 1992). Therefore, Experiments 7.1 and 7.2 examined whether the amygdala is also involved in extinction in the developing rat by using a temporary inactivation technique. Rats at 23 and 16 days of age (in Experiments

7.1 and 7.2, respectively) were first trained to fear an auditory CS by pairing it with a shock US. Twenty-four hrs later, rats at each age received intra-amygdala infusions of either saline or bupivacaine prior to extinction training. They were then tested for extinction retention 24 hrs later, drug free.

Experiment 7.1

Method

Procedure

All rats received cannulation surgery one day before training.

Training. For training on Day 1, P23 rats received 3 pairings of the CS and US as described in Chapter 2.

Extinction. On Day 2, rats received extinction training consisting of 30 CS presentations in the absence of shock as described in Chapter 2. Six minutes before extinction, rats received bilateral amygdala infusion of either Saline or Bupivacaine.

Test. On Day 3, all rats were tested for baseline (1 min) and CS-elicited fear (2 mins) as described in Chapter 2.

116

Figure 4.1 Histological reconstruction of coronal sections showing cannula placements in the amygdala for Experiments 7.1 () and 7.2 ("). Although surgery was conducted at different ages in each experiment, cannula placements were comparable across experiments. 117

Results and Discussion

Group Saline exhibited substantial levels of freezing to the CS in the first block of extinction that decreased substantially by the last block, whereas group Bupivacaine exhibited very low levels of freezing throughout extinction, indicating that the amygdala was inactivated in this group (Figure 4.2). A mixed-design ANOVA of the extinction data yielded significant main effects of Block [F (1, 16) = 21.32, p < .0001] and Group [F (1, 16) = 37.51, p < .0001], as well as a Block x Group interaction [F (1,

16) = 9.10, p < .01].

Figure 4.2 Mean (± SEM) percentage freezing to the CS by P24 rats during extinction training in Experiment 7.1. Rats that were microinfused with saline exhibited significant levels of freezing in the first block that decreased substantially by the last block. Rats that were microinfused with bupivacaine showed negligible amount of freezing throughout extinction training.

118

The mean (± SEM) levels of CS-elicited freezing at test are shown in Figure 4.3.

Rats that received a pre-extinction infusion of bupivacaine exhibited substantially higher levels of freezing at test (drug-free) compared to rats that had received saline.

ANCOVA yielded a significant effect of Drug [F (1, 15) = 15.11, p < .001]. This result shows that temporary inactivation of the amygdala during extinction training disrupts long-term extinction in 24-day-old rats.

Figure 4.3 Mean (± SEM) CS-elicited freezing by rats during test in Experiment 7.1. At test, rats that received pre-extinction infusion of bupivacaine exhibited substantially higher levels of freezing (drug-free) compared to rats that had been infused with saline. Saline (n = 9), and Bupivacaine (n = 9). * Indicates a significant difference to the Saline group.

119

Experiment 7.2

Method

Procedure

All rats received cannulation surgery one day before training.

Training. For training on Day 1, P16 rats received 6 pairings of the CS and US as described in Chapter 2.

Test. On Day 3, all rats were tested for baseline (1 min) and CS-elicited fear (2 mins) as described in Chapter 2.

Results and Discussion

Figure 4.4 Mean (± SEM) percentage freezing to the CS by P17 rats during extinction training in Experiment 7.2. 120

As in Experiment 7.1, group Saline exhibited a substantial level of freezing in the first block of extinction that decreased substantially by the last minute, whereas group

Bupivacaine exhibited very low levels of freezing throughout extinction (Figure 4.4). A mixed-design ANOVA of the extinction data yielded significant main effects of Block

[F (1, 11) = 28.24, p < .0001] and Group [F (1, 11) = 35.36, p < .0001], as well as a

Block x Group interaction [F (1, 11) = 11.89, p < .005].

Figure 4.5 Mean (± SEM) CS-elicited freezing by P17 rats during test in Experiment 7.2. At test, rats that received pre-extinction infusion of bupivacaine exhibited substantially higher levels of freezing (drug-free) compared to rats that had been given of saline. Saline (n = 6), and Bupivacaine (n = 7). * Indicates a significant difference to the other group.

121

The mean and SEM levels of CS-elicited freezing at test are shown in Figure 4.5.

Temporary inactivation of the amygdala during extinction training impaired long-term extinction at test, as shown by higher levels of freezing in rats that had received an infusion of bupivacaine compared to those given saline. ANCOVA revealed a significant effect of Drug [F (1, 10) = 10.94, p < .01].

122

Experiments 8.1 & 8.2. Temporary inactivation of the amygdala during re-extinction in the developing rat

Experiments 7.1 and 7.2 demonstrated that extinction in the developing rat is dependent on the amygdala, as it is with adult rats. Contrary to the experiments described in

Chapters 2 and 3, which report developmental differences in extinction, the results from

Experiments 7.1 and 7.2 suggest that the role of the amygdala in extinction is the same, or at least very similar, across development. To further examine the role of the amygdala in extinction in the developing rat, in the next set of experiments I examined the effect of temporary inactivation of the amygdala in re-extinction of conditioned fear.

Re-extinction of conditioned fear

In re-extinction, rats are re-conditioned to the previously extinguished CS and then receive re-extinction to the same CS. This preparation has been utilised in previous studies to elucidate the neural mechanisms underlying the initial extinction learning

(e.g., Morgan et al., 2003; M. Weber, R. F. Westbrook, P. Carrive, & R. Richardson, unpublished observations). For example, in Morgan et al. (2003), adult rats were trained to fear an auditory CS by pairing it with shock. Some rats then received electrolytic lesions of the vmPFC. Following recovery from surgery, rats were subjected to extinction trials over multiple days. Contrary to previous findings, the results showed that post-training vmPFC lesions had little effect on extinction rate. However, when rats were re-conditioned to the same CS and then re-extinguished, lesioned rats showed significant resistance to extinction during re-extinction trials. This study conveys how examining re-extinction can reveal aspects of initial extinction learning.

Interestingly, in the adult rat, the amygdala appears to be critical for extinction only when extinction happens for the first time (M. Weber et al., unpublished 123

observations). That is, when a fear-eliciting context in adult rats was extinguished, re- conditioned, and then re-extinguished, rats exhibited low levels of freezing (i.e., good extinction) the next day regardless of whether the amygdala was inactivated with bupivacaine or not at the time of re-extinction. In fact, there is increasing evidence supporting the notion that the amygdala may only be necessary for extinction the first time around. For example, microinfusion of the NMDA receptor antagonist AP5 or

GABA agonist midazolam into the BLA disrupts extinction but not re-extinction in the adult rat (G. Hart, V. Laurent, & R. F. Westbrook, unpublished observations; Laurent,

Marchand, & Westbrook, in press). From these studies, it appears that the amygdala is involved in the initial learning of the CS-no US memory acquired in extinction training, but once that memory has been acquired then the amygdala is no longer needed for subsequent extinction training episodes, at least in the adult rat.

In Experiments 8.1 and 8.2, I examined whether a similar transition from amygdala-dependence to amygdala-independence also occurs in re-extinction in the developing rat. In Experiments 8.1 and 8.2, 23- and 16-day-old rats (respectively) were conditioned, extinguished, re-conditioned to the same CS, and then re-extinguished; some rats had their amygdala inactivated at the time of re-extinction. According to the findings so far reported in this thesis, P24 rats are essentially adult-like (e.g., they show renewal and reinstatement). Therefore, I hypothesised that re-extinction in 24-day-old rats would be amygdala-independent. The primary question in these experiments was whether re-extinction would be amygdala-independent in P17 rats. 124

Figure 4.6 Histological reconstruction of coronal sections showing cannula placements in the amygdala for Experiments 8.1 () and 8.2 ("). Although surgery was conducted at different ages in each experiment, cannula placements were comparable across experiments. 125

Experiment 8.1

Methods

Three groups of P23 rats were fear conditioned (Day 1) and extinguished (Day 2) as described in Experiment 7.1. On Day 3, all rats had cannulas surgically implanted. For re-training on Day 4, rats received 2 CS-US pairings. On Day 5, two groups received re- extinction training (30 non-reinforced presentations of the CS; i.e., the same as initial extinction training). Six mins before re-extinction, rats in one group received bilateral amygdala infusions of Saline while rats in the other group were infused with

Bupivacaine. Rats in the non-re-extinguished Control group received infusions (half with saline and half with bupivacaine) and then were exposed to the experimental chamber for an equal amount of time as the other groups, but without any CS presentations. On Day 6, all rats were tested for baseline and CS-elicited freezing as described in previous experiments.

Results and Discussion

For the initial extinction session (drug-free) at 24 days of age, all groups exhibited substantial levels of freezing in the first minute that decreased substantially by the last minute (Figure 4.7). A mixed-design ANOVA of these data yielded a significant main effect of Block [F (1, 22) = 81.53, p < .0001], but no effect of Group [F (2, 22) = 1.25, p = .31] or Block x Group interaction (F < 1).

126

Figure 4.7 Mean (± SEM) percentage freezing to the CS by P24 rats during the initial extinction training (drug-free) in Experiment 8.1. There were no significant group differences.

For re-extinction, group Saline exhibited substantial levels of freezing in the first minute of extinction that decreased substantially by the last minute whereas group

Bupivacaine exhibited very low levels of freezing throughout (Figure 4.8). A mixed- design ANOVA of these data yielded significant main effects of Block [F (1, 15) =

19.85, p < .0001] and Group [F (1, 15) = 33.31, p < .0001], as well as a Block x Group interaction [F (1, 15) = 9.28, p < .01]. 127

Figure 4.8 Mean (± SEM) percentage freezing to the CS by P24 rats during re-extinction in Experiment 8.1. Group Saline exhibited good extinction, whereas group Bupivacaine showed negligible amount of freezing throughout re-extinction training.

The mean (± SEM) levels of CS-elicited freezing at test are shown in Figure 4.9.

Temporary inactivation of the amygdala at the time of re-extinction had no effects on subsequent performance at test, as shown by comparable, low levels of freezing in groups Saline and Bupivacaine and the high levels of freezing in the non-re- extinguished Control group. ANCOVA of the test data revealed a significant effect of

Group [F (2, 21) = 4.01, p < .05]. Subsequent post-hoc comparisons, with Tukey’s HSD test, showed that group Control froze significantly more than the other two groups (p

< .05), which did not differ. As found previously in the adult rat (e.g., Weber et al., unpublished observations), this result indicates that the amygdala is not involved in extinction when extinction is occurring for the second time in P24 rats. 128

Figure 4.9 Mean (± SEM) percentage freezing to the CS by P24 rats during test in Experiment 8.1. Temporary inactivation of the amygdala at re-extinction had no effects on re-extinction, as shown by low levels of freezing in group Saline and Bupivacaine compared to the non-re- extinguished Control group. Control (n = 8), Saline (n = 10), and Bupivacaine (n = 7). * Indicates a significant difference to the other groups.

Experiment 8.2

Method

On Day 1, 3 groups of P16 rats received 6 CS-US pairings. The rest of the procedures were identical to Experiment 8.1.

Results and Discussion

For the initial extinction session (drug-free) at P17, all groups exhibited substantial levels of freezing in the first minute of extinction that decreased substantially by the last minute (Figure 4.10). A mixed-design ANOVA of these data yielded a significant main 129

effect of Block [F (2, 18) = 82.59, p < .0001] but no effect of Group (F < 1) or Block x

Group interaction (F < 1).

Figure 4.10 Mean (± SEM) percentage freezing to the CS by P17 rats during the initial extinction training (drug-free) in Experiment 8.2. There were no significant group differences.

For re-extinction, group Saline exhibited substantial levels of freezing in the first minute of re-extinction that decreased substantially by the last minute, whereas group

Bupivacaine exhibited very low levels of freezing throughout (Figure 4.11). A mixed- design ANOVA of these data yielded a significant main effects of Block [F (1, 14) =

8.00, p < .01] and Group [F (1, 14) = 20.49, p < .0001], but a non-significant Block x

Group interaction [F (1, 14) = 1.92, p = .19].

130

Figure 4.11 Mean (± SEM) percentage freezing to the CS by P17 rats during re-extinction in Experiment 8.2. Group Saline exhibited good extinction, whereas group Bupivacaine showed negligible amount of freezing throughout re-extinction training.

The mean and SEM levels of CS-elicited freezing at test are shown in Figure

4.12. Temporary inactivation of the amygdala disrupted re-extinction in P17 rats, as shown by comparable, high levels of freezing in groups Bupivacaine and Control and low levels of freezing in group Saline. ANCOVA of the test data revealed a significant effect of Group [F (2, 17) = 10.48, p < .001]. Subsequent post-hoc comparisons, with

Tukey’s HSD test, showed that group Saline froze significantly less than either of the other two groups (p < .005), which did not differ. This result indicates that re-extinction is amygdala-dependant in P17 rats.

131

Figure 4.12 Mean (± SEM) percentage freezing to the CS by P17 rats during test in Experiment 8.2. Temporary inactivation of the amygdala at re-extinction disrupted re-extinction, as shown by high levels of freezing in group Bupivacaine compared to the Saline group. Control (n = 5), Saline (n = 8), and Bupivacaine (n = 8). * Indicates a significant difference to the other groups.

Taken together, the results of Experiments 8.1 and 8.2 reveal another developmental dissociation in extinction of conditioned fear. Specifically, re-extinction was found to be amygdala-independent in P24 rats, which replicates previous findings in the adult rat.

Interestingly, re-extinction was still amygdala-dependent in P17 rats.

132

Experiment 9. The effect of age at initial extinction on temporary inactivation of the amygdala during re-extinction in the developing rat

Previous findings show that disrupting amygdala function during re-extinction training has no effects on re-extinction in the adult rat. Experiment 8.1 replicated this finding in

24-day-old rats. However, Experiment 8.2 demonstrated that temporarily inactivating the amygdala at re-extinction disrupted re-extinction in 17-day-old rats. These results show another developmental dissociation in extinction of conditioned fear, and highlight that the neural mechanisms underlying extinction may be fundamentally different across development.

It should be noted, however, that training, extinction, re-training, re-extinction, and test all occurred at different ages in Experiments 8.1 and 8.2 (i.e., it wasn’t merely the age at re-extinction that differed). Therefore, the outcomes these two experiments could be due to age differences at the time of (1) fear conditioning, (2) initial extinction training, or (3) re-extinction. Firstly, it may be that conditioning at 23 days of age is different from conditioning at 16 days of age, such that if conditioning occurs at 16 days of age, extinction and re-extinction always requires the amygdala. A second explanation for the developmental differences observed in Experiments 8.1 and 8.2 is that initial extinction at 17 days of age is fundamentally different from initial extinction at 24 days of age. For example, initial extinction at 24 days of age involves learning a CS-no US association, and subsequent re-extinction simply involves retrieval of the original CS-no

US memory. However, initial extinction at 17 days of age may involve unlearning, and therefore, subsequent re-extinction still requires the amygdala. From this perspective, one would predict that if the initial extinction happens at 17 days of age and re- extinction happens at 27 days of age, re-extinction would still require the amygdala because the initial extinction that happened at 17 days of age involved unlearning. The 133

final possible explanation of the developmental difference reported in Experiments 8.1 and 8.2 is that re-extinction that happens prior to about 24 days of age depends on the amygdala, whereas re-extinction occurring after about 24 days of age does not depend on the amygdala. From this perspective, one would predict that if the initial extinction happens at 17 days of age and re-extinction happens at 27 days of age, re-extinction would be amygdala-independent because re-extinction occurred after 24 days of age.

To test these alternative potential accounts, Experiment 9 again examined temporary inactivation of the amygdala at re-extinction. Unlike Experiments 8.1 and 8.2,

Experiment 9 kept the training, re-training, re-extinction, and test age constant across groups (16, 26, 27, and 28 days of age, respectively). The only thing that varied across groups was the age at the time of the initial extinction training session (either 17 or 24 days of age; see Table 4.2 for design). If the training age is the critical factor determining whether re-extinction is amygdala-dependent or -independent, then rats in both groups should perform similarly; that is, based on the results of Experiment 8.2, bupivacaine should block re-extinction regardless of the initial extinction age. In contrast, if age at the time of the initial extinction is the critical factor, then rats extinguished at 17 days of age should exhibit amygdala-dependent re-extinction while rats extinguished at 24 days should exhibit amygdala-independent extinction. Finally, if age at the time of re-extinction is the critical factor, then re-extinction should be amygdala-independent regardless of the age at initial extinction, based on the results of

Experiment 8.1.

Method

Experiment 9 was a 2 x 2 factorial design: one factor was Drug (Saline or Bupivacaine at re-extinction); the other was initial Extinction Age (P17 or P24). At 16 days of age, 134

four groups of rats were fear conditioned with 9 CS-US pairings. Nine pairings were given because of the longer training-extinction interval in this experiment. Rats were placed in the conditioning chambers, and after a two-minute adaptation period, the CS was presented for 10 s and co-terminated with the shock US (0.6 mA, 1 s). The ITI ranged from 85s to 135s with a mean of 110s. Then two groups (P17-Saline and P17-

Bupivacaine) were given extinction training at 17 days of age, whilst the other two groups (24-Saline and 24-Bupivacaine) received extinction training at 24 days of age.

Extinction training, whenever it occurred, was 30 non-reinforced presentations of the

CS as described in previous experiments in this thesis. All rats received cannula placement surgery at 25 days of age. At 26 days of age, all rats were re-trained with 2

CS-US pairings. The next day, rats received bilateral amygdala infusions of either

Saline or Bupivacaine 6 mins before re-extinction Re-extinction training consisted of 30 non-reinforced presentations of the CS as described in Experiment 8.1. All rats were tested for baseline and CS-elicited freezing one day after re-extinction training.

Groups Train Extinction Re-Train Re-Extinction Test

P17-Saline 16 17 26 27 (saline) 28

P17-Bupivacaine 16 17 26 27 (bupivacaine) 28

P24-Saline 16 24 26 27 (saline) 28

P24-Bupivacaine 16 24 26 27 (bupivacaine) 28

Table 4.2 Experiment 9 design. Numbers refer to the rat’s age in days. This experiment kept the training, re-training, re-extinction, and test age constant across groups. The only age that varied across groups was the age at the time of the initial extinction training session (either 17 or 24 days of age). 135

Figure 4.13 Histological reconstruction of coronal sections showing cannula placements in the amygdala for Experiment 9. 136

Results and Discussion

For the initial extinction session (drug-free) that occurred either at 17- or 24-days of age, all groups exhibited substantial levels of freezing in the first minute of extinction that decreased substantially by the last minute (Figure 4.14). A mixed-design ANOVA of these data yielded a significant main effect of Block [F (1, 25) = 99.14, p < .0001], but no effect of Extinction Age [F (1, 25) = 2.19, p = .152] nor Block x Extinction Age interaction [F (1, 25) = 2.38, p = .136]; there were no other significant main effects or interactions (Fs < 1). The null effect of Extinction Age indicates that rats displayed high levels of freezing even when extinction occurred at P24 (i.e., 8 days after the original conditioning episode), showing that these rats did not forget over the retention interval.

Figure 4.14 Mean (± SEM) percentage freezing to the CS during the initial extinction training (drug-free) in Experiment 9. There were no significant group differences (i.e., the rates of extinction were similar across different ages of initial extinction training). . 137

Figure 4.15 Mean (± SEM) percentage freezing to the CS during re-extinction in Experiment 9. Saline groups exhibited good extinction, whereas Bupivacaine groups showed negligible amount of freezing throughout re-extinction training.

