Effects of Propranolol on Cognition and Eye Contact in ASD 03.25.10

Effects of Propranolol on Cognition and Eye Contact in Autism Spectrum Disorder (ASD)

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

Sanjida Shoma Saklayen

Integrated Biomedical Sciences Graduate Program

The Ohio State University

2010

Dissertation Committee:

David Q. Beversdorf, Advisor

Howard Gu, Advisor

Sandra Kostyk

Wolfgang Sadee

Sanjida Shoma Saklayen

2010

Abstract

Propranolol, a nonselective beta blocker, produces noradrenergic blockade with central and peripheral nervous system effects. While propranolol is often prescribed for hypertension, it is also commonly prescribed for situational anxiety (stage fright, test anxiety, etc) due to its central activity. Previous work in this lab has examined the effect of propranolol on cognitive flexibility tasks. In a previous study in this lab, performance on simple cognitive flexibility tasks was shown to be increased in autistic individuals who take propranolol, whereas controls only exhibited improvement in difficult tasks. Our aim was to examine other possible benefits of propranolol on cognition. To do this, we compared performance on verbal fluency tasks, which require cognitive flexibility, between autistic and control individuals, under propranolol and placebo conditions.

Furthermore, it is characteristic of individuals with autism to exhibit poor eye contact with others from an early age. Recent physiological evidence suggests that direct eye contact may be stressful to those affected by autism. Stress is well known to activate the noradrenergic system. Therefore, an agent that could reliably decrease the stress related to eye contact by acting to block noradrenergic activation may be beneficial to

those affected with autism. Thus, since decreased eye contact in autistic individuals may be linked to stress and propranolol is known to decrease social stress, we proposed to determine whether autism-affected individuals would increase their eye contact when given propranolol. We hypothesized that propranolol administration, through its action of decreasing the stress response, would lead patients with autism to spend more proportionate time making eye contact, compared to placebo administration. Eye contact was measured using an ASL eyetracker and dynamic video stimuli of 16 novel faces at each of two drug condition visits. Eyetracker data was analyzed using the EyeNal and FixPlot programs by ASL.

Fourteen autism subjects with age/IQ/gender matched controls were tested in the verbal fluency study and the same fourteen autism subjects participated in the eyetracking study. Results indicate significant improvements in semantic fluency in autism subjects given propranolol, relative to the placebo condition. 2x2 ANOVA in the semantic fluency task revealed a trend for an interaction effect of drug and group as well as a significant main effect of drug, driven by the ASD group. In the eyetracking study, individuals with ASD and controls had similar amounts of eye-to-eye gaze under the placebo condition, which was unexpected. However, both groups improved significantly in the propranolol condition. 2x2 ANOVA in the eyetracker task revealed a trend for an interaction effect of drug and group as well as a main effect of drug.

iii

Dedication

Dedicated to my grandparents:

The late-Syed Fariduddin Saki, “Nana”

The late-Syeda Shamsun Nahar, “Nani”

The late-Azimuddin Ahmed, “Dadu”

Salema Azim, “Dadi”

Acknowledgements

*To my co-advisors - Dr.David Beversdorf (“Dr. B”) and Dr. Howard Gu, and my committee members, current and former - Dr. Sandra Kostyk, Dr. David Saffen, and Dr. Wolfgang

Sadee, thank you for your commitment to provide me with excellent training during my graduate