For re-extinction, both Saline groups exhibited substantial levels of freezing in the first block of extinction that decreased substantially by the last block, whereas both

Bupivacaine groups exhibited very low levels of freezing throughout (Figure 4.15). A mixed-design ANOVA of these data yielded significant main effects of Block [F (1, 25)

= 23.32, p < .0001] and Drug [F (1, 25) = 43.62, p < .0001], as well as a Block x Drug interaction [F (1, 25) = 17.10, p < .0001]. There were no effects of Extinction Age [F (1,

25) = 1.95, p = .175], nor Drug x Extinction Age interaction [F (1, 25) = 2.84, p = .104]; there were no other significant main effects or interactions (Fs < 1). These results indicate that re-extinction was comparable across Saline groups regardless of when 138

initial extinction occurred, and that Bupivacaine groups had their amygdala inactivated as indicated by the absence of freezing.

Figure 4.16 Mean (± SEM) percentage freezing to the CS during test in Experiment 9. Temporary inactivation of the amygdala disrupted re-extinction when the initial extinction training occurred at 17 days of age, but when the initial extinction training session occurred at 24 days of age, inactivation of the amygdala at re-extinction had no effects. P17-Saline (n = 7), P17-Bupivacaine (n = 8), P24-Saline (n = 7), and P17-Bupivacaine (n = 7). * Indicates a significant difference to the other groups.

The mean and SEM levels of CS-elicited freezing at test are shown in Figure

4.16. Temporary inactivation of the amygdala disrupted re-extinction when the initial extinction training occurred at 17 days of age. However, when the initial extinction training session occurred at 24 days of age, inactivation of the amygdala at re-extinction 139

had no effects (i.e., the rats with the inactivated amygdala at re-extinction exhibited as much extinction at test as did rats with an active amygdala). ANCOVA revealed a significant main effect of Extinction Age [F (1, 24) = 7.08, p < .05], and an Extinction

Age x Drug interaction [F (1, 24) = 8.01, p < .01]. The effect of Drug was not significant [F (1, 24) = 4.01, p = .057]. The significant interaction confirms that temporary inactivation of the amygdala at re-extinction has differential effects depending on the age at the time of initial extinction. Subsequent post hoc comparisons, with Tukey’s HSD test, showed that group P17-Bupivacaine froze significantly more than all the other groups (p < .005); no other group differences were found. These results indicate that amygdala-dependence or -independence of re-extinction is critically determined by the age at initial extinction.

140

Discussion

The experiments reported in Chapters 2 and 3 demonstrated that 17-day-old rats fail to exhibit a variety of extinction-related phenomena whereas 24-day-olds do. To extend these findings, the experiments in Chapter 4 examined the effect of temporarily inactivating the amygdala on extinction and re-extinction in the developing rat. It was observed in Experiments 7.1 and 7.2 that extinction retention was impaired in both 24- and 17-day-old rats if the amygdala is inactivated during extinction training. This is the first demonstration of amygdala involvement in extinction at these ages. Experiment 8.1 showed that the amygdala was not involved in re-extinction of a CS that had been previously extinguished and then re-trained in P24 rats. In contrast, re-extinction was still amygdala-dependent in P17 rats in Experiment 8.2. Interestingly, it was shown in

Experiment 9 that the amygdala-dependence of re-extinction was determined by the age at initial extinction, not the age at conditioning, re-conditioning, re-extinction, or test.

That is, when age was kept constant across these other stages, re-extinction was amygdala-independent if initial extinction occurred at 24 days of age but amygdala- dependent if initial extinction occurred at 17 days of age.

Considering that the amygdala is a site for acquisition, consolidation, and storage of fear memories in adult rats (e.g., Gale et al., 2004; Phelps & LeDoux, 2005;

Davis, 2006), it is not surprising that fear extinction also involves the amygdala in 24- and 17-day-old rats. The amygdala appears to develop very early, because studies have shown that various neurotransmitter systems that are involved in conditioned fear in adult rats are also involved in fear conditioning in the developing rat (e.g., Kim et al.,

2006; Weber, McNally, & Richarson, 2006; Langton et al., 2007). For example, pre- training systemic injections of the NMDA antagonist MK-801 prevents fear conditioning in 17- and 24-day-old rats (Langton et al., 2007). NMDA neuro- 141

transmission has been long-known to be important for fear conditioning in the adult rat

(e.g., Kandel, 2001). Other neurotransmitter systems such as the GABAergic and the endogenous opioid system also participate in conditioned fear in the infant rat (Kim et al., 2006; Weber et al., 2006). The involvement of these neurotransmitter systems in fear conditioning in infant rats suggest that the amygdala may function in an adult-like manner for conditioning in these rats.

One study has explicitly shown that the amygdala participates in conditioned fear early in development (Sullivan et al., 2000). In this study, P8 and P12 rats were injected with 14C-2-deoxyglucose (radio-labelled glucose analogue) before being conditioned to an odour CS. P8 and P12 rats were chosen based on the previous finding that showed reliable odour conditioning in P12 rats but not in P8 rats (e.g., Sullivan,

Wilson, Lemon, & Gerhardt, 1994). Sullivan et al. (2000) reported that uptake of 14C-

2-deoxyglucose in the amygdala was enhanced in conditioned P12 rats compared with non-conditioned control rats. On the other hand, the uptake of the labelled glucose analogue by the amygdala did not differ across conditioning conditions in P8 pups. To put simply, metabolic activity was enhanced in the amygdala as a result of conditioning in P12 rats but not in P8 rats. These results indicate that the amygdala participates in fear conditioning in rats by 12 days of age. Hence, it is likely that fear memories, once acquired, are stored in the amygdala in P24 and P17 rats. Therefore, extinction of conditioned fear at these ages would be amygdala-dependent.

Although the amygdala is involved in extinction in both 24- and 17-day-old rats,

Experiments 8.1, 8.2, and 9 suggest that the processes mediating extinction at these ages may be different. These experiments showed that if extinction occurs at 24 days of age, re-extinction is amygdala-independent whereas the amygdala is still necessary for re- extinction if extinction occurs at 17 days of age. The findings with the 24-day-old rats 142

replicate those in adult rats (e.g., M. Weber et al., unpublished observations). From this, it appears that the amygdala is involved in the initial learning of the CS-no US memory acquired in extinction training, but once that memory has been acquired then the amygdala is no longer needed for subsequent extinction training episodes, at least in

P24 and adult rats.

As extinction appears to be a new learning in the adult rat, perhaps it is not surprising that extinction for the second time is different from extinction the first time.

This is because of studies that suggest acquisition of conditioned fear is also different when it happens for the second time. For example, Sanders and Fanselow (2003) showed that microinfusion of the NMDA antagonist 5-amino-phosphonovaleric acid

(APV) into the hippocampus impaired contextual fear conditioning. However, if rats first received conditioning in context A, and then received infusion of APV prior to conditioning in context B, fear conditioning to context B was not impaired. In short, fear acquisition to context B became NMDA-independent by the prior conditioning experience in context A.

Research using step-down inhibitory avoidance procedure also suggests that learning for the second time may involve different neural processes than initial learning.

In this procedure, rats are first placed on a platform that stands above a grid floor. A training trial involves the rats receiving a single shock when in contact with the grid floor. At test, learning is indicated by a longer latency to step-down in shocked rats compared to non-shocked rats. This learning can be enhanced by giving rats more trials.

In Cammarota, Bevilaqua, Medina, and Izquierdo (2004) it was shown that microinfusion of the protein synthesis inhibitor anisomycin into the BLA or hippocampus blocked step-down inhibitory avoidance learning when given before the first trial. Interestingly, giving anisomycin before the second trial did not prevent extra 143

learning occurring from the second trial. Another study using the same procedure showed that infusion of APV into the hippocampus after training did not block learning if the rat had been pre-exposed to the training apparatus (Roesler et al., 1998). Taken together, these studies indicate that learning the first time may be fundamentally different to learning the second time. Just as there are differences between learning and re-learning, there also should be differences between extinction and re-extinction.

Finding that re-extinction is amygdala-dependent in P24 rats, along with previous findings on reinstatement, renewal, FG7142, and MK-801 effects in the developing rat (Chapters 2 & 3; Langton et al., 2007; Yap & Richardson, 2007), indicates that the processes mediating extinction in 24-day-old rats is the same, or at least very similar, to those that mediate extinction in the adult rat.

Extinction in the 17-day-old rat and current models of extinction

In the experiments presented in this chapter, it was observed that the amygdala is necessary for re-extinction if extinction occurs at 17 days of age (Experiment 8.2 &

9), which is contrary to what is found in older rats. This suggests that fundamentally different neural mechanisms are involved in extinction at this age. Current models of extinction implicate the amygdala, mPFC, and hippocampus as components of the neural circuitry of extinction (e.g., Quirk et al., 2006; Sotres-Bayon et al., 2006; Myers

& Davis, 2007). Interestingly, both the mPFC and hippocampus are late-maturing structures in the rat. For example, the cortical layers of the mPFC attain their adult proportional width around 24 days of age (Van Eden & Uylings, 1985). Also, the hippocampus has long been known for its delayed maturation (e.g., Wilson, 1984); and behavioural studies show that young rats (e.g., 18-day-olds) are impaired in learning about context (e.g., Rudy, 1993; Rudy & Morledge, 1994). Hence, the observed 144

developmental differences in extinction may be due to 17-day-old rats having a different neural circuitry for extinction. The results reported in Chapters 2, 3, as well as some other recent findings support this notion. Specifically, renewal and reinstatement are not observed in 17-day-old rats, suggesting that the hippocampus is not involved in extinction in these rats (Chapters 2 & 3; Yap & Richardson, 2007). Further, the failure to observe GABAergic modulation of extinction expression in 17-day-old rats (Chapter

3) indicates the absence of the mPFC-mediated inhibition of the amygdala for extinction in these rats. Finally, using immuno-labelling of the long-term cell plasticity marker phosphorylated MAPK (pMAPK), I have obtained preliminary evidence showing post- extinction pMAPK activation in the vmPFC in 24-day-old rats but not in 17-day-old rats (J. H. Kim & R. Richardson, unpublished observations). Taken together, these findings suggest that the hippocampus and the mPFC are not involved in extinction of conditioned fear in P17 rats.

Suggesting that a different neural circuitry underlying extinction at 17 days of age can explain the findings reported in the current chapter on the amygdala- dependence of re-extinction in two ways. The first explanation is that the CS-no US extinction memory is stored in the amygdala in 17-day-old rats. In the adult models of extinction, the storage of the extinction memory is not localised in one neural structure, but is posited to be dependent on the entire amygdala-mPFC-hippocampus circuit (e.g.,

Barad et al., 2006; Farinelli et al., 2006; Quirk & Mueller, 2008). At 17 days of age, the initial extinction memory may be stored solely in the amygdala due to vmPFC and hippocampus immaturity, as a consequence, inactivation of the amygdala disrupted re- extinction. The second explanation is that due to the different neural circuit involved in extinction at P17, extinction involves unlearning of the original CS-US association in

P17 rats. In rats 24 days of age, initial extinction involves learning a CS-no US 145

association, hence subsequent re-extinction may simply involves retrieval of the original CS-no US memory, therefore, the amygdala is then no longer necessary for re- extinction. On the other hand, if initial extinction at 17 days of age involves unlearning, subsequent re-extinction may still require the amygdala because re-extinction cannot be a retrieval of the CS-no US memory. Therefore, when initial extinction training was given at 17 days of age and re-extinction at 27 days of age in Experiment 9, re- extinction was amygdala-dependent because the initial extinction at 17 days of age involved unlearning. This explanation is based on the idea that the involvement of the hippocampus and the mPFC allows flexibility in learning so that the CS-US and CS-no

US memories can co-exist (e.g., Rhodes & Killcross, 2004). Not having these structures participate in extinction at 17 days of age may provide no such flexibility, so that extinction leads to erasure of the original CS-US memory.

This unlearning account for extinction in P17 rats is supported by the recent finding that pre-extinction systemic injection of the NMDA antagonist MK-801 does not disrupt extinction in 17-day-old rats (Langton et al., 2007). NMDA-dependence of extinction learning in the adult rat is taken as evidence for the new learning account of extinction (e.g., Lattal et al., 2006). Hence, NMDA-independent extinction in 17-day- olds rats suggests that extinction is not new learning in these rats. Further, extinction expression is not mediated by GABAergic inhibitory activity in 17-day-old rats

(Chapter 3), which implies that extinction is not inhibition in rats this age. As mentioned in Chapter 3, recent reviews on extinction now suggest both new learning and unlearning as mechanisms for extinction in the adult rat (Barad et al., 2006;

Delamater, 2004; Lattal et al., 2006; Myers & Davis, 2007). It may be the case that when extinction occurs early in development, the balance between unlearning and new learning processes of extinction is simply shifted, in that extinction relies more on 146

unlearning rather than new learning. A potential way of testing whether unlearning occurs in extinction at 17 days of age is to record neural activity in LA neurons after extinction. After CS–US pairings, CS-elicited neural activity in LA neurons increases in the adult rat; during extinction, the responses of many of these cells return to pretraining levels (Quirk et al., 1995; Repa et al., 2001). However, Repa et al. (2001) found a population of LA cells in which CS-elicited activity remained elevated throughout extinction training. Repa et al. (2001) inferred that these cells represent the original CS–

US memory, demonstrating that extinction does not erase the original CS–US association in the adult rat. Investigating whether this population of cells show a different response after extinction in 17-day-old rats compared to older rats would be helpful in determining whether unlearning is a mechanism for extinction in the developing rat.

Considering that extinction processes may be dissociated across development, it is clear that more work needs to be done in examining extinction in the developing rat.

Such research not only provides a unique way of assessing extinction processes, but is important also because of its clinical implications. Contrary to the long-held belief that fear acquired early in life is particularly resistant to the effects of extinction (e.g.,

Jacobs & Nadel, 1985), the findings reported in the present thesis suggest that such fear can be much more effectively extinguished if treated early enough in development. 147

Chapter Five.

General Discussion

The experiments reported in this thesis examined extinction of conditioned fear in the developing rat. The experiments described in Chapter 2 investigated whether there is a developmental difference in the reinstatement of an extinguished fear response following a post-extinction administration of the US. Experiment 1 showed that a US treatment led to the recovery of extinguished fear in P24 rats. This effect was also context-dependent in rats this age as they failed to exhibit the reinstatement effect if they were tested in a context different from where the US had been given (Experiment

2). In contrast to these findings, P17 rats did not display reinstatement of extinguished fear (Experiments 1 & 2). The failure to observe reinstatement in P17 rats was not due to the US treatment being ineffective in rats this age because the same treatment alleviated spontaneous forgetting (Experiments 3.1, 3.2, & 3.3). Specifically,

Experiment 3.1 showed that rats trained at 16 days of age with weak conditioning exhibited spontaneous forgetting over a 2 day retention interval. Experiments 3.2 and

3.3 then showed that this forgotten memory recovered when the same US treatment used in Experiments 1 and 2 was given prior to test. These latter findings indicate that the failure to observe reinstatement of extinguished fear in P17 rats (Experiments 1 & 2) is not due to a general ineffectiveness of the US in mediating retrieval of a fear memory in rats this age. Rather, it appears that P17 rats are impaired in showing the post- extinction reinstatement effect.

The series of experiments in Chapter 3 then examined renewal and GABAergic involvement in extinction in P24 and P17 rats. Experiment 4 showed that P24 rats displayed renewal when tested in a different context to where they had received 148

extinction training, whereas P17 rats did not. The younger rats exhibited reduced fear

(i.e., extinction) regardless of the context in which they were tested. Experiments 5.1 and 5.2 replicated the results of Experiment 4 (i.e., a developmental difference in context-modulation of extinction), and additionally showed that pre-test injection of

FG7142 recovered extinguished fear in P24 rats but not in P17 rats. The failure to observe a recovery of an extinguished fear response following pre-test administration of

FG7142 in P17 rats was not due to the dose selected (Experiment 5.3). Further, the failure to see any effect of FG7142 on extinction in P17 rats cannot be attributed to a general lack of responsiveness to this drug in P17 rats, as FG7142 was found to be effective in alleviating spontaneous forgetting in rats this age (Experiments 6.1 & 6.2).

Taken together, these results show that extinction in P24 rats is susceptible to context manipulations and pre-test alterations in GABAergic neurotransmission while extinction in P17 rats is not influenced by either.

Chapters 2 and 3 together provided strong evidence for a developmental dissociation in the mechanisms underlying fear extinction: when extinction occurs at 17 days of age, rats fail to show renewal or reinstatement of extinguished fear, whereas when extinction occurs at 24 days of age, rats do show these phenomena. Also, extinction in P24 rats is susceptible to pre-test alterations in GABAergic neurotransmission whilst extinction in P17 rats is not susceptible to such a manipulation.

To investigate whether there is any evidence of a developmental dissociation in the neural structures involved in extinction of conditioned fear, the experiments described in Chapter 4 examined the effect of temporary inactivation of the amygdala on extinction and re-extinction in the developing rat. It was observed that extinction retention is impaired in both P24 and P17 rats if the amygdala is inactivated during extinction training (Experiments 7.1 & 7.2). In Experiments 8.1 and 8.2, P24 and P17 149

rats (respectively) were conditioned, extinguished, and then re-conditioned to the same

CS. Following re-conditioning the CS was re-extinguished; at this time, some rats at each age had their amygdala temporarily inactivated. Re-extinction was amygdala- independent in P24 rats. However, re-extinction was still amygdala-dependent in P17 rats. Further, Experiment 9 showed that when age was kept constant across conditioning, re-conditioning, re-extinction and test, re-extinction was amygdala-independent if initial extinction occurred at 24 days of age but amygdala-dependent if initial extinction occurred at 17 days of age. That is, amygdala involvement in re-extinction was critically determined by the age when the initial extinction occurred.

In every one of these experiments the magnitude of original fear learning as measured by the levels of CS-elicited freezing during extinction did not differ between

P17 and P24 rats. The rate of extinction also did not differ between different ages, as confirmed by statistical analyses. Therefore, the developmental dissociations observed in extinction across the experiments in this thesis are not due to potential quantitative differences in fear learning or the rate of extinction in these ages. Furthermore, these developmental dissociations are not due to potential qualitative differences in fear learning at different ages because Experiment 9 showed that even when rats are conditioned at the same age, a developmental dissociation is observed if the initial extinction occurs at different ages. Taken together, the developmental differences observed in the present research suggest different processes underlie extinction at P17 days of age compared to older ages. 150

I. EXTINCTION IS FUNDAMENTALLY DIFFERENT ACROSS

DEVELOPMENT –THEORETICAL ACCOUNTS

It was observed in the present thesis that 24-day-old rats show reinstatement and renewal. Further, a pre-test injection of FG7142 leads to a return of an extinguished fear response in P24 rats. On the other hand, 17-day-old rats did not show any of these extinction-related phenomena. These developmental dissociations in extinction have important theoretical implications because the loss in conditioned responding in P17 rats may be indicative of erasure or unlearning of the original CS-US association acquired during conditioning. That is, extinction in P17 rats may be qualitatively different from extinction in older rats. Alternatively, these developmental differences in extinction may reflect quantitative differences. Specifically, extinction at 17 days of age could be facilitated, or ‘deepened’, extinction, in which the CS is extremely well- extinguished, so that subsequent return of CS-elicited fear is much less likely in these rats. This section of the general discussion will largely focus on distinguishing between these two possible explanations for extinction in P17 rats.

Extinction may be unlearning in 17-day-old rats

Unlearning refers to the permanent loss in associative strength between the CS and the US. For example, the Rescorla-Wagner model (Rescorla & Wagner, 1972) states that the amount learned about a CS on a conditioning trial (V), is a function of the difference, or error, between the actual outcome () and the expected outcome of the conditioning trial (V) (i.e., V = V). The expected outcome of the conditioning trial is the summed associative strengths of all CSs present on that trial. To summarise, the Rescorla-Wagner model states that the error between what actually happens and 151

what is expected (V) drives learning. When the error is positive (i.e., V), so that the actual outcome of the trial exceeds the expected outcome, associative learning occurs (V = +). When the error is zero (i.e., V), so that the actual and expected outcomes are the same, no learning occurs (V = 0). In the case of extinction, this discrepancy is negative (i.e., V) because the expected outcome exceeds the actual outcome (because the CS is not reinforced), so ‘erasure’ of the CS-US association occurs (V = -).

However, this unlearning account of fear extinction was rejected largely based on the findings of spontaneous recovery, reinstatement, and renewal. These phenomena illustrate how extinguished CRs can be restored without additional CS-US pairings, showing that extinction does not lead to erasure of the original CS-US association.

Hence, the failure to observe reinstatement and renewal in P17 rats at least raises the possibility that extinction may be unlearning in rats at this age.

Additionally, data from experiments reported in Chapter 3 suggest that spontaneous recovery is observed in P24 rats but not in P17 rats. That is, although CS- elicited freezing was at a similar level at the end of extinction training at both ages, only

24-day-old rats exhibited a recovery, albeit slight, in level of CS-elicited freezing when tested the next day. To more thoroughly examine this assertion, extinction and test data from P24-Same group in Experiment 4 and Vehicle-Same group in Experiment 5.1 were compiled and analysed for evidence of spontaneous recovery in rats this age.

Similarly, extinction and test data from P17-Same group in Experiment 4 and Vehicle-

Same group in Experiment 5.2 were compiled and analysed to examine spontaneous recovery in rats this age. These groups were chosen because they received both extinction training and test in the same context, and there were no additional treatments such as reinstatement or FG7142 injection in these groups. That is, compared to other 152

Figure 5.1 P24 rats (n = 18) displayed significantly more CS-elicited freezing at test compared to the last block of extinction training, indicating some spontaneous recovery.

Figure 5.2 P17 rats displayed comparable CS-elicited freezing at test compared to the last block of extinction training, failing to show any spontaneous recovery (n = 17). 153

experiments in the present thesis, the conditions of these groups were the closest to the conditions in a typical test for spontaneous recovery. With the combined data, I compared the levels of freezing exhibited by rats of different ages at the end of extinction and at test (Figures 5.1 & 5.2). Performance during extinction training was essentially identical across the two ages. However, it appears that P24 rats froze significantly more at test compared to at the end of extinction (i.e., they showed some spontaneous recovery over the 24 hr interval; Figure 5.1). No such recovery was observed in P17 rats (Figure 5.2). A repeated-measures ANOVA comparing performance at the end of extinction and at test confirmed this description of the combined data: F (1, 17) = 11.49, p <.005, for P24 rats; and F < 1, for P17 rats.

Obviously a proper investigation into age differences in spontaneous recovery after extinction is necessary before making any concrete conclusions, but this post-hoc analysis suggests that P17 rats do not show spontaneous recovery whereas P24 rats do.

Taken together, all of these data suggests that extinction at 17 days of age may be unlearning, as evidenced by the lack of renewal, reinstatement, and spontaneous recovery after extinction.

Extinction may be ‘deepened extinction’ in 17-day-old rats

Alternatively, the failure to see these phenomena in 17-day-old rats may be due to ‘deepened extinction’ (Rescorla, 2006) rather than unlearning. Deepened extinction refers to enhanced extinction when a previously extinguished CS is further extinguished in compound with another previously extinguished CS. For example, Rescorla (2006) trained rats to fear a light CS and a noise CS by pairing each one separately with shock.

Subsequently, rats displayed conditioned suppression in the presence of the light or the noise CS. Then each CS was extinguished individually until conditioned suppression 154

was no longer observed in the presence of either CS. Some rats then further received compound extinction, in which the extinguished CSs were presented together without any shock. On the initial presentations of the compound consisting of the two extinguished CSs, rats exhibited renewed conditioned suppression; over non-reinforced trials the renewed CR gradually decreased. This return of responding when two extinguished CSs are presented in compound had been previously reported by Reberg

(1972). A control group in the experiment by Rescorla (2006) received additional extinction to only the noise CS. After these additional extinction sessions (with either the compound or the noise CS), the rats were tested 7 days later for spontaneous recovery. Interestingly, rats showed significantly reduced spontaneous recovery to the noise CS if it had been subsequently extinguished in compound with the light, compared to if it had been additionally extinguished on its own. That is, the compound extinction procedure deepened, or enhanced, extinction. Further, when a pre-test US reminder was given, rats that had received additional compound extinction exhibited less reinstatement than did rats that had received additional extinction with an individual CS. These findings on reduced spontaneous recovery and reinstatement observed after compound extinction is similar to the findings on extinction in 17-day- old rats reported in the present thesis.

The deepened extinction observed in Rescorla (2006) was explained in light of contemporary error-correction models (e.g., Rescorla, 2003). Contemporary error- correction models are based on the Rescorla-Wagner model but instead of stating that erasure of CS-US association occurs when there is negative error between the expected outcome and the actual outcome (i.e., V), these models state that inhibitory learning occurs. Therefore, in Rescorla (2006), extinction learning was significantly greater in rats that received the compound extinction procedure compared to rats that 155

received single stimulus extinction because Vnoise + Vlight) led to a more negative error (i.e., more inhibitory learning) than Vnoise).

Interestingly, these contemporary error-correction models would also predict that extinction in P17 rats may lead to deepened extinction. This is because these models also incorporate the context as a stimulus that can acquire associative strength in a given learning trial [i.e., VCS + Vcontext)]. As mentioned before, rats readily learn about the context at 23 days of age, but not at 18 days of age (e.g., Rudy, 1993). Hence, at 24 days of age, both context and the CS can gain inhibitory strength during extinction

[i.e., VCS24 + Vcontext) =V]. In contrast, only the CS gains inhibitory strength in extinction at 17 days of age [i.e., VCS17) =V]. In the experiments reported in this thesis, rats of both ages received identical extinction training, therefore, any changes in inhibitory strength acquired in extinction is shared between the CS and the context in

P24 rats, whereas the same amount of associative strength is entirely attributed to the

CS in P17 rats [i.e., VCS24 + Vcontext) VCS17)]. Consequently, inhibitory strength acquired by the CS in extinction training would be larger in 17-day-old rats than in 24- day-old rats, leading to deepened extinction in P17 rats. As a result, P17 rats would be expected to exhibit reduced reinstatement or renewal of extinguished fear.

According to this argument, lesions of the hippocampus should also deepen extinction in the adult rat because such lesions would remove the context as a cue that participates in extinction learning. Indeed, it has been shown that electrolytic and excitotoxic lesions of the hippocampus can abolish reinstatement (e.g., Wilson et al.,

1995; Frohardt et al., 2000) and renewal (e.g., Ji & Maren, 2005). However, there is also evidence suggesting that disrupted hippocampal function during extinction training blocks extinction learning (e.g., Corcoran et al., 2005; Farinelli et al., 2006). More studies are needed to resolve this issue. 156

Deepened extinction as a mechanism underlying extinction in P17 rats, however, cannot explain the failure of FG7142 to restore extinguished conditioned fear in rats this age (see Experiments 5.2 & 5.3). If extinction at 17 days of age leads to enhanced inhibitory learning to the CS, one might expect that these rats would be even more susceptible to the effects of FG7142, a drug that reduces GABAergic inhibitory activity.

Further, there is no reason as to why deepened extinction would lead to a developmental dissociation in amygdala involvement in re-extinction. It was observed in Experiments

8.1, and 8.2 that re-extinction is amygdala-independent in P24 rats (as it is with adult rats), whereas re-extinction in P17 rats requires the amygdala. Such difference in the neural structure involved in re-extinction suggests that extinction at 17 days of age cannot merely be quantitatively different from extinction at older ages. Nonetheless, to thoroughly examine the idea that extinction at 17 days of age is deepened extinction, it would be interesting to test whether the CS becomes “extra” inhibitory after extinction in P17 rats compared to the extinguished CS in P24 rats. For example, summation and retardation tests of inhibitory strength of a CS could be used. Such tests will clarify whether extinction is quantitatively different across ages.

There are other theories that may account for the developmental dissociations in extinction reported in this thesis. For instance, in the ‘recency to primacy theory’ of learning, Lubow and De la Casa (2005) suggested that when an animal learns two competing associations (e.g., CS-US vs CS-no US in the case of extinction), the animal expresses the most recent memory at the immediate retention test because the test context is most similar to the context of recent learning. As time passes, however, the context changes from that of the most recent learning and the animal starts to express the memory that was first acquired. This is because there is an inherent positive bias 157

towards the retention of the earliest memory (i.e., the primacy effect). The recency to primacy theory successfully explains spontaneous recovery observed after extinction as well as ‘super’ latent inhibition observed in the adult rat (Lubow & De la Casa, 2005).

Super latent inhibition refers to enhanced latent inhibition (i.e., decreased CR) when the rat is tested at a long retention interval following latent inhibition training. As context change is important for the recency to primacy shift to occur according to Lubow and

De la Casa (2005), one could argue that such a shift does not occur in the P17 rat due to its relative inability in processing contextual information. That is, the P17 rat always expresses the most recent memory. Therefore, the failure to see renewal, reinstatement, or the FG7142 effect in the experiments reported in this thesis may be due to P17 rats always expressing the most recent memory (i.e., extinction). However, if P17 rats do indeed always express the most recent memory, then they should also fail to express latent inhibition. However, data from our laboratory indicate that rats this age exhibit latent inhibition (Yap & Richardson, 2005). It therefore appears that the developmental differences in extinction found in the present research are not due to any bias in the primacy or the recency effect in memory retention in P17 rats.

Another potential explanation for the developmental dissociations in extinction of conditioned fear observed in the present research relies on non-associative accounts of extinction. There is evidence showing that non-associative factors may influence extinction in the adult rat, including changes in the ways in which the animal processes the CS (e.g., Kamprath & Wotjak, 2004; Pavlov 1927). For example, habituation is a type of non-associative learning involving a decrement in responding to a stimulus as a consequence of repeated exposure to that stimulus. In the present experiments, the decrease in CR after extinction may represent habituation to the CS because extinction training was comprised of 30 presentations of the CS. Therefore, one could argue that 158

the developmental differences observed in the present research were due to differences in habituation exhibited by P24 and P17 rats. Specifically, the difference would be that in P24 rats, habituation to the CS is lost due to the passage of time or various post- extinction manipulations, whereas in P17 rats, habituation is not lost by such manipulations. However, this explanation contradicts previous findings on non- associative learning in the developing rat. For example, Parsons, Fagan, and Spear

(1973) showed that retention of habituation was correlated with age; once habituation was acquired to a novel stimuli, 15-day-old rats failed to retain habituation at 1 hr retention interval, whereas 25-day-old rats continued to exhibit habituation even after

24 hrs. According to this finding, P17 rats should recover an extinguished fear response much more readily than P24 rats. This was clearly not the case in the present thesis.

Therefore, the developmental dissociations in extinction observed in the current thesis are unlikely to be due to developmental differences in habituation.

Taken together, in light of the theoretical models of learning discussed above, the results from the present thesis suggest that extinction processes are fundamentally distinct in 17-day-old rats compared to older rats. It appears that various theories that account for extinction and its related phenomena in the adult rat fail to explain the present findings on extinction in the 17-day-old rat. Interestingly, the theoretical model that most successfully accounts for the findings on extinction in P17 rats is the

Rescorla-Wagner model that suggests extinction leads to the unlearning of the CS-US association. If unlearning is indeed involved in extinction at 17 days of age but not at older ages, then it is likely that maturation of the brain that occur during development is responsible for such fundamental differences found in extinction processes across development. In the next section, some of the neural differences between P17 rats and older rats that may lead to developmental dissociations in extinction are discussed. 159

II. EXTINCTION IS FUNDAMENTALLY DIFFERENT ACROSS

DEVELOPMENT –NEURAL ACCOUNTS

The current neural models of extinction state that connections between the amygdala, mPFC, and the hippocampus are involved in extinction of conditioned fear in the adult rat (Garcia, 2002; Herry & Garcia, 2002; Hobin et al., 2003; LeDoux & Gorman, 2001;

Milad & Quirk, 2002; Morgan & LeDoux, 1995; Quirk et al., 2000; Sotres-Bayon et al.,

2006). To briefly summarise, these models suggest that the mPFC inhibits the amygdala by activating inhibitory GABAergic interneurons and/or projecting neurons (in ITC cell mass) during the expression of extinction. Additionally, the hippocampus can inhibit or excite the mPFC and thereby modulate the expression of extinction. For example, when the rat is outside of extinction context, renewal may occur because the hippocampus no longer excites the mPFC, which then no longer inhibits the amygdala.

In the present thesis, it was demonstrated that rats display reinstatement, renewal, and FG7142-induced recovery of extinguished fear if extinction occurs at 24 days of age. These results strongly suggest that the hippocampus-mPFC-amygdala network functions in an adult-like manner in rats this age. On the other hand, if extinction occurs at 17 days of age, rats do not show reinstatement, renewal, or FG7142-induced recovery of extinguished fear. These findings strongly suggest that the neural circuit that mediate extinction in P17 rats is very different from that involved in older rats.

As one possibility, the failure to see renewal and reinstatement in P17 rats may be due to the hippocampus not modulating the mPFC at this age. In other words, the developmental differences in extinction observed in the present study may be a consequence of the well-established developmental difference in contextual conditioning (e.g., Rudy, 1993). However, this possibility cannot account for the failure to observe a return of extinguished fear following pre-test injection of FG7142 in P17 160

rats. This particular result suggests that GABAergic inhibition is not involved in the processes mediating extinction in rats this age. As previous findings indicate that

GABAergic inhibition in the amygdala is mediated by the mPFC (e.g., Milad et al.,

2004), the lack of FG7142 effect on extinction in P17 rats suggests that the mPFC may not be involved in extinction in P17 rats.

The mPFC, in fact, is a late-maturing structure like the hippocampus. For example, Kolb and Nonneman (1978) examined the functional maturation of the mPFC by permanently destroying the mPFC in rats at different stages of development. Then they observed the effects of these lesions when the rats became adults on mPFC- dependent behavioural tasks. One mPFC-dependent task examined was spatial reversals, in which food-deprived rats were trained to choose an alley in a T-maze that contained food. This task is called “spatial reversals” because the food is put into a different alley on each trial, so that the rats are required to reverse their choice of an alley between trials. Interestingly, Kolb and Nonneman (1978) found that if mPFC lesions were made at 5 or 9 days of age, the ability of rats to learn this task as adults was unaffected.

However, if lesions were made at 25, 35, or 40 days of age, then rats were impaired at switching responses between trials. Kolb and Nonneman (1978) concluded from these results that the mPFC is functionally immature until approximately 25 days of age.

They reached this conclusion based on the assumption that if a neural structure is lesioned before it has reached its functional maturity, other neural structures can compensate for its loss. As a consequence, behavioural tasks that are normally dependent on that neural structure remain unaffected. On the other hand, if the lesion is made after a neural structure has reached its functional maturity, then other structures do not compensate for its loss. Therefore, behavioural tasks that are dependent on that particular neural structure are disrupted. 161

However, it is difficult to determine whether the mPFC is functionally mature at

17 days of age from Kolb and Nonneman’s (1978) study because no lesions were made between 9-25 days of age. Clearer evidence on mPFC maturation comes from a study by Van Eden and Uylings (1985) that showed the cortical layers of the mPFC is immature at 18 days of age, but attain their adult proportional width by 24 days of age.

From that study, it appears that the mPFC is mature at 24 days of age, but not at 18 days of age. This idea is indirectly supported by a recent study that showed P21 rats were impaired in learning a mPFC-dependent task (i.e., spatial reversals), whereas P26 rats could learn this task (Watson, Sullivan, Frank, & Stanton, 2006). Taken together, these studies suggest that in P24 rats the mPFC is mature enough to participate in extinction, whereas in P17 rats the mPFC is not mature and does not participate in extinction.

Moreover, our laboratory has obtained preliminary evidence suggesting that extinction of conditioned fear in P17 rats does not involve the mPFC (J. H. Kim & R.

Richardson, unpublished observations). In that study, P17 and P24 rats were conditioned and extinguished to an auditory CS. Rats in No Extinction groups at each age were conditioned but were only exposed to the experimental chamber without any

CS presentations for an equal amount of time as the extinguished rats. One hour after the extinction session (or context exposure for the no extinction Control group), all rats were perfused and the brains were collected for immuno-labelling of the long-term cell plasticity marker pMAPK. This extinction-perfusion time interval was chosen based on a previous study that examined post-extinction pMAPK in the BLA of adult rats (Herry et al., 2006). We found that rats of both ages exhibited similar levels of CS-elicited freezing during extinction (Figure 5.3). Immunohistochemistry data revealed that extinguished P24 rats had elevated pMAPK neuron count in the amygdala and the IL compared to P24 rats that did not receive extinction (Figures 5.4, 5.5, & 5.6). This result 162

Figure 5.3 Mean freezing (+ or – SEM) by rats in response to the CS during the extinction training in J. H. Kim and R. Richardson (unpublished observations). One extinction block consisted of 3 CS presentations.

Figure 5.4 Total pMAPK stained neuron count in the IL and amygdala in J. H. Kim and R. Richardson (unpublished observations) [P17-No Extinction (n = 4); P17-Extinction (n = 3); P24-No Extinction (n = 4); and P24-Extinction (n = 3)]. 163

Figure 5.5 Phosphorylated MAPK cell staining in the IL in J. H. Kim and R. Richardson (unpublished observations). (A) Non-extinguished P24 rats. (B) Extinguished P24 rats. Area in the white square is magnified in (C); scale bar = 100 m. (D) Non-extinguished P17 rats. (E) Extinguished P17 rats; scale bar = 300 m. 164

Figure 5.6 Phosphorylated MAPK neuron staining in the amygdala in J. H. Kim and R. Richardson (unpublished observations). The neurons were counted from the BLA, CeA and the bed nucleus of the stria terminalis intra-amygdaloid division (STIA). (A) Non-extinguished P24 rats. (B) Extinguished P24 rats. (C) Non-extinguished P17 rats. (D) Extinguished P17 rats; scale bar = 300 m. 165

is consistent with previous findings on MAPK involvement in the amygdala and the mPFC in extinction in adult mice and rats (Herry et al., 2006; Hugues et al., 2004). In

P17 rats, however, the elevated pMAPK count as a result of extinction was only found in the amygdala (Figures 5.4, 5.5, & 5.6). This finding strongly indicates that the IL is involved in extinction in P24 rats but not in P17 rats. It would be interesting to examine the effect of intra-IL infusion of a MAPK inhibitor before extinction training.

According to these preliminary data, MAPK inhibition in the IL may disrupt extinction in P24 rats but not in P17 rats.

The findings of J. H. Kim and R. Richardson (unpublished observations), along with evidence of mPFC immaturity at P17 (e.g., Van Eden & Uylings, 1985), indicate that a likely account for the developmental dissociations in extinction reported in the present thesis is that a different neural circuit mediates extinction in P17 rats compared to that which mediates extinction in P24 and older rats. Specifically, the neural circuit involved in extinction in P24 rats may be adult-like, involving the mPFC, amygdala, and the hippocampus (Figure 5.7A). In contrast, extinction in P17 rats may be amygdala-dependent but is mPFC- and hippocampus-independent (Figure 5.7B). That is, due to the non-involvement of the mPFC and hippocampus in extinction in P17 rats, the neural mechanisms required for extinction at this age may rely largely on the amygdala.

This reliance on the amygdala for extinction in P17 rats lead to the following two possible accounts for extinction of learned fear at this age: the storage of extinction memory solely in the amygdala; or the erasure of original fear memory that is stored in the amygdala. The second possibility is based on the idea that the involvement of the hippocampus and the mPFC allows flexibility in learning so that the CS-US and CS-no

US memories can co-exist (e.g., Rhodes & Killcross, 2004). Not having these structures 166

participate in extinction at 17 days of age may provide no such flexibility, so that extinction leads to erasure of the original CS-US memory.

Figure 5.7 (A) Neural structures involved in extinction of conditioned fear in P24 rats are likely to be similar to those involved in extinction in the adult rat (adapted from Sotres-Bayon et al., 2004). As with adult rats, extinction in P24 rats probably involves hippocampal modulation of the mPFC, which in turn controls LA projections to the CeA. The mPFC also modulates the GABAergic ITC cells, which projects to the CeA. (B) Extinction may be hippocampus- and mPFC-independent in P17 rats, which may result in greater dependence on the amygdala for extinction learning at this age. 167

Extinction memory may be stored in the amygdala in 17-day-old rats

The neural models of extinction that have been developed on work done with the adult rat posit that the storage of extinction memory is not localised in one neural structure but is dependent on the entire amygdala-mPFC-hippocampus circuit (e.g., Barad et al.,

2006; Quirk & Mueller, 2008). Finding that re-extinction is amygdala-independent in the adult rat is consistent with this idea (e.g., Laurent et al., in press; M. Weber et al., unpublished observations) because if extinction memory was stored solely in the amygdala, then extinction should always be amygdala-dependent.

It appears that extinction memory is also stored across the amygdala-mPFC- hippocampus circuit in P24 rats because it was demonstrated in Chapter 4 that P24 rats display amygdala-independent re-extinction. In contrast, re-extinction was amygdala- dependent in P17 rats (Experiments 8.2 & 9). This finding suggests that at 17 days of age, extinction memory may be stored solely in the amygdala. A potential mechanism for extinction memory storage in the amygdala is the GABAergic ITC cells because these cells exhibit NMDA-mediated long-term plasticity (Royer & Pare, 2002; 2003).

For example, CS presentations in extinction training leads to BLA activation; then excitatory inputs from BLA to ITC cell masses can trigger activation of ITC cells. Over many CS-alone presentations, a long-term increase in the efficacy of the CS in activating the ITC cells may occur. Consequently, after extinction, CS presentation leads to activation of the ITC cells that inhibit CeA output neurons.

However, this possibility is unlikely for at least two reasons. First, it was demonstrated in Chapter 3 that pre-test alterations in GABA neurotransmission does not affect extinction in P17 rats. Such finding implies that ITC cells (which are GABAergic) are not involved in extinction at this age. Second, it was recently demonstrated that extinction in P17 rats is NMDA-independent (Langton et al., 2007). Therefore, although 168

ITC cells appear to be capable of long-term memory storage because these neurons express NMDA-mediated plasticity in the adult rat (e.g., Royer & Pare, 2002), at 17 days of age, extinction memory is probably not stored in these neurons because extinction is NMDA-independent at this age.

Another possible mechanism for the storage of extinction memory in P17 rats is

LVGCC activation. As mentioned previously, intracellular molecular processes necessary for long-term memory can be triggered by LVGCC activation independently of NMDA neurotransmission (Figure 1.2). Further, it has been shown in the adult rat that pre-training injections of LVGCC inhibitors completely block extinction in the adult rat, and this process appears to be unique for extinction and not acquisition of conditioned fear (Cain et al., 2002). No-one has yet examined whether LVGCC activation is involved in extinction at 17 days of age. Demonstrating that LVGCC blockade impairs extinction in P17 rats may provide evidence that extinction at this age is a new learning that requires long-term cellular plasticity.

Extinction may be erasure in 17-day-old rats

The second possible consequence of extinction in P17 rats relying largely, if not solely, on the amygdala is that extinction involves erasure of the original CS-US association stored in the amygdala. Evidence supporting this view comes from Experiments 8.1, 8.2, and 9 that showed re-extinction was amygdala-independent in P24 rats but amygdala- dependent in P17 rats. It appears that in P24 rats re-extinction simply involves a retrieval of the original CS-no US memory acquired during the initial extinction. As it is believed that the extinction memory is stored across the mPFC-amygdala-hippocampus network, the retrieval of the CS-no US memory does not necessarily require the amygdala; therefore, re-extinction is amygdala-independent in P24 rats. In contrast, re- 169

extinction was amygdala-dependent in P17 rats. This could be due to the CS-no US memory being stored solely in the amygdala in these rats, as discussed previously. Or, this could be due to re-extinction not being a retrieval of the CS-no US memory because there was no such memory to retrieve because initial extinction at 17 days of age was erasure rather than new learning.

A potential mechanism for erasure in extinction is ‘depontentiation’ (e.g., Lin,

Lee, Huang, Wang, & Gean, 2005; Lin, Lee, & Gean, 2003a). Depotentiation refers to a return of potentiated synapses (i.e., in the case of LTP) back to baseline synaptic efficacy when low-frequency stimulation is applied to afferent pathways within a few minutes following LTP induction (Bashir & Collingridge, 1994). Depotentiation is different to long-term depotentiation (LTD) in that LTD refers to below-baseline synaptic efficacy that can be triggered independently of LTP; depotentiation, on the other hand, returns potentiated synaptic efficacy (as a result of LTP) back to baseline.

Therefore, depotentiation can only be triggered within minutes after LTP. Interestingly, depotentiation entails molecular processes that are the opposite to those of LTP. As mentioned earlier, LTP is a neural correlate for long-term memory and involves phosphorylation of intracellular transcription factors such as MAPK (Figure 1.2) and down-regulation of phosphatase (e.g., calcineurin) activity (e.g., Bliss & Collingridge,

1993). On the other hand, depotentiation is correlated with dephosphorylation of MAPK and up-regulation of calcineurin (e.g., Lin et al., 2001; Lin et al., 2003b). In other words, low-frequency stimulation following LTP leads to depotentiation of LTP, which counteracts molecular processes that lead to long-term cellular plasticity that are necessary for acquisition of memory. Taken together, depotentiation appears to be a potential neural mechanism underlying erasure of memory. 170

Indeed, Gean and his colleagues have provided convincing evidence suggesting that depotentiation reflects erasure of conditioned fear in the adult rat (Lin et al., 2001;

Lin et al., 2003a; 2003b; 2003c). They showed that low-frequency stimulation of the amygdala in vivo 10 minutes following fear conditioning ‘quenched’ (i.e., abolished) fear expression as measured by FPS (Lin et al., 2003b). That is, depotentiation directly correlated with changes in behaviour. Thus, synaptic depotentiation could be an intra- amygdala process that decreases fear. Interestingly, extinction and depotentiation appear to share a variety of molecular processes (Lin et al., 2003a; 2003b). Hence, Lin et al. (2003a) proposed that it is possible for fear extinction to cause erasure of the original fear memory via processes such as depotentiation. However, Lin et al. (2003a) noted that depotentiation may not contribute much during extinction because in the laboratory extinction training is typically given 24 hrs after fear conditioning.

Depotentiation, on the other hand, is triggered by low-frequency stimulation administered within minutes of fear conditioning.

Based on these findings on synaptic depotentiation and the quenching of fear,

Myers, Ressler, and Davis (2006) hypothesised that if extinction occurs at a time when depotentiation can be induced (i.e., 10 minutes after fear acquisition), then extinction may trigger depotentiation. In other words, extinction given 10 minutes after fear conditioning may rely on depotentiation processes and erase conditioned fear. This hypothesis also depends on the idea that newly-formed memories undergo a time- dependent process during which they are converted from a short-term, labile state into a long-term, permanent state (i.e., consolidation theory). Myers et al. (2006) predicted that at 10 mins following conditioning, consolidation of fear memory has only begun.

Therefore, extinction training given at this time may counter the consolidation of fear memory, resulting in erasure of the fear memory. They tested this idea using FPS to a 171

light CS, and reported that when extinction was given 72 hrs post-conditioning, rats showed reinstatement, renewal, and spontaneous recovery; however, when extinction was given 10 mins post-conditioning, rats showed none of these phenomena. Myers et al. (2006) suggested that extinction that occurs 10 mins after fear acquisition caused erasure via depotentiation.

The findings of Myers et al. (2006) have significant implications for the present thesis, which demonstrated that P17 rats do not show renewal, reinstatement, nor

FG7142-mediated recovery of extinguished fear. It may be the case that consolidation of fear memory occurs over a much longer time frame in P17 rats compared to P24 and adult rats, so that extinction training 24 hrs post-conditioning in the P17 rat leads to erasure (i.e., depotentiation) in a similar way to extinction 10 mins post-conditioning in the adult rat (Myers et al., 2006).

Accepting depotentiation as a potential mechanism for extinction in P17 rats, however, is inconsistent with some of the existing studies on extinction in the developing rat because depotentiation is NMDA- and LVGCC-dependent (Malleret et al., 2001). Our laboratory has demonstrated that extinction is NMDA-independent in

P17 rats (Langton et al., 2007), therefore, depotentiation at 17 days of age must depend of LVGCC-dependent mechanisms. However, if extinction at 17 days of age causes erasure of the original conditioned fear due to depotentiation (because extinction at this age is similar to extinction given at 10 mins after conditioning in the adult rat), this depotentiation cannot be LVGCC-dependent. This is because a recent study showed that extinction training conducted immediately after fear conditioning is not LVGCC- dependent (Cain, Godsil, Jami, & Barad, 2005). Cain et al., (2005) suggested that extinction is different depending on the conditioning-extinction interval, in that early extinction is independent of LVGCCs, whereas late extinction requires them. It appears 172

that the relationship between early extinction and depotentiation needs to be delineated more specifically in the adult rat, in order to determine whether those processes are involved in extinction in the developing rat.

Another potential mechanism for erasure is opioid neurotransmission. In adult rats, multiple neurotransmitter systems are involved in extinction (see Myers & Davis,

2007, for review). The major ones include the NMDA, GABA, and endogeneous opioid systems. Out of these neurotransmitter systems, the endogenous opioid system is likely to be involved in extinction at 17 days of age for several reasons. Firstly, we already know that extinction is NMDA- and GABA-independent in P17 rats (Chapter 3;

Langton et al., 2007). Secondly, endogenous opioids regulate learning, memory, and motivation in mammals (see McNally & Akil, 2002, for review), and this neurotransmitter system has been found to regulate extinction in adult rats (e.g.,

Hernandez & Powell, 1980; 1983; McNally, Pigg, & Weidemann, 2005; McNally &

Westbrook, 2003a). For example, McNally and Westbrook (2003a) found that systemic administration of the -opioid receptor antagonist naloxone before extinction training profoundly attenuated acquisition of extinction as measured by freezing. This effect was mediated by central (not peripheral) opioid receptors because injection of naloxone methiodide, which does not readily cross the blood-brain barrier, had no effect on extinction. Additionally, McNally and Westbrook (2003a) noted that endogenous opioids appear to be critical for the acquisition of extinction because naloxone administration immediately after extinction training or before test had no effect, indicating that opioid receptors are not involved in the consolidation or expression of extinction.

Thirdly, the endogenous opioid system may be involved in erasure of conditioned fear during extinction in P17 rats because there is evidence showing that 173

the activation of opioid system disrupts consolidation of memory in adult rats. For example, postconditioning injections of opioid receptor agonists disrupt long-term memory (e.g., McNally & Westbrook, 2003b; Rudy, Kuwagama, & Pugh, 1999). The endogenous opoioids also play an important role in experimentally-induced amnesia.

Specifically, Rudy and his colleagues proposed that amnesia due to post-conditioning social isolation is a failure in consolidation of memory due to opioid release because opioid receptor antagonists prevent the amnesia (Rudy et al., 1999). From this, it could be said that opioid neurotransmission can ‘erase’ memory because it prevents and/or reverses consolidation of conditioned memory.

Therefore, acquisition of extinction in P17 rats may lead to erasure of the original conditioned fear by mainly utilising the opioid system rather than the multiple neurotransmitter systems that are involved in extinction in the adult rat. In fact, the endogenous opioid system is active from birth (Blass, Jackson, & Smotherman, 1991), and has been found to modulate memory in infant rats (Weber et al., 2006).

Furthermore, the endogenous opioid system is particularly active early in development because behaviours specific to infancy, such as suckling and play, have been found to activate the opioid system. For example, intraoral infusions of milk in 10-day-old rats reduce distress vocalisations and increase analgesia to aversive stimulation, effects that are prevented by opioid receptor antagonists (Blass & Fitzgerald, 1988). Play behaviour appears to also activate the opioid system - opioid receptor antagonists reduce social play and opioid agonists increase it (Vanderschuren, Niesink, Spruijt, & Van Ree, 1995;

Vanderschuren, Spruijt, Hol, Niesink, & Van Ree, 1996). From these studies, endogenous opioid activity appears to be higher in infant rats compared to older rats.

Considering the effectiveness of opioids in inducing post-conditioning amnesia, the involvement of a particularly active endogenous opioid system in extinction at P17 rats 174

may lead to erasure of the original conditioned fear. Using an opioid receptor antagonist such as naloxone, it would be interesting to examine whether the opioid system is critical for extinction acquisition in 17-day-old rats.

It should be noted that other neurotransmitter systems such as the adrenergic system and endogenous cannabinoid system are also involved in extinction in the adult rat (see Myers & Davis, 2007, for review). However, the role of adrenergic and endocannabinoid system in extinction appears to depend on GABA neurotransmission

(e.g., Azad et al., 2003; Berlau & McGaugh, 2006; Marsicano et al., 2002). Therefore, those neurotransmitter systems may not be important for extinction in P17 rats because it was demonstrated in the present thesis that alterations in GABAergic activity does not affect in extinction at 17 days of age. Nevertheless, the potential importance of adrenergic, endocannabinoid, and other neurotransmitter systems in extinction in P17 rats cannot be determined without additional research.

Overall, the experiments from the present thesis suggest that the mechanisms underlying extinction are fundamentally distinct in 17-day-old rats compared to those of older rats. It appears that P17 rats may utilise a different neural circuitry for extinction perhaps because of hippocampus and mPFC immaturity. A different neural circuitry for extinction in these rats may result in the extinction memory to be stored differently.

Alternatively, engaging in a different neural circuitry for extinction at 17 days of age may cause extinction to erase the original fear memory. Unfortunately, it is yet difficult to propose any concrete ideas on how erasure may occur because not much is known about the neural processes underlying erasure of memory. Nevertheless, it does appear that erasure of memory is possible (e.g., Shema, Charlton, & Dudai, 2007). Furthermore, research on non-associative learning in the invertebrate Aplysia californica has shown 175

that just as long-term memory requires physiological growth of new synaptic connections, as the memory fades over time these connections retract (e.g., Bailey &

Chen, 1988; 1989; Bailey, Montarolo, Chen, Kandel, & Schacher, 1992). However, virtually nothing is known on how these synaptic connections retract. Further research into erasure of memory appears necessary, especially taking into account the potential significance of erasure as a possible mechanism for extinction early in development.

III. CONCLUDING REMARKS

The developmental approach to the study of memory provides unique opportunities to assess the neurobiological processes underlying how an animal learns, remembers, and forgets. In the present thesis, extinction of conditioned fear was examined in the developing rat. It was demonstrated, for the first time, that 24-day-old rats are essentially adult-like in extinction, in that they showed reinstatement, renewal, and a recovery in an extinguished fear response following pre-test FG7142 injection. Further, re-extinction was amygdala-independent in rats this age, consistent with previous findings with the adult rat. In contrast, it was found that extinction at 17 days of age is fundamentally different from extinction in older rats. Specifically, P17 rats did not show reinstatement, renewal, nor the FG7142 effect. Additionally, re-extinction of conditioned fear was dependent on the amygdala in these rats. These findings were largely interpreted as extinction erasing the original fear memory in the 17-day-old rat, due to differences in the neural structures involved in extinction in these rats compared to older rats.

The novel findings in this thesis do not provide any definitive answers as to how exactly extinction processes are different in infancy compared to adulthood. Rather, this 176

thesis promotes more questions on how maturation of different neural structures during development affects learning, memory, and forgetting. For instance, as well as showing developmental differences in extinction of conditioned fear, the present experiments also show dissociations in spontaneous forgetting and extinction (e.g., spontaneously forgotten fear can be returned following US reminder whereas extinguished fear is unaffected by the US reminder in P17 rats; Chapter 2). These findings highlight that memory and its related phenomena involves complex interactions between many neural structures and different neurotransmitter systems, so that the involvement of one process for one type of memory does not necessarily indicate that the same process is involved in the same way for another type of memory.

More studies on extinction in the developing rat would also be informative in determining how the brain processes aversive and appetitive memories differently. It is worth noting that extinction of an appetitive behaviour in the developing rat may be different to fear extinction because studies by Gonzalez-Lima and his colleagues suggested that the mPFC and hippocampus are involved in extinction of a conditioned appetitive response at 17 days of age (Nair, Berndt, Barrett, & Gonzalez-Lima, 2001;

Nair & Gonzalez-Lima, 1999). To my knowledge, there is only one study that examined renewal and reinstatement following extinction of an appetitive memory in the developing rat. In Carew and Rudy (1991), 17- and 20-day old rats received pairings of a CS and infusion of 2% sucrose solution (delivered directly into the rat’s mouth through a cheek cannula). Learning was measured by the amount of time the rat spent mouthing (defined as opening and closing of the jaws) during CS presentations. When rats were extinguished and tested in different contexts, renewal was observed in P20 rats but not in P17 rats. This finding is consistent with the findings on renewal in the present thesis. Interestingly, Carew and Rudy (1991) showed that rats at both ages 177

exhibited reinstatement when a pre-test US was given, a finding contradictory to the finding on reinstatement in the present thesis. However, it should be noted that in Carew and Rudy (1991), the younger rats were trained for three days starting at P17; and these rats received extinction on day 4 of the experiment (i.e., at 21 days of age). Hence, it is yet unknown whether an appetitive task extinguished at 17 days of age can be reinstated.

Considering that not much is known about extinction of a Pavlovian conditioned appetitive response, investigating extinction in the developing rat across aversive and appetitive tasks would be a fruitful area for future research.

Finally, the present findings suggesting that distinct processes may mediate extinction at different stages of development have significant clinical implications. It has been proposed that fear acquired early in development has enduring effects that are resistant to extinction (Jacobs & Nadel 1985; 1999). Longitudinal data also suggest that anxiety disorders are the most prevalent type of mental illness in childhood, and if left untreated, persist into adulthood (Newman et al., 1996). Also, while there is no doubt that the most successful treatments for anxiety disorders are exposure-based therapies, relapse is a common problem in adult populations (Hofmann, 2007; Rachman 1989).

However, the present research suggests that exposure therapy in young children may be more effective compared to the same treatments in adults. In the least, exposure therapy in young children may make them less susceptible to relapse. The most dramatic interpretation of the present research is that fear acquired early in development may be permanently erased if exposure therapy is given at an early age. Considering such clinical significance, future developmental studies on extinction may need to focus on translational research into anxiety disorder patients in pre-adolescent populations. 178

References

Akirav, I., Raizel, H., & Maroun, M. (2006). Enhancement of conditioned fear extinction by infusion of the GABAA agonist muscimol into the rat prefrontal cortex and amygdala. European Journal of Neuroscience, 23, 758-764.

Alberts, J. R., & Gubernick, D. J. (1984). Early learning as ontogenetic adaptation for ingestion by rats. Learning & Motivation, 15, 334-359.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Anagnostaras, S. G., Gale, G. D., & Fanselow, M. S. (2001). Hippocampus and contextual fear conditioning: recent controversies and advances. Hippocampus, 11, 8-17.

Anagnostaras, S. G., Maren, S., & Fanselow, M. S. (1999). Temporally-graded retrograde amnesia of contextual fear after hippocampal damage in rats: Within- subjects examination. Journal of Neuroscience, 19, 1106–1114.

Australian Code of Practice for the Care and Use of Animals for Scientific Purposes (7th ed.) (2004). Canberra: Australian Government Publishing Service.

Azad, S. C., Eder, M., Marsicano, G., Lutz, B., Zieglgansberger, W., & Rammes, G. (2003). Activation of the cannabinoid receptor type 1 decreases glutamatergic and GABAergic synaptic transmission in the lateral amygdala of the mouse. Learning & Memory, 10, 116-128.

Bailey, C. H., & Chen, M. (1988). Long-term memory in Aplysia modulates the total number of varicosities of single identified sensory neurons. The Proceedings of the National Academy of Sciences, 85, 2373-2377.

Bailey, C. H., & Chen, M. (1989). Time course of structural changes at identified sensory neuron synapses during long-term sensitization in Aplysia. Journal of Neuroscience, 9, 1774-1780.

Bailey, C. H., Montarolo, P., Chen, M., Kandel, E. R., & Schacher, S. (1992). Inhibitors of protein and RNA synthesis block structural changes that accompany long-term heterosynaptic plasticity in Aplysia. Neuron, 9, 749-58.

Bailey, D. J., Kim, J. J., Sun, W., Thompson, R. F., & Helmstetter, F. J. (1999). Acquisition of fear conditioning in rats requires the synthesis of mRNA in the amygdala. Behavioral Neuroscience, 113, 276-282.

Baker, J. D., & Azorlosa, J. L. (1996) The NMDA antagonist MK-801 blocks the extinction of Pavlovian fear conditioning. Behavioral Neuroscience, 110, 618-620.

Barad, M., Gean, P. W., & Lutz, B. (2006). The role of the amygdala in the extinction of conditioned fear. Biological Psychiatry, 60, 322-328. 179

Bashir, Z. I., & Collingridge, G. L. (1994). An investigation of depotentiation of longterm potentiation in the CA1 region of the hippocampus. Experimental Brain Research, 100, 437– 443.

Baudry, M., Arst, D., Oliver, M. & Lynch, G. (1981). Development of glutamate binding sites and their regulation by calcium in rat hippocampus, Developmental Brain Research, 1, 37-48.

Bauer, E. P., Schafe, G. E., & LeDoux, J. E. (2002). NMDA receptors and L-type voltage-gated calcium channels contribute to long-term potentiation and different components of fear memory formation in the lateral amygdala. Journal of Neuroscience, 22, 5239–5249.

Berlau, D. J., & McGaugh, J. L. (2006). Enhancement of extinction memory consolidation: The role of the noradrenergic and GABAergic systems within the basolateral amygdala. Neurobiology of Learning & Memory, 86, 123-132.

Berretta, S., Pantazopoulos, H., Caldera, M., Pantazopoulos, P., & Pare, D. (2005). Infralimbic cortex activation increases c-Fos expression in intercalated neurons of the amygdala. Neuroscience, 132, 943-953.

Blanchard, R. J., & Blanchard, D. C. (1969). Passive and active reactions to fear- eliciting stimuli. Journal of Comparative and Physiological Psychology, 68, 129- 135.

Blass, E. M., & Fitzgerald, E. (1988). Milk-induced analgesia and comforting in 10- day-old rats: Opioid mediation. Pharmacology Biochemistry & Behavior, 29, 9– 13.

Blass, E. M., Jackson, A. M., & Smotherman, W. P. (1991). Milk-induced, opioid- mediated antinociception in rats at the time of cesarean delivery. Behavioral Neuroscience, 105, 677–686.

Bliss, T. V., & Collingridge, G. L. (1993). A synaptic model of memory: Long-term potentiation in the hippocampus. Nature, 361, 31-39.

Boulis, N. M., & Davis, M. (1989) Footshock-induced sensitization of electrically elicited startle reflexes. Behavioral Neuroscience, 103, 504-508.

Bouton, M. E. (1984). Differential control by context in the inflation and reinstatement paradigms. Journal of Experimental Psychology: Animal Behavior Processes, 10, 56-74.

Bouton, M. E. (1991). Context and retrieval in extinction and in other examples of interference in simple associative learning, In L. Dachowski, & C. F. Flaherty (Eds.), Current Topics in Animal Learning: Brain, Emotion, and Cognition (pp. 25–53). Hillsdale, NJ: Erlbaum.

Bouton, M. E. (2002). Context, ambiguity, and unlearning: Sources of relapse after behavioral extinction. Biological Psychiatry, 52, 976-986. 180

Bouton, M. E. & Bolles, R. C. (1979a). Role of conditioned contextual stimuli in reinstatement of extinguished fear. Journal of Experimental Psychology: Animal Behavior Processes, 5, 368-378.

Bouton, M. E., & Bolles, R. C. (1979b) Contextual control of the extinction of conditioned fear. Learning & Motivation, 10, 445-466.

Bouton, M. E., & King, D. A. (1983). Contextual control of the extinction of conditioned fear: Tests for the associative value of the context. Journal of Experimental Psychology: Animal Behavior Processes, 9, 248-265.

Bouton, M. E., & Ricker, S. T. (1994). Renewal of extinguished responding in a second context. Animal Learning & Behavior, 22, 317-324.

Bouton, M. E., & Swartzentruber, D. (1989). Slow reacquisition following extinction: Context, encoding, and retrieval mechanisms. Journal of Experimental Psychology: Animal Behavior Processes, 15, 43-53.

Bouton, M. E., Westbrook, R. F., Corcoran, K. A., & Maren, S. (2006). Contextual and temporal modulation of extinction: Behavioral and biological mechanisms. Biological Psychiatry, 60, 352-360.

Burgos-Robles, A., Vidal-Gonzalez, I., Santini, E., & Quirk, G. J. (2007). Consolidation of fear extinction requires NMDA receptor-dependent bursting in the ventromedial prefrontal cortex. Neuron, 53, 871-880.

Cain, C. K., Blouin, A. M., & Barad, M. (2002). L -type voltage-gated calcium channels are required for extinction, but not for acquisition or expression, of conditional fear in mice. Journal of Neuroscience, 22, 9113-9121.

Cain, C. K., Godsil, B. P., Jami, S., & Barad, M. (2005). The L-type calcium channel blocker nifedipine impairs extinction, but not reduced contingency effects, in mice. Learning & Memory, 12, 277-284.

Calandreau, L., Desmedt, A., Decorte, L., & Jaffard, R. (2005). A different recruitment of the lateral and basolateral amygdala promotes contextual or elemental conditioned association in Pavlovian fear conditioning. Learning & Memory, 12, 383-388.

Cammarota, M., Bevilaqua, L. R., Medina, J. H., & Izquierdo, I. (2004). Retrieval does not induce reconsolidation of inhibitory avoidance memory. Learning & Memory, 11, 572-578.

Campbell, B. A., & Ampuero, M. X. (1985). Dissociation of autonomic and behavioral components of conditioned fear during development in the rat. Behavioral Neuroscience, 99, 1089-1102.

Campbell, B. A., & Campbell, E. H. (1962). Retention and extinction of learned fear in infant and adult rats. Journal of Comparative & Physiological Psychology, 55, 1-8. 181

Campbell, B. A., & Jaynes, J. (1966). Reinstatement. Psychological Review, 73, 478- 480.

Campbell, B. A., & Spear, N. E. (1972). Ontogeny of memory. Psychological Review, 79, 215-236.

Campeau, S., Hayward, M. D., Hope, B. T., Rosen, J. B., Nestler, E. J., & Davis, M. (1991). Induction of the c-fos proto-oncogene in rat amygdala during unconditioned and conditioned fear. Brain Research, 565, 349-352.

Carew, M. B. & Rudy, J. W. (1991). Multiple functions of context during conditioning: A developmental analysis. Developmental Psychobiology, 24, 191-209.

Carr, D.B. & Sesack, S.R. (1996). Hippocampal afferents to the rat prefrontal cortex: Synaptic targets and relation to dopamine terminals. Journal of Comparative Neurology, 369, 1–5.

Cartwright-Hatton, S., McNicol, K., & Doubleday, E. (2006). Anxiety in a neglected population: Prevalence of anxiety disorders in pre-adolescent children. Clinical Psychology Review, 26, 817-833.

Cartwright-Hatton, S., Roberts, C., Chitsabesan, P., Fothergill, C., & Harrington, R. (2004). Systematic review of the efficacy of cognitive behaviour therapies for childhood and adolescent anxiety disorders. British Journal of Clinical Psychology, 43, 421-436.

Catterall, W. A., & Mackie, K. (1996). Local anaesthetics. In J. G. Hardman, L. E. Limbird, P. B. Molinoff, R. W. Ruddon, & A. G. Gilman (Eds.), Goodman and Gilman’s the pharmacological basis of therapeutics (pp. 331–347). New York: McGraw-Hill.

Conde, F., Maire Lepoivre, E., Audinat, E., & Crepel, F. (1995). Afferent connections of the medial frontal cortex of the rat. II. Cortical and subcortical afferents. Journal of Comparative Neurology, 352, 567–593.

Corcoran, K. A., Desmond, T. J., Frey, K. A., & Maren, S. (2005). Hippocampal inactivation disrupts the acquisition and contextual encoding of fear extinction. Journal of Neuroscience, 25, 8978-8987.

Corcoran, K. A., & Maren, S. (2001). Hippocampal inactivation disrupts contxtual retrieval of fear memory after extinction. The Journal of Neuroscience, 21, 1720– 1726.

Corcoran, K.A., & Maren, S. (2004). Factors regulating the effects of hippocampal inactivation on renewal of conditional fear after extinction. Learning & Memory, 11, 598 – 603.

Corcoran, K. A., & Quirk, G. J. (2007). Activity in prelimbic cortex is necessary for the expression of learned, but not innate, fears. Journal of Neuroscience, 27, 840-844. 182

Costa, E, Guidotti, A., & Mao, C. C. (1975). Evidence for involvement of GABA in the action of benzodiazepine: studies on rat cerebellum. Advances in Biochemical Psychopharmacology, 14, 113-130.

Crain, B., Cotman, C., Taylor, D. & Lynch, G. (1973). A quantitative electron microscopic study of synaptogenesis in the dentate gyrus of the rat, Brain Research, 63, 195-204.

Davis, M. (1979) Diazepam and flurazepam: effects on conditioned fear as measured with the potentiated startle paradigm. Psychopharmacology, 62, 1-7.

Davis, M. (1992). The role of the amygdala in fear and anxiety. Annual Review of Neuroscience, 15, 353-375.

Davis, M. (2000). The role of the amygdala in conditioned and unconditioned fear and anxiety. In J. P. Aggleton (Ed.), The Amygdala, vol. 2, (pp. 213–287). Oxford, UK: Oxford University Press.

Davis, M. (2006). Neural systems involved in fear and anxiety measured with fear- potentiated startle. American Psychologist, 61, 741-756.

Davis, M., Falls, W. A., Campeau, S., & Kim, M. (1993). Fear-potentiated startle: A neural and pharmacological analysis. Behavioural Brain Research, 58, 175-98.

Davis, M., & Myers, K. (2002). The role of glutamate and gamma-aminobutyric acid in fear extinction: clinical implications for exposure therapy. Biological Psychiatry, 52, 998-1007.

Davis, M., & Whalen, P. J. (2001). The amygdala: vigilance and emotion. Molecular Psychiatry, 6, 13-34.

Delamater, A. R. (1996). Effects of several extinction treatments upon the integrity of Pavlovian stimulus–outcome associations. Animal Learning & Behavior, 24, 437– 449.

Delamater, A. R. (2004). Experimental extinction in Pavlovian conditioning: Behavioral and neuroscience perspectives. The Quarterly Journal of Experimental Psychology, 57B, 97-132.

Falls, W. A., Miserendino, M. J., & Davis, M. (1992). Extinction of fear-potentiated startle: blockade by infusion of an NMDA antagonist into the amygdala. Journal of Neuroscience, 12, 854–863.

Fanselow, M. S., & Kim, J. J. (1992). The benzodiazepine inverse agonist DMCM as an unconditioned stimulus for fear-induced analgesia: implications for the role of GABAA receptors in fear-related behavior. Behavioral Neuroscience, 106, 336- 344. 183

Fanselow, M. S., & Rudy, J. W. (1998). Convergence of experimental and developmental approaches to animal learning and memory processes. In T. J. Carew, R. Menzel, & C. J. Shatz (Eds.), Mechanistic relationship between development and learning: Reports of the Dahlem workshop on mechanistic relationship between development and learning, Berlin, January 19-25, 1997 (pp.15 - 28). New York: John Wiley & Sons Ltd.

Farb, C., Aoki, C., Milner, T., Kaneko, T., & LeDoux, J. E. (1992). Glutamate immunoreactive terminals in the lateral amygdaloid nucleus: A possible substrate for emotional memory. Brain Research, 593, 145-158.

Farinelli, M., Deschaux, O., Hugues, S., Thevenet, A., & Garcia R. (2006). Hippocampal train stimulation modulates recall of fear extinction independently of prefrontal cortex synaptic plasticity and lesions. Learning & Memory, 13, 329- 34.

Feigley, D. A., & Spear, N. E. (1970). Effect of age and punishment condition on long- term retention by the rat of active- and passive-avoidance learning. Journal of Comparative & Physiological Psychology, 73, 515-526.

Freud, S. (1920). A general introduction to psychoanalysis. New York, NY: Horace Liveright.

Freud, S., & Breuer, (1893). On the psychical mechanism of hysterical phenomena: Preliminary communications. In J. Strachey & A. Strachey (Eds.), Studies in hysteria, 1974 (pp. 53-69). Harmonsworth Penguin.

Frohardt, R. J., Guarraci, F. A., & Bouton, M. E. (2000). The effects of neurotoxic hippocampal lesions on two effects of context after fear extinction. Behavioral Neuroscience, 114, 227-240

Frysztak, R. J., & Neafsey, E. J. (1994). The effect of medial frontal cortex lesions on cardiovascular conditioned emotional responses in the rat. Brain Research, 643, 181-193.

Gale, G. D., Anagnostaras, S. G., Godsil, B. P., Mitchell, S., Nozawa, T., Sage, J. R., Wiltgen, B., & Fanselow, M. S. (2004). Role of basolateral amygdala in the storage of fear memories across the adult lifetime of rats. Journal of Neuroscience, 24, 3810-3815.

Garcia, R. (2002). Postextinction of conditioned fear: Between two CS-related memories. Learning & Memory, 9, 361–363.

Garcia, R., Chang, C., & Maren, S. (2006). Electrolytic lesions of the medial prefrontal cortex do not interfere with long-term memory of extinction of conditioned fear. Learning & Memory, 13, 14-17.

Gewirtz, J. C., Falls, W. A., & Davis, M. (1997). Normal conditioned inhibition and extinction of freezing and fear-potentiated startle following electrolytic lesions of medial prefrontal cortex in rats. Behavioral Neuroscience, 111, 712-726. 184

Good, M., & Honey, R. C. (1991). Conditioning and contextual retrieval in hippocampal rats. Behavioral Neuroscience, 105, 499-509.

Gottlieb, G., & Klopfer, P. H. (1962). The relation of developmental age to auditory and visual imprinting. Journal of Comparative & Physiological Psychology, 55, 821- 826. Grillon, C., Baas, J. P., Lissek, S., Smith, K., & Milstein, J. (2004). Anxious Responses to Predictable and Unpredictable Aversive Events. Behavioral Neuroscience, 118, 916-924.

Haefely, W. E. (1990). The GABAA-benzodiazepine receptor: biology and pharmacology. In G. D. Burrows, M. Roth, & R. Noyes. Jr. (Eds.), Handbook of Anxiety, vol. 3: the neurobiology of anxiety (pp.165-188). Elsevier: New York.

Haefely, W., Kyburz, E., Gerecke, M., & Mohler, H., (1985). Recent advances in the molecular pharmacology of benzodiazepines receptors and in the structure-activity relationship of their agonists and antagonists. Advances in Drug Research, 14, 165- 322.

Haralambous, T., & Westbrook, R. F. (1999). An infusion of bupivacaine into the nucleus accumbens disrupts the acquisition but not the expression of contextual fear conditioning. Behavioral Neuroscience, 113, 925-940.

Harlow, H. F. (1959). The development of learning in the rhesus monkey. American Scientist, 47, 459-479.

Harris, J. A., & Westbrook, R. F. (1994). Effects of midazolam and naloxone in rats tested for sensitivity/reactivity to formalin pain in a familiar, novel or aversively conditioned environment. Psychopharmacology. 115, 65-72.

Harris, J. A., & Westbrook, R. F. (1995). Effects of benzodiazepine microinjection into the amygdala or periaqueductal gray on the expression of conditioned fear and hypoalgesia in rats. Behavioral Neuroscience, 109, 295-304.

Harris, J. A., & Westbrook, R. F. (1996). Midazolam impairs the acquisition of conditioned analgesia if rats are tested with an acute but not a chronic noxious stimulus. Brain Research Bulletin, 39, 227-233.

Harris, J. A., & Westbrook, R. F. (1998a). Benzodiazepine-induced amnesia in rats: Reinstatement of conditioned performance by noxious stimulation on test. Behavioral Neuroscience, 112, 183-192.

Harris, J. A., & Westbrook, R. F. (1998b). Evidence that GABA transmission mediates context-specific extinction of learned fear. Psychopharmacology, 140, 105–115.

Henri, V., & Henri, C. (1895). On our earliest recollections of childhood. Psychological Review, 2, 215-216. 185

Hernandez, L. L., & Powell, D. A. (1980). Effects of naloxone on Pavlovian conditioning of eyeblink and heart rate responses in rabbits. Life Sciences, 27, 863–869.

Hernandez, L. L., & Powell, D. A. (1983). Naloxone induces multiple effects on aversive Pavlovian conditioning. Behavioral Neuroscience, 97, 478–491.

Herry, C., & Garcia, R. (2002). Prefrontal cortex long-term potentiation, but not long- term depression, is associated with the maintenance of extinction of learned fear in mice. Journal of Neuroscience, 22, 577-83.

Herry, C. & Garcia, R. (2003). Behavioral and paired-pulse facilitation analyses of long-lasting depression at excitatory synapses in the medial prefrontal cortex in mice. Behavioral Brain Research, 146, 89–96.

Herry, C., Trifilieff, P., Micheau, J., Luthi, A., & Mons, N. (2006). Extinction of auditory fear conditioning requires MAPK/ERK activation in the basolateral amygdala. European Journal of Neuroscience, 24, 261-269.

Herry, C., Vouimba, R. M., & Garcia, R. (1999). Plasticity in the mediodorsal thalamo- prefrontal cortical transmission in behaving mice. Journal of Neurophysiology, 82, 2827–2832.

Hijzen, T. H. & Slangen, J. L. (1989). Effects of midazolam, DMCM and lindane on potentiated startle in the rat. Psychopharmacology, 99, 362-365.

Hirsh, R. (1980). The hippocampus, conditional operations, and cognition. Physiological Psychology, 8, 175-182.

Hobin, J. A., Goosens, K. A., & Maren, S. (2003). Context-dependent neuronal activity in the lateral amygdala represents fear memories after extinction. The Journal of Neuroscience, 23, 8410–8416.

Hofer, M. (1991). Microsurgery: Window on hidden developmental processes. In H. Shair, G. Barr, & M. Hofer (Eds.), Developmental psychobiology: New methods and changing concepts (pp. 19-31). New York: Oxford University Press.

Hofmann, S. G. (2007). Enhancing exposure-based therapy from a translational research perspective. Behavioral Research Therapy, 45, 1987-2001.

Holt, W., & Maren, S. (1999). Muscimol inactivation of the dorsal hippocampus impairs contextual retrieval of fear memory. Journal of Neuroscience, 19, 9054– 9062.

Honey, R. C., & Good, M. (1993). Selective hippocampal lesions abolish the contextual specificity of latent inhibition and conditioning. Behavioral Neuroscience, 107, 23-33. 186

Hugues, S., Chessel, A., Lena, I., Marsault, R., & Garcia, R. (2006). Prefrontal infusion of PD098059 immediately after fear extinction training blocks extinction- associated prefrontal synaptic plasticity and decreases prefrontal ERK2 phosphorylation. Synapse, 60, 280-287.

Hugues, S., Deschaux, O., & Garcia, R. (2004). Postextinction infusion of a mitogen- activated protein kinase inhibitor into the medial prefrontal cortex impairs memory of the extinction of conditioned fear. Learning & Memory, 11, 540-543.

Hunt, P. & Campbell, B. A. (1997). Developmental dissociation of the components of conditioned fear. In M. E. Bouton & M. S. Fanselow (Eds.), Learning, motivation and cognition: The functional behaviorism of Robert C. Bolles (pp. 53-74). Washington, D.C.: American Psychological Association. Hunt, P.S., Hess, M.F., & Campbell, B.A. (1998). Inhibition of the expression of conditioned cardiac responses in the developing rat. Developmental Psychobiology, 33, 221-233.

Hyson, R. L., & Rudy, J. W. (1984). Ontogenesis of learning: 11. Variations in rats reflexive and learning responses to acoustic stimulation. Developmental Psychobiology, 17, 263-283.

Ishikawa, A., & Nakamura, S. (2003). Convergence and interaction of hippocampal and amygdalar projections within the prefrontal cortex in the rat. The Journal of Neuroscience, 23, 9987–9995.

Jacobs, W. J., & Nadel, L. (1985). Stress-induced recovery of fears and phobias. Psychological Review, 92, 512-531.

Jacobs, W. J., & Nadel, L. (1999). The first panic attack: A neurobiological theory. Canadian Journal of Experimental Psychology, 53, 92-107.

Jacobson, S. (1963). Sequence of myelinization in the brain of the albino rat: Cerebral cortex, thalamus and related structures. Journal of Comparative Neurology, 121, 5-29.

Ji, J. & Maren, S. (2005). Electrolytic lesions of the dorsal hippocampus disrupt renewal of conditional fear after extinction. Learning & Memory, 12, 270 – 276.

Kamprath, K., & Wotjak, C. T. (2004). Nonassociative learning processes determine expression and extinction of conditioned fear in mice. Learning & Memory, 11, 770-786.

Kandel, E. R. (2001). The molecular biology of memory storage: a dialogue between genes and synapses. Science, 294, 1030-1038.

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM–IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593–602. 187

Kessler, R.C., Chiu, W. T., Demler, O. & Walters, E. E. (2005). Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 617-627.

Kim, J. H., McNally, G. P., & Richardson, R. (2006). Recovery of fear memories in rats: role of gamma-amino butyric acid (GABA) in infantile amnesia. Behavioral Neuroscience, 120, 40-48.

Kim, J. J., & Fanselow, M. S. (1992). Modality-specific retrograde amnesia of fear. Science, 256, 675–677.

Kolb, B. (1984). Functions of the frontal cortex of the rat: A comparative review. Brain Research Reviews, 8, 65-98.

Kolb, B., & Nonneman, A. J. (1978). Sparing of function in rats with early prefrontal cortex lesions. Brain Research, 151, 135-148.

Konorski, J. (1948). Conditioned Reflexes and Neuron Organization. Cambridge: University Press.

Kucharski, D., & Spear, N. E. (1984). Conditioning of aversion to an odor paired with peripheral shock in the developing rat. Developmental Psychobiology, 17, 465- 479.

Langton, J. L., Kim, J. H., Nicholas, J., & Richardson, R. (2007). The effect of NMDA- receptor antagonist MK801 on the acquisition and extinction of learned fear in the developing rat. Learning & Memory, 14, 665-668.

Lattal, K. M., Radulovic, J., & Lukowiak, K. (2006). Extinction: does it or doesn’t it? The requirement of altered gene activity and new protein synthesis. Biological Psychiatry, 60, 344–351.

Laurent, V., Marchand, A. R., & Westbrook, R. F. (in press). The basolateral amygdala is necessary for learning but not relearning extinction of context conditioned fear. Learning & Memory.

Ledgerwood, L., Richardson, R., & Cranney, J. (2003). Effects of D-cycloserine on extinction of conditioned freezing. Behavioral Neuroscience, 117, 341-349.

LeDoux, J. E. (1993). Emotional memory systems in the brain. Behavioural Brain Research, 58, 69-79.

LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–184.

LeDoux, J. E., Farb, C. F., & Ruggiero, D.A. (1990). Topographic organization of neurons in the acoustic thalamus that project to the amygdala. Journal of Neuroscience, 10, 1043-1054. 188

LeDoux, J. E., & Gorman, J. M. (2001). A call to action: Overcoming anxiety through active coping. American Journal of Psychiatry, 158, 1953-1955.

LeDoux, J. E., Iwata, J., Cicchett, P., & Reis, D. J. (1988). Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. Journal of Neuroscience, 8, 2517-2529.

Lee, H., & Kim, J. J. (1998). Amygdalar NMDA receptors are critical for new fear learning in previously fear-conditioned rats. Journal of Neuroscience, 18, 8444– 8454.

Lin, C. H., Lee, C. C., & Gean P. W. (2003a). Involvement of a calcineurin cascade in amygdala depotentiation and quenching of fear memory. Molecular Pharmacology, 63, 44 –52.

Lin, C. H., Lee, C. C., Huang, Y. C., Wang, S. J. & Gean, P. W. (2005). Activation of roup II metabotropic glutamate receptors induces depotentiation in amygdala slices and reduces fear-potentiated startle in rats. Learning & Memory, 12, 130- 137.

Lin, C. H., Yeh, S. H., Leu T. H., Chang, W. C., Wang, S. T., & Gean, P. W. (2003b). Identification of Calcineurin as a Key Signal in the Extinction of Fear Memory. Journal of Neuroscience, 23, 1574-1579

Lin, C. H., Yeh, S. H., Lu, K. T., & Gean, P. W. (2003c). The Similarities and Diversities of Signal Pathways Leading to Consolidation of Conditioning and Consolidation of Extinction of Fear Memory. Journal of Neuroscience, 23, 8310- 8317

Lin, C. H., Yeh, S. H., Lu, K. T., Leu T. H., Chang, W. C., & Gean, P. W. (2001). A role for the PI-3 kinase signaling pathway in fear conditioning and synaptic plasticity in the amygdala. Neuron, 31, 841– 851.

Lu, K.T., Walker, D. L., & Davis, M. (2001). Mitogen -activated protein kinase cascade in the basolateral nucleus amygdala is involved in extinction of fear-potentiated startle. Journal of Neuroscience, 21, RC162: 1-5.

Lubow, R. E., & De la Casa, L. G. (2005). Time-induced super-latent inhibition is dependent on the distinctiveness of the retention-interval context from the other experimental contexts. Learning & Motivation, 36, 322-330.

Malleret, G., Haditsch, U., Genoux, D., Jones, M. W., Bliss, T. V., Vanhoose, A. M., Weitlauf, C., Kandel, E. R., Winder, D. G., & Mansuy, I. M. (2001). Inducible and reversible enhancement of learning, memory, and long-term potentiation by genetic inhibition of calcineurin. Cell, 104, 675–686. 189

Maren, S., Ferrario, C. R., Corcoran, K. A., Desmond, T. J., & Frey, K. A. Protein synthesis in the amygdala, but not the auditory thalamus, is required for consolidation of Pavlovian fear conditioning in rats. European Journal of Neuroscience, 18, 3080-3088.

Maren, S., & Holt, W. (2000). The hippocampus and contextual memory retrieval in Pavlovian conditioning. Behavioral Brain Research, 110, 97–108.

Marsicano, G., Wotjak, C. T., Azad, S. C., Bisogno, T., Rammes, G., Cascio, M. G., Hermann, H., Tang, J., Hofmann, C., Zieglgaansberger, W., Di Marzo, V., & Lutz, B. (2002). The endogenous cannabinoid system controls extinction of aversive memories. Nature, 418, 530-534.

Martin, L. T., & Alberts, J. R. (1982). Associative learning in neonatal rats revealed by cardiac response patterns. Journal of Comparative & Physiological Psychology, 96, 668-675.

McDonald, A. J. (1985). Immunohistochemical identification of -aminobutyric acid- containing neurons in the rat basolateral amygdala. Neuroscience Letters, 53, 203- 207.

McDonald, A. J., Mascagni, F., & Guo, L. (1996). Projections of the medial and lateral prefrontal cortices to the amygdala: a Phaseolus vulgaris leucoagglutinin study in the rat. Neuroscience, 71, 55-75.

McNally, G. P., & Akil, H. (2002). Opioid peptides and their receptors: Overview and function in pain modulation. In K. Davis, D. Charney, J. T. Coyle, & C. Nemeroff (Eds.), Neuropsychopharmacology: A fifth generation of progress (pp. 35–46). New York: Williams & Wilkins

McNally, G. P., Lee, B., Chiem, J. Y., & Choi, E. A. (2005). The midbrain periaqueductal gray and fear extinction. Opioid receptor subtype and roles of cAMP, protein kinase A, and mitogen-activated protein kinase. Behavioral Neuroscience, 119, 1023-1033.

McNally, G. P., Pigg, M., & Weidemann, G. (2004). Opioid receptors in the midbrain periaqueductal gray regulate extinction of Pavlovian fear conditioning. Journal of Neuroscience, 24, 6912–6919.

McNally, G. P., & Westbrook, R. F. (2003a). Opioid receptors regulate the extinction of Pavlovian fear conditioning. Behavioral Neuroscience, 117, 1292–1301.

McNally, G. P., & Westbrook, R. F. (2003b). Temporally-graded, context-specific retrograde amnesia and its alleviation by context preexposure: Effects of postconditioning exposures to morphine in the rat. Journal of Experimental Psychology: Animal Behavior Processes, 29, 130–142.

Milad, M. R., & Quirk, G. J. (2002). Neurons in medial prefrontal cortex signal memory for fear extinction. Nature, 420, 70–74. 190

Milad, M. R., Vidal-Gonzalez, I., & Quirk, G. J. (2004). Electrical stimulation of medial prefrontal cortex reduces conditioned fear in a temporally specific manner. Behavioral Neuroscience, 118, 389–394.

Millhouse, O. E. (1986). The intercalated cells of the amygdala. Journal of Comparative Neurology, 247, 246-271.

Mineka, S. & Zinbarg, R. (2006). A contemporary learning theory perspective on the etiology of anxiety disorders. American Psychologist, 61, 10-26.

Morgan, M. A., & LeDoux, J. E. (1995). Differential contribution of dorsal and ventral medial prefrontal cortex to the acquisition and extinction of conditioned fear in rats. Behavioral Neuroscience, 109, 681-688.

Morgan, M. A., Romanski, L. M., & LeDoux, J. E. (1993). Extinction of emotional learning: contribution of medial prefrontal cortex. Neuroscience Letters, 163, 109–113.

Morgan, M. A., Schulkin, J., LeDoux, J. E. (2003). Ventral medial prefrontal cortex and emotional perseveration: the memory for prior extinction training. Behavioral Brain Research, 146, 121–130.

Myers, K. M., & Davis, M. (2007). Mechanisms of fear extinction. Molecular Psychiatry, 12, 120–150.

Myers, K. M., Ressler, K. J., & Davis, M. (2006). Different mechanisms of fear extinction dependent on length of time since fear acquisition. Learning & Memory, 13, 216-223.

Nair, H. P., Berndt, J. D., Barrett, D., & Gonzalez-Lima, F. (2001). Maturation of extinction behavior in infant rats: Large-scale regional interactions with medial prefrontal cortex, orbitofrontal cortex, and anterior cingulate cortex. Journal of Neuroscience, 21, 4400-4407.

Nair, H. P., & Gonzalez-Lima, F. (199). Extinction of behavior in infant rats: Development of functional coupling between septal, hippocampal, and ventral tegmental regions. Journal of Neuroscience, 19, 8646-8655.

Newman, D. L., Moffitt, T. E., Caspi, A., Magdol, L., Silva, P., & Stanton, W. R. (1996). Psychiatric disorder in a birth cohort of young adults: prevalence, comorbidity, clinical significance, and new case incidence from ages 11 to 21. Journal of consulting and Clinical Psychology, 64, 552-562.

Olsen, R. W. (2002) GABA. In K. L. Davis, D. Charney, J. T. Coyle, & C. Nemeroff (Eds.), Neuropsychopharmacology: The fifth generation of progress (pp. 159-168). American College of Neuropsychopharmacology.

Pare, D., Quirk, G. J., & LeDoux, J. E. (2004). New vistas on amygdala networks in conditioned fear. Journal of Neurophysiology, 92, 1–9. 191

Parsons, P. J., Fagan, T., & Spear, N. E. (1973). Short-term retention of habituation in the rat: A developmental study from infancy to old age. Journal of Comparative and Physiological Psychology, 84, 545-553.

Pavlov, I. P. (1927). Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex. London: Oxford University Press.

Pavlov, I. P. (1932). The reply of a physiologist to psychologists. Psychological Review, 39, 91-127.

Paxinos, G. & Watson, C. (1998). The rat brain in stereotaxic coordinates. Academic Press, New York.

Phelps, E. A., & LeDoux, J. E. (2005). Contributions of the amygdala to emotion processing: From animal models to human behaviour. Neuron, 48, 175–187.

Phillips, R. G., & LeDoux, J. E. (1992). Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behavioral Neuroscience, 106, 274–285.

Pitkanen, A., Savander, V., & LeDoux, J. E. (1997). Organization of intra-amygdaloid circuitries in the rat: An emerging framework for understanding functions of the amygdala. Trends in Neurosciences, 20, 517-523.

Price, J. L. (2003). Comparative aspects of amygdala connectivity. Annals of the New York Academy of Sciences, 985, 50-58.

Pugh, C. R., & Rudy, J. W. (1996). A developmental analysis of contextual fear conditioning. Developmental Psychobiology, 29, 87-100.

Quirk, G. J. (2002) Memory for extinction of conditioned fear is long-lasting and persists following spontaneous recovery. Learning & Memory, 9, 402-407.

Quirk, G. J., Armony, J. L., & LeDoux, J. E. (1997). Fear conditioning enhances different temporal components of tone-evoked spike trains in auditory cortex and lateral amygdala. Neuron, 19, 613-624.

Quirk, G. J., Garcia, R., & Gonzalez-Lima, F. (2006). Prefrontal mechanisms in extinction of conditioned fear. Biological Psychiatry, 60, 337–343.

Quirk, G. J., Likhtik, E., Pelletier, J. G., & Pare, D. (2003). Stimulation of medial prefrontal cortex decreases the responsiveness of central amygdala output neurons. Journal of Neuroscience, 23, 8800-8807.

Quirk, G. J., & Mueller, D. (2008). Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology, 33, 56-72.

Quirk, G. J., Repa C. J., & LeDoux, J. E. (1995). Fear conditioning enhances short- latency auditory responses of lateral amygdala neurons: parallel recordings in the freely behaving rat. Neuron, 15, 1029-1039. 192

Quirk, G. J., Russo, G. K., Barron, J. L., & Lebron, K. (2000). The role of ventromedial prefrontal cortex in the recovery of extinguished fear. Journal of Neuroscience, 20, 6225-6231.

Rachman, S. (1989). The return of fear: Review and prospect. Clinical Psychology Review, 9, 147-168.

Reberg, D. (1972). Compound tests for excitation in early acquisition and after prolonged extinction of conditioned suppression. Learning & Motivation, 3, 246- 258.

Repa, J. C., Muller, J., Apergis, J., Desrochers, T. M., Zhou, Y., & LeDoux, J. E. (2001). Two different lateral amygdala cell populations contribute to the initiation and storage of memory. Nature Neuroscience, 4, 724-731.

Rescorla, R. A. (2003). Contemporary study of Pavlovian conditioning. The Spanish Journal of Psychology, 6, 185-195.

Rescorla, R. A. (2006). Deepened extinction from compound stimulus presentation. Journal of Experimental Psychology: Animal Behavior Processes, 32, 135-144.

Rescorla, R. A., & Heth, C. D. (1975): Reinstatement of fear to an extinguished conditioned stimulus. Journal of Experimental Psychology: Animal Behavior Processes, 104, 88 –96.

Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and non-reinforcement. In A. H. Prokasy (Ed.), Classical Conditioning II: Current Research and Theory (pp. 64 - 99). New York: Appleton-Century-Croft.

Ressler, K., & Davis, M. (2003). Genetics of childhood disorders: L. Learning and memory, Part 3: Fear conditioning. Journal of American Academy of Child and Adolescent Psychiatry, 42, 612-615.

Rhodes, S. E. V., & Killcross, S. (2004). Lesions of rat infralimbic cortex enhance recovery and reinstatement of an appetitive Pavlovian response. Learning & Memory, 11, 611-616.

Richardson, R., Duffield, T. Q., Bailey, G. K., & Westbrook, R. F. (1999). Reinstatement of fear to an extinguished conditioned context. Animal Learning & Behavior, 27, 399-415.

Richardson, R. & Fan, M. (2002). Behavioral expression of learned fear in rats is appropriate to their age at training, not their age at testing. Animal Learning & Behavior, 30, 394 – 404.

Richardson, R., Paxinos, G., & Lee, J. (2000). The ontogeny of conditioned odor- potentiation of startle. Behavioral Neuroscience, 114, 1167 – 1173. 193

Richardson, R., Riccio, D. C., & Devine, L. (1984). ACTH-induced recovery of extinguished avoidance responding. Physiological Psychology, 12, 184-192.

Richardson, R., Tronson, N., Bailey, G., & Parnas, A.S. (2002). Extinction of conditioned odor potentiation of startle. Neurobiology of Learning & Memory, 78, 426–440.

Roesler, R., Vianna, M., Sant’Anna, M. K., Kuyven, C. R., Kruel, A. V., Quevedo, J., & Ferreira, M. B. C. (1998). Intrahippocampal infusion of the NMDA receptor antagonist AP5 impairs retention of an inhibitory avoidance task: protection from impairment by pretraining or preexposure to the task apparatus. Neurobiology of Learning & Memory, 69, 87-91.

Rosen, J. B., Hitchcock, J. M., Sananes, C. B., Miserendino, M. J., & Davis, M. (1991). A direct projection from the central nucleus of the amygdala to the acoustic startle pathway: Anterograde and retrograde tracing studies. Behavioural Neuroscience, 105, 817-25.

Rosen, J. B., & Schulkin, J. (1998). From normal fear to pathological anxiety. Psychological Review, 105, 325-350.

Royer, S., & Pare, D. (2002). Bidirectional synaptic plasticity in intercalated amygdala neurons and the extinction of conditioned fear responses. Neuroscience, 115, 455– 462.

Royer, S., & Pare, D. (2003). Conservation of total synaptic weight through balanced synaptic depression and potentiation. Nature, 422, 518-522.

Rudy, J. W. (1993). Contextual conditioning and auditory cue conditioning dissociate during development. Behavioral Neuroscience, 197, 887-891.

Rudy, J. W., & Cheatle, M. D. (1979). Ontogeny of associative learning: Acquisition of odor aversions by neonatal rats. In N. E. Spear & B. A. Campbell (Eds.), Ontogeny of learning and memory (pp. 157–188). Mahwah, NJ: Erlbaum.

Rudy, J. W., Kuwagama, K., & Pugh, C. R. (1999). Isolation reduces contextual but not auditory-cue fear conditioning: A role for endogenous opioids. Behavioral Neuroscience, 113, 316–323.

Rudy, J. W., & Morledge, P. (1994). Ontogeny of contextual fear conditioning in rats: implications for consolidation, infantile amnesia, and hippocampal system function. Behavioral Neuroscience, 108, 227-234.

Sah, P., Faber, E. S., Lopez De Armentia, M., & Power, J. (2003). The amygdaloid complex: Anatomy and physiology. Physiological Review, 83, 803–834.

Sanders, M. J., & Fanselow, M. S. (2003). Pre-training prevents context fear conditioning deficits produced by hippocampal NMDA receptor blockade. Neurobiology of Learning & Memory, 80, 123-129. 194

Sanders, S. K., & Shekhar, A., (1991). Blockade of GABAA receptors in the region of the anterior basolateal amygdala of rats elicits increases in heart rate and blood pressure. Brain Research, 576, 101-110.

Santini, E., Ge, H., Ren, K., De Ortiz, S. P., & Quirk, G. J. (2004). Consolidation of Fear Extinction Requires Protein Synthesis in the Medial Prefrontal Cortex. Journal of Neuroscience, 24, 5704-5710.

Santini, E., Muller, R. U., & Quirk, G. J. (2001). Consolidation of extinction learning involves transfer from NMDA-independent to NMDA-dependent memory. Journal of Neuroscience, 21, 9009-9017.

Schafe, G. E., Nader, K., Blair, H. T., & LeDoux, J. E. (2001). Memory consolidation of Pavlovian fear conditioning: A cellular and molecular perspective. Trends in Neurosciences, 24, 540-546.

Schlessinger, A., Cowan, W. & Gottlieb, D. (1975). An autoradiographic study of the time of origin and the pattern of granule cell migration in the dentate gyrus of the rat. Journal of Comparative Neurology, 159, 149-176.

Schulenburg, C. J., Riccio, D. C., & Stikes, E. R. (1971). Acquisition and retention of a passive-avoidance response as a function of age in rats. Journal of Comparative and Physiological Psychology, 74, 75-83.

Sesack, S. R., Deutch, A. Y., Roth, R. H., & Bunney, B. S. (1989). Topographical organization of the efferent projections of the medial prefrontal cortex in the rat: An anterograde tract-tracing study with Phaseolus vulgaris leucoagglutinin. Journal of Comparative Neurology, 290, 213-242.

Shema, R., Sacktor, T. C., & Dudai, Y. (2007). Rapid erasure of long-term memory associations in the cortex by an inhibitor of PKM? Science, 317, 951-953.

Shi, C., & Davis, M. (2001). Visual pathways involved in fear conditioning measured with fear-potentiated startle: Behavioral and anatomic studies. Journal of Neuroscience, 21, 9844–9855.

Silva, R. R., Gallagher, R., & Minami, H. (2006). Cognitive-behavioral treatments for anxiety disorders in children and adolescents. Primary Psychiatry, 13, 68-76.

Smith, Y., Pare, J. F., & Pare, D. (2000). Differential innervation of parvalbumin- immunoreactive interneurons of the basolateral amygdaloid complex by cortical and intrinsic inputs. Journal of Comparative Neurology, 416, 496-508.

Sotres-Bayon, F., Bush, D.E., & LeDoux, J.E. (2004). Emotional perseveration: An update on prefrontal-amygdala interactions in fear extinction. Learning & Memory, 11, 525–535.

Sotres-Bayon F., Bush, D. E., & LeDoux, J. E. (2007). Acquisition of fear extinction requires activation of NR2B-containing NMDA receptors in the lateral amygdala. Neuropsychopharmacology, 32, 1929-40. 195

Sotres-Bayon, F., Cain, C. K., & LeDoux, J. E. (2006). Brain mechanisms of fear extinction: Historical perspectives on the contribution of prefrontal cortex. Biological Psychiatry, 60, 329-336.

Spear, N. E., & Parsons, P. J., (1976). Analysis of a reactivation treatment: Ontogenic determinants of alleviated forgetting. In R. T. Davis, D. L. Medin, & W. A. Roberts (Eds.), Processes of Animal Memory (pp. 135-165). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Spear, N. E., & Rudy, J. W. (1991). Tests of the ontogeny of learning and memory: Issues, methods, and results. In H. N. Shair, G. A. Barr, & M. A. Hofer (Eds.), Developmental psychobiology: New methods and changing concepts (pp. 84-113). New York, NY: Oxford University Press.

Sullivan, R. M., Landers, M., Yeaman, B., & Wilson, D. A. (2000). Good memories of bad events in infancy: ontogeny of conditioned fear and the amygdala. Nature, 407, 38–39.

Sullivan, R. M., Wilson, D. A., Lemon, C., & Gerhardt, G. A. (1994). Bilateral 6- OHDA lesions of the locus coeruleus impair associative olfactory learning in newborn rats. Brain Research, 643, 306 –309.

Van Eden, C. G., & Uylings, H. B. (1985). Cytoarchitectonic development of the prefrontal cortex in the rat. Journal of Comparative Neurology, 241, 253-267.

Vanderschuren, L. J., Niesink, R. J., Spruijt, B. M., & Van Ree, J. M. (1995). Mu- and kappa-opioid receptor-mediated opioid effects on social play in juvenile rats. European Journal of Pharmacology, 276, 257–266.

Vanderschuren, L. J., Spruijt, B. M., Hol, T., Niesink, R. J., & Van Ree, J. M. (1996). Sequential analysis of social play behavior in juvenile rats: Effects of morphine. Behavioural Brain Research, 72, 89–95.

Walker, D. L., Ressler, K. J., Lu, K. T., & Davis, M. (2002). Facilitation of conditioned fear extinction by systemic administration or intra-amygdala infusions of D- cycloserine as assessed with fear-potentiated startle in rats. Journal of Neuroscience, 22, 2343-2351.

Watson, D. J., Sullivan, J, R., Frank, J. G., & Stanton, M. E. (2006). Serial reversal learning of position discrimination in developing rats. Developmental Psychobiology, 48, 79-94.

Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3, 1-14.

Weber, M., McNally, G., & Richardson, R. (2006). Opioid receptors regulate retrieval of infant fear memories: Effects of naloxone on infantile amnesia. Behavioral Neuroscience, 120, 702-709. 196

Weber, M., & Richardson, R. (2001). Centrally-administered corticotropin-releasing factor and peripheral injections of strychnine hydrochloride potentiate the acoustic startle response in preweanling rats. Behavioral Neuroscience, 115, 1273-1282.

Weber, M. & Richardson, R. (2004). Pretraining inactivation of the caudal pontine reticular nucleus impairs the acquisition of conditioned fear-potentiated startle to an odor, but not light. Behavioral Neuroscience, 118, 965 – 974.

Weber, M., Watts, N., & Richardson, R. (2003). High illumination levels potentiate the acoustic startle response in preweanling rats. Behavioral Neuroscience, 117, 1458- 1462.

Westbrook, R. F., Iordanova, M., & McNally, G., Richardson, R., & Harris, J. (2002). Reinstatement of fear to an extinguished conditioned stimulus: Two roles for context. Journal of Experimental Psychology: Animal Behavior Processes, 28, 97 –110.

Wilensky, A. E., Schafe, G. E., Kristensen, M. P., & LeDoux, J. E. (2006). Rethinking the fear circuit: The central nucleus of the amygdala is required for the acquisition, consolidation, and expression of Pavlovian fear conditioning. Journal of Neuroscience, 26, 12387-12396.

Wilson, A., Brooks, D. C., & Bouton, M. E. (1995). The role of the rat hippocampal system in several effects of context in extinction. Behavioral Neuroscience, 109, 828-836

Wilson, D. A. (1984). A comparison of the postnatal development of post-activation potentiation in the neocortex and dentate gyrus of the rat. Developmental Brain Research, 16, 61-68.

Yap, C. S. L. & Richardson, R. (2005). Latent inhibition in the developing rat: An examination of context-specific effects. Developmental Psychobiology, 47, 55-65.

Yap, C. S. L. & Richardson, R. (2007). Extinction in the developing Rat: An examination of renewal effects. Developmental Psychobiology, 49, 565-575. 197

Appendix A. Raw Data4

CHAPTER 2

Experiment 1

Extinction data

P16-Reminder Extinction Block Rat 1 2 3 4 5 6 7 8 9 10 1 89 89 67 44 56 0 22 11 0 0 2 83 89 44 33 22 44 67 44 22 67 3 89 100 100 100 44 0 0 0 78 56 4 67 100 11 78 44 22 44 11 33 19 5 67 53 44 11 44 56 22 0 33 0 6 100 100 56 0 11 0 11 11 67 89 7 61 100 100 100 89 100 67 0 33 22

P16-No Reminder Extinction Block Rat 1 2 3 4 5 6 7 8 9 10 1 89 100 89 33 11 11 0 44 22 0 2 64 89 56 33 33 33 44 22 67 44 3 50 56 22 56 11 0 11 33 44 22 4 78 100 33 11 11 33 22 0 0 83 5 67 81 11 22 11 22 22 0 67 78 6 100 100 33 33 0 11 0 11 56 22 7 100 100 67 11 0 0 0 0 25 8 8 89 100 100 100 100 100 100 100 100 92

4Extinction data are rounded to whole numbers. Test data are not rounded. 198

P23-Reminder Extinction Block Rat 1 2 3 4 5 6 7 8 9 10 1 67 17 33 44 44 33 33 11 56 67 2 81 100 78 33 22 22 0 0 0 0 3 53 67 89 78 67 0 22 0 0 0 4 11 22 0 0 0 0 0 0 8 0 5 100 100 0 0 0 33 44 56 75 11 6 78 100 67 100 100 67 22 0 67 22 7 67 100 11 11 22 0 22 0 33 83 8 42 100 89 100 11 11 33 56 89 100

P23-No Reminder Extinction Block Rat12345678910 1 56 64 100 100 22 11 0 22 33 0 2 33 89 78 78 33 11 67 33 33 22 3 69 100 100 100 100 100 78 78 44 11 4 78 100 0 0 0 0 0 0 8 44 5 89 100 33 33 0 0 0 0 100 100 6 78 81 44 0 22 44 11 0 78 0 7 56 42 11 33 56 11 11 33 0 25 8 58 89 100 78 33 89 22 0 33 56

Test data

P16-Reminder P16-No Reminder P16-Not trained-Reminder Rat Baseline CS Rat Baseline CS Rat Baseline CS 1 50 42.5 1 15 20 1 10 5 2 20 37.5 2 50 70 2 10 22.5 3 003 20 35 3 50 40 4 40 67.5 4 50 12.5 4 0 17.5 5 30 30 5 10 7.5 5 5 17.5 6 5 82.5 6 20 90 6 25 42.5 7 50 0 7 10 17.5 7 5 12.5 8 45 100 8 10 2.5

199

P23-Reminder P23-No Reminder P23-Not trained-Reminder Rat Baseline CS Rat Baseline CS Rat Baseline CS 1 25 67.5 1 10 22.5 1 030 2 570 2 0 22.5 2 50 17.5 3 35 72.5 3 15 82.5 3 50 27.5 4 540 4 5354 10 10 5 50 100 5 5205 10 35 6 080 6 10 45 6 10 5 7 075 7 0 12.5 7 35 32.5 8 15 55 8 10 45 8 10 27.5

Experiment 2

P16-Same Extinction Block Test Rat 1 10 Baseline CS 1 44 33 50 45 2 22 0 15 5 3 100 11 5 72.5 4 100 11 50 45 5 100 100 5 60 6 33 42 0 0 7 42 8 0 20 8 100 56 15 70 9 100 22 45 40 10 100 0 5 70

P16-Different Extinction Block Test Rat 1 10 Baseline CS 1 66 22 10 67.5 2 47 55 0 22.5 3 33 0 25 42.5 4 58 22 50 72.5 5 100 78 0 77.5 6 100 42 15 45 7 56 0 45 75 8 67 17 15 5 9 56 0 15 65 10 22 0 0 0

200

P23-Same Extinction Block Test Rat 1 10 Baseline CS 1 83 0 25 77.5 2 81 11 40 82.5 3 83 44 25 70 4 78 8 0 60 5 100 17 10 70 6 66 78 25 62.5 7 78 17 40 75 8 67 45 0 50 9 61 91 30 92.5 10 33 17 25 57.5

P23-Different Extinction Block Test Rat 1 10 Baseline CS 1 50 22 0 10 2 67 39 5 35 3 56 8 15 7.5 4 78 11 0 37.5 5 89 11 15 32.5 6 33 11 10 60 7 100 75 30 42.5 8 89 0 0 25 9 79 42 10 7.5 10 78 0 0 45

Experiment 3.1

Immediate 48 hrs Rat Baseline CS Rat Baseline CS 1 060 1 0 17.5 2 040 2 20 37.5 3 045 3 05 4 0 32.5 4 05 5 045 5 0 12.5 6 5 62.5 6 0 27.5 7 089 7 15 7.5 8 035 8 520

201

Experiment 3.2

No Reminder Reminder Not Trained-Reminder Rat Baseline CS Rat Baseline CS Rat Baseline CS 1 0 37.5 1 25 65 1 015 2 0 2.5 2 25 87.5 2 50 25 3 5253 30 87.5 3 010 4 004 07.54 05 5 0 17.5 5 0355 015 6 0106 056 05 7 007 5257 50 8 0 2.5 8 0258 55

Experiment 3.3

No Reminder Reminder Rat Baseline CS Rat Baseline CS 1 0 27.5 1 25 55 2 025 2 25 80 3 510 3 30 82.5 4 0 92.5 4 095 5 50 67.5 5 40 90 6 05 6 20 50 7 25 42.5 7 40 92.5 8 50 8 40 52.5 9 535 9 35 77.5 10 15 52.5 10 040

202

CHAPTER 3

Experiment 4

P16-Same Extinction Block Test Rat 1 10 Baseline CS 1 100 39 0 2.5 2 92 0 20 25 3 56 0 10 17.5 4 67 0 0 15 5 89 0 0 40 6 58 61 10 20 7 58 0 0 10 8 28 11 0 27.5

P16-Different Extinction Block Test Rat 1 10 Baseline CS 1 33 100 0 2.5 2 33 25 0 5 3 67 33 15 0 4 100 0 20 32.5 5 89 8 30 42.5 6 75 42 5 17.5 7 56 0 5 27.5

P23-Same Extinction Block Test Rat 1 10 Baseline CS 1 78 56 45 72.5 2 83 11 25 30 3 92 25 5 35 4 50 0 0 5 5 83 0 0 42.5 6 92 19 10 10 7 100 50 10 55 8 89 25 50 25 9 53 44 25 67.5

203

P23-Different Extinction Block Test Rat 1 10 Baseline CS 1 22 22 40 87.5 2 100 25 35 80 3 44 22 20 70 4 83 33 45 90 5 83 0 15 68 6 83 0 50 90 7 75 0 0 32.5 8 75 11 20 87.5 9 100 33 50 95 10 64 33 35 67.5

Experiment 5.1

Vehicle-Same Extinction Block Test Rat 1 10 Baseline CS 1 56 22 5 22.5 2 92 0 10 52.5 3 56 33 25 70 4 48 8 0 10 5 100 0 40 80 6 100 55 20 55 7 33 11 25 50 8 100 33 15 90 9 44 22 0 10 10 89 33 20 35

204

Vehicle-Different Extinction Block Test Rat 1 10 Baseline CS 1 78 0 5 75 2 67 0 50 80 3 78 0 25 85 4 72 56 40 95 5 33 22 40 72.5 6 100 8 20 57.5 7 100 11 35 95 8 100 0 40 90 9 44 11 30 100 10 87 11 40 85 11 92 0 35 72.5

FG7142-Same Extinction Block Test Rat 1 10 Baseline CS 1 92 11 35 57.5 2 92 19 25 87.5 3 100 100 40 100 4 92 8 30 52.5 5 67 0 30 85 6 75 11 0 35 7 78 19 35 85 8 100 0 20 60 9 89 67 50 85 10 100 25 0 100

205

FG7142-Different Extinction Block Test Rat 1 10 Baseline CS 1 92 78 20 75 2 92 8 50 47.5 3 89 0 50 97.5 4 100 11 50 97.5 5 78 22 30 70 6 100 11 15 87.5 7 57 19 25 87.5 8 89 92 35 100 9 56 11 0 90 10 44 0 10 67.5 11 81 25 50 47.5

Experiment 5.2

Vehicle-Same Extinction Block Test Rat 1 10 Baseline CS 1 70 44 0 20 2 11 0 0 2.5 3 100 44 0 0 4 67 22 10 62.5 5 83 0 35 0 6 100 8 20 50 7 89 0 0 42.5 8 78 25 0 0 9 78 78 0 17.5

206

Vehicle-Different Extinction Block Test Rat 1 10 Baseline CS 1 61 0 0 47.5 2 89 0 5 30 3 89 11 0 0 4 56 28 20 52.5 5 75 17 0 0 6 100 67 0 52.5 7 44 0 30 42.5 8 89 0 0 55 9 89 11 0 17.5

FG7142-Same Extinction Block Test Rat 1 10 Baseline CS 1 100 22 0 40 2 67 0 0 10 3 100 78 25 37.5 4 78 45 15 27.5 5 50 11 0 5 6 100 50 25 72.5 7 56 67 0 10 8 78 0 5 27.5 9 92 83 0 35

FG7142-Different Extinction Block Test Rat 1 10 Baseline CS 1 67 0 0 35 2 56 0 0 5 3 89 0 50 47.5 4 89 8 20 47.5 5 67 44 5 12.5 6 78 100 0 77.5 7 33 0 20 70 8 75 33 0 67.5 9 81 0 0 12.5 10 67 100 0 5 11 25 58 0 2.5

207

Experiment 5.3

Control-Vehicle Test Rat Baseline CS 1 0 12.5 2 30 90 3 095 4 040 5 30 97.5 6 0 42.5 7 5 52.5 8 040

1mg/kg FG7142 Extinction Block Test Rat 1 10 Baseline CS 1 75 0 10 17.5 2 89 11 25 10 3 67 8 0 15 4 100 0 5 20 5 34 0 15 22.5 6 78 50 5 22.5 7 31 0 0 0

5mg/kg FG7142 Extinction Block Test Rat 1 10 Baseline CS 1 78 0 0 15 2 83 0 5 32.5 3 83 0 5 42.5 4 67 33 0 15 5 89 56 20 75 6 22 0 5 5 7 100 17 0 0 8 0 22 0 2.5

208

10mg/kg FG7142 Extinction Block Test Rat 1 10 Baseline CS 1 83 0 0 2.5 2 75 0 0 40 3 83 0 0 37.5 4 100 11 0 27.5 5 55 0 20 10 6 78 0 15 22.5 7 47 52 5 5 8 8 0 0 7.5

Experiment 6.1

Vehicle FG7142 Rat Baseline CS Rat Baseline CS 1 0 2.5 1 0 12.5 2 10 17.5 2 25 45 3 015 3 040 4 0 2.5 4 5 17.5 5 0 7.5 5 015 6 15 10 6 560 7 0 7.5 7 5 27.5 8 050

209

Experiment 6.2

Vehicle 1mg/kg FG7142 Rat Baseline CS Rat Baseline CS 1 510 1 015 2 12.5 27.5 2 0 2.5 3 0 7.5 3 025 4 7.5 10 4 15 0 5 5 17.5 5 0 22.5 6 10 12.5 6 0 27.5 7 2.5 2.5 7 015 8 0 47.5

5mg/kg FG7142 10mg/kg FG7142 Rat Baseline CS Rat Baseline CS 1 045 1 0 37.5 2 030 2 025 3 025 3 0 12.5 4 010 4 0 32.5 5 15 7.5 5 515 6 5 22.5 6 0 32.5 7 0 42.5 7 025 8 2.5 35 8 10 40 9 080

210

CHAPTER 4

Experiment 7

Saline Extinction Block Test Rat 1 10 Baseline CS 1 100 22 15 52.5 2 75 0 35 90 3 25 33 0 25 4 100 0 10 13 5 44 44 0 10 6 100 39 5 75 7 89 44 15 17.5 8 44 0 10 15 9 100 47 45 72.5

Bupivacaine Extinction Block Test Rat 1 10 Baseline CS 1 0 0 50 100 2 0 0 15 97.5 3 44 0 50 100 4 33 8 10 75 5 00580 6 22 0 25 90 7 8 0 50 95 8 0 11 10 77.5 9 17 11 40 97.5

211

Experiment 7.2

Saline Extinction Block Test Rat 1 10 Baseline CS 1 78 0 0 10 2 33 36 5 12.5 3 56 0 10 27.5 4 89 44 0 20 5 67 0 0 10 6 67 0 20 5

Bupivacaine Extinction Block Test Rat 1 10 Baseline CS 1 0 0 0 57.5 2 22 0 45 87.5 3 22 0 5 32.5 4 0 0 20 57.5 5 0 0 10 97.5 6 11 0 25 57.5 7 22 0 30 30

Experiment 8.1

Control Extinction Test Rat 1 10 Baseline CS 1 78 0 5 100 2 100 33 35 75 3 100 11 50 82.5 4 100 0 45 92.5 5 78 33 35 95 6 100 56 50 87.5 7 100 11 40 97.5 8 100 56 30 95

212

Saline Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 89 44 78 56 35 60 2 89 33 100 0 5 50 3 44 11 33 22 45 75 4 78 0 78 0 35 40 5 89 67 89 0 5 50 6 50 66 100 25 20 80 7 75 11 33 0 0 5 8 89 11 100 0 50 85 9 100 11 33 33 35 92.5 10 100 0 100 0 0 15

Bupivacaine Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 100 100 0 8 5 40 2 100 0 33 0 50 77.5 3 92 0 0 0 30 37.5 4 78 33 0 8 0 0 5 92 67 19 0 5 55 6 92 11 11 0 25 87.5 7 89 56 33 0 50 67.5

Experiment 8.2

Control Extinction Test Rat 1 10 Baseline CS 1 100 0 40 77.5 2 78 0 0 75 3 78 22 10 72.5 4 89 33 0 95 5 89 11 45 100

213

Saline Extinction Re-Extinction Test Rat 1 10 1 10 Baseline CS 1 89 8 67 0 5 52.5 2 100 0 83 22 20 67.5 3 98 47 89 67 10 60 4 22 11 33 11 10 15 5 83 100 83 67 10 0 6 67 0 89 22 20 45 7 89 0 75 56 45 42.5 8 78 22 92 100 30 67.5

Bupivacaine Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 100 89 53 0 5 65 2 78 0 67 0 35 70 3 67 22 0 0 30 82.5 4 67 11 0 11 30 95 5 78 22 22 0 45 95 6 100 33 0 8 35 100 7 1000002080 8 100 22 22 47 50 80

Experiment 9

P17-Saline Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 89 0 100 44 10 0 2 89 0 89 0 50 45 3 100 33 100 0 20 42.5 4 67 0 89 100 50 35 5 44 0 56 56 50 62.5 6 56 22 100 22 25 50 7 78 44 78 22 20 57.5

214

P17-Bupivacaine Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 89 0 22 11 50 90 2 78 0 33 33 50 77.5 3 100 44 0 0 5 97.5 4 33 0 0 0 45 97.5 5 33 22 0 0 25 70 6 44 0 11 0 15 55 7 100 33 11 0 45 82.5 8 56 33 0 0 25 75

P24-Saline Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 89 0 89 0 20 45 2 67 0 89 78 20 95 3 22 44 44 33 45 40 4 56 0 22 0 30 27.5 5 67 11 67 33 45 60 6 44 11 22 0 0 12.5 7 33 0 100 0 50 15

P24-Bupivacaine Extinction Re-extinction Test Rat 1 10 1 10 Baseline CS 1 56 11 0 0 30 30 2 44 11 11 0 45 70 3 44 11 33 33 50 67.5 4 33 0 22 22 35 40 5 33 0 0 0 7.5 17.5 6 100 78 11 0 40 17.5 7 56 0 0 0 0 5

215

Appendix B. Summary Tables of Statistical Analyses

CHAPTER 2

Experiment 1

Extinction: Repeated measures ANOVA summary table

Source SS df MS F p Block 122134.348 9 13570.483 18.395 < .0001 Block* Group 2384.838 9 264.982 .359 .953 Block*Age 2398.906 9 266.545 .361 .952 Block*Group*Age 3237.872 9 359.764 .488 .882 Error 179267.738 243 737.727

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Group 68.276 1 68.276 185.988 .886 Age 1426.414 1 1426.414 .021 .513 Group*Age 1006.518 1 1006.518 .440 .582 Error 87521.459 27 3241.536 .311

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 2630.281 5 526.056 1.862 .122 Within Groups 11580.357 41 282.448 Total 14210.638 46

CS: ANCOVA summary

Source SS df MS Fp Co-variate 1956.261 1 1956.261 3.652 .063 Age 1739.162 1 1739.162 3.247 .079 Group 7317.778 2 3658.889 6.831 .003 Age*Group 3635.534 2 1817.767 3.394 .044 Error 21424.315 40 535.608

216

CS: P23 post-hoc ANOVA summary

Source SS df MS Fp Group 9427.083 2 4713.542 15.214 < .0001 Error 6506.250 21 309.821

Tukey HSD

Post-hoc N Subset5 12 P23-No Reminder 8 35.6250 P23-Reminder 8 70.0000 P23-Not trained Reminder 8 23.1250

CS: P16 post-hoc ANOVA summary

Source SS df MS Fp Group 1524.500 2 762.250 .854 .441 Error 17860.826 20 893.041

Experiment 2

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 31708.562 1 31708.562 67.671 < .0001 Block*Test Context 39.978 1 39.978 .085 .772 Block*Age 435.978 1 435.978 .930 .341 Block*Test Context*Age 233.859 1 233.859 .499 .485 Error 16400.000 35 468.571

Extinction: Between measures ANOVA summary table

Source SS df MS F p Test Context 1226.451 1 1226.451 1.478 .232 Age 404.829 1 404.829 .488 .490 Test Context*Age 73.154 1 73.154 .088 .768 Error 29050.511 35

5Groupsbelongingindifferentsubsetsindicatethattherearesignificantdifferencesbetweenthem.The valuesinsubsetsindicatethegroupmean. 217

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 1012.500 3 337.500 1.277 .297 Within Groups 9515.000 36 264.306 Total 10527.500 39

CS: ANCOVA summary

Source SS df MS Fp Co-variate 1506.641 1 1506.641 3.119 .086 Age 389.293 1 389.293 .806 .375 Test Context 1996.267 1 1996.267 4.133 .050 Age*Test Context 3715.712 1 3715.712 7.693 .009 Error 16904.234 35

CS: P23 post-hoc independent samples t-test

tdfp Mean difference SE Difference -5.756 18 < .0001 -39.55000 6.87154

CS: P16 post-hoc indpendent samples t-test

tdfp Mean difference SE Difference .359 18 .724 4.50000 12.54381

Experiment 3.1

Baseline: Independent samples t-test

tdfp Mean difference SE Difference -1.507 14 .154 -4.37500 2.90282

CS: ANCOVA summary

Source SS df MS Fp Co-variate 269.901 1 269.901 1.126 .308 Training-Test Interval 4936.051 1 4936.051 20.599 .001 Error 3115.192 13 239.630

218

Experiment 3.2

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 418.750 2 209.375 1.296 .295 Within Groups 3393.750 21 161.607 Total 3812.500 23

CS: ANCOVA summary

Source SS df MS Fp Co-variate 4071.450 1 4071.450 14.656 .001 Group 3352.429 2 1676.215 6.034 .009 Error 5555.894 20 277.795

Tukey HSD

Post-hoc N Subset 12 No Reminder 8 11.8750 Reminder 8 42.1875 Not trained-Reminder 8 10.0000

Experiment 3.3

Baseline: Independent samples t-test

tdfp Mean difference SE Difference 2.147 18 .046 15.00000 6.98809

CS: ANCOVA summary

Source SS df MS Fp Co-variate 1339.406 1 1339.406 2.291 .148 Group 3003.893 1 3003.893 5.138 .037 Error 9938.719 17 584.631

219

CHAPTER 3

Experiment 4

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 39006.872 1 39006.872 60.659 < .0001 Block* Test Context 195.175 1 195.175 .304 .586 Block*Age 754.625 1 754.625 1.173 .287 Block*Test Context*Age 1513.481 1 1513.481 2.354 .135 Error 19291.689 30 643.056

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Test Context 7.637 1 7.637 .016 .900 Age 664.528 1 664.528 1.386 .248 Test Context*Age 86.311 1 86.311 .180 .674 Error 14381.012 30 479.367

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 3412.183 3 1137.394 5.290 .005 Within Groups 6450.317 30 215.011 Total 9862.500 33

CS: ANCOVA summary

Source SS df MS Fp Co-variate 3141.451 1 3141.451 13.039 .001 Test Context 4272.642 1 4272.642 17.734 <.0001 Group 1167.296 1 1167.296 4.845 .036 Test Context*Group 2634.003 1 2634.003 10.933 .003 Error 6986.768 29 240.923

Tukey HSD

Post-hoc N Subset 12 P16-Same 8 19.6875 P16-Different 7 18.2143 P23-Same 9 38.0556 P23-Different 10 76.8000

220

Experiment 5.1

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 71570.634 1 71570.634 154.411 < .0001 Block* Test Context 96.443 1 96.443 .208 .651 Block*Drug .316 1 .316 .001 .979 Block*Test Context*Drug 773.839 1 773.839 1.670 .204 Error 17613.336 38 463.509

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Test Context 287.543 1 287.543 .474 .496 Drug 1872.900 1 1872.900 3.084 .087 Test Context*Age 22.900 1 22.900 .038 .847 Error 23074.318 38 607.219

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 1699.924 3 566.641 2.509 .073 Within Groups 8583.409 38 225.879 Total 10283.333 41

CS: ANCOVA summary

Source SS df MS Fp Co-variate 1259.698 1 1259.698 3.044 .089 Test Context 2221.782 1 2221.782 5.368 .026 Drug 1096.054 1 1096.054 2.648 .112 Test Context*Drug 1659.601 1 1659.601 4.010 .050 Error 37

Tukey HSD

Post-hoc N Subset 12 Vehicle-Same 10 47.5000 Vehicle-Different 11 82.5000 FG7142-Same 10 74.7500 FG7142-Different 11 78.8636

221

Experiment 5.2

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 41664.381 1 41664.381 66.973 < .0001 Block* Test Context 39.620 1 39.620 .064 .802 Block*Drug 1621.403 1 1621.403 2.606 .116 Block*Test Context*Drug 344.260 1 344.260 .553 .462 Error 21151.677 34 622.108

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Test Context 1080.809 1 1080.809 1.284 .265 Drug 766.063 1 766.063 .910 .347 Test Context*Age 247.967 1 247.967 .295 .591 Error 28612.869 34 841.555

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 32.955 3 10.985 .065 .978 Within Groups 5704.545 34 167.781 Total 5737.500 37

CS: ANCOVA summary

Source SS df MS Fp Co-variate 2267.572 1 2267.572 4.135 .050 Test Context 671.326 1 671.326 1.224 .277 Drug 133.929 1 133.929 .244 .624 Test Context*Drug 125.496 1 125.496 .229 .636 Error 18095.055 33 548.335

Experiment 5.3

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 34800.162 1 34800.162 56.595 < .0001 Block* Group 203.071 2 101.536 .165 .849 Error 12297.929 20 614.896

222

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Group 105.132 2 52.566 .077 .926 Error 13625.607 20 681.280

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 105.213 3 35.071 .371 .775 Within Groups 2554.464 27 94.610 Total 2659.677 30

CS: ANCOVA summary

Source SS df MS Fp Co-variate 3527.964 1 3527.964 9.241 .005 Group 8577.042 3 2859.014 7.489 .001 Error 9925.831 26 381.763

Tukey HSD

Post-hoc N Subset 12 Control 8 58.7500 1mg/kg-FG7142 7 15.3571 5mg/kg-FG7142 8 19.0625 10mg/kg-FG7142 8 23.4375

Experiment 6.1

Baseline: Independent samples t-test

t df p Mean difference SE Difference -1.507 14 .154 -4.37500 2.90282

CS: ANCOVA summary

Source SS df MS Fp Co-variate 269.901 1 269.901 1.126 .308 Training-Test Interval 4936.051 1 4936.051 20.599 .001 Error 3115.192 13 239.630

223

Experiment 6.2

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 418.750 2 209.375 1.296 .295 Within Groups 3393.750 21 161.607 Total 3812.500 23

CS: ANCOVA summary

Source SS df MS Fp Co-variate 4071.450 1 4071.450 14.656 .001 Group 3352.429 2 1676.215 6.034 .009 Error 5555.894 20 277.795

Tukey HSD

Post-hoc N Subset 12 No Reminder 8 11.8750 Reminder 8 42.1875 Not trained-Reminder 8 10.0000

224

CHAPTER 4

Experiment 7.1

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 8160.111 1 8160.111 21.320 < .0001 Block*Drug 3481.000 1 3481.000 9.095 .008 Error 6123.889 16 382.743

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Drug 15708.444 1 15708.444 37.505 < .0001 Error 6701.444 16 418.840

Baseline: Independent samples t-test

tdfp Mean difference SE Difference -1.625 16 .124 -13.33333 8.20738

CS: ANCOVA summary

Source SS df MS Fp Co-variate 3019.253 1 3019.253 7.932 .013 Drug 5752.801 1 5752.801 15.114 .001 Error 5709.302 15 380.620

Experiment 7.2

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 6343.795 1 6343.795 28.244 < .0001 Block*Drug 2671.487 1 2671.487 11.894 .005 Error 2470.667 11 224.606

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Drug 7323.795 1 7323.795 35.355 < .0001 Error 2278.667 11 207.152

225

Baseline: Independent samples t-test

tdfp Mean difference SE Difference -1.894 11 .085 -13.45238 7.10152

CS: ANCOVA summary

Source SS df MS Fp Co-variate 91.142 1 91.142 .223 .647 Drug 4462.461 1 4462.461 10.938 .008 Error 4079.691 10 407.969

Experiment 8.1

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 43120.950 1 43120.950 81.527 < .0001 Block*Group 622.116 2 311.058 .588 .564 Error 11636.164 22 528.917

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Group 1254.870 2 627.435 1.253 .305 Error 11016.050 22 500.730

Re-extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 10740.813 1 10740.813 19.852 < .0001 Block*Group 5018.461 1 5018.461 9.276 .008 Error 8115.657 15 541.044

Re-extinction: Between measures ANOVA summary table

Source SS df MS Fp Group 10672.941 1 10672.941 33.311 < .0001 Error 4806.000 15 320.400

226

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 922.786 2 461.393 1.347 .281 Within Groups 7533.214 22 342.419 Total 8456.000 24

CS: ANCOVA summary

Source SS df MS Fp Co-variate 4450.337 1 4450.337 10.289 .004 Group 3472.857 2 1736.428 4.014 .033 Error 9083.567 21 432.551

Tukey HSD

Post-hoc N Subset 12 Control 8 90.6250 Saline 10 55.3000 Bupivacaine 7 52.2857

Experiment 8.2

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 39995.084 1 39995.084 82.594 < .0001 Block*Group 547.617 2 273.809 .565 .578 Error 8716.288 18 484.238

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Group 254.717 2 127.359 .182 .835 Error 12579.188 18 698.844

Re-extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 4140.500 1 4140.500 8.991 .010 Block*Group 882.000 1 882.000 1.915 .188 Error 6447.500 14 460.536

227

Re-extinction: Between measures ANOVA summary table

Source SS df MS Fp Group 16471.125 1 16471.125 20.492 < .0001 Error 11252.750 14 803.768

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 762.143 2 381.071 1.509 .248 Within Groups 4545.000 18 252.500 Total 5307.143 20

CS: ANCOVA summary

Source SS df MS Fp Co-variate 561.937 1 561.937 1.779 .200 Group 6619.080 2 3309.540 10.478 .001 Error 5369.782 17 315.870

Tukey HSD

Post-hoc N Subset 12 Control 5 84.0000 Saline 8 43.7500 Bupivacaine 8 83.4375

Experiment 9

Extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 33191.309 1 33191.309 99.138 < .0001 Block*Drug 312.120 1 312.120 .932 .344 Block*Extinction Age 796.535 1 796.535 2.379 .136 Block*Drug*Extinction Age 4.793 1 4.793 .014 .906 Error 8370.009 25 334.800

228

Extinction: Between measures ANOVA summary table

Source SS df MS Fp Drug .936 1 .936 .001 .971 Extinction Age 1474.991 1 1474.991 2.185 .152 Drug*Extinction Age 98.567 1 98.567 .146 .706 Error 16874.152 25 674.966

Re-extinction: Repeated measures ANOVA summary table

Source SS df MS Fp Block 9236.627 1 9236.627 23.317 < .0001 Block*Drug 6772.120 1 6772.120 17.095 < .0001 Block*Extinction Age 135.936 1 135.936 .343 .563 Block*Drug*Extinction Age 95.890 1 95.890 .242 .627 Error 9903.438 25 396.138

Re-extinction: Between measures ANOVA summary table

Source SS df MS Fp Drug 26328.549 1 26328.549 43.624 < .0001 Extinction Age 1178.724 1 1178.724 1.953 .175 Drug*Extinction Age 1716.152 1 1716.152 2.843 .104 Error 15088.438 25 603.538

Baseline: Overall ANOVA

Source SS df MS Fp Between Groups 46.613 3 15.538 .049 .985 Within Groups 7998.214 25 319.929 Total 8044.828 28

CS: ANCOVA summary

Source SS df MS Fp Co-variate 1769.274 1 1769.274 3.838 .062 Drug 1847.743 1 1847.743 4.008 .057 Extinction Age 3264.539 1 3264.539 7.082 .014 Drug*Extinction Age 3693.106 1 3693.106 8.012 .009 Error 11063.173 24 460.966

229

Tukey HSD

Post-hoc N Subset 12 P17-Saline 7 41.8571 P17-Bupivacaine 8 80.6250 P24-Saline 7 42.1429 P24-Bupivacaine 7 35.3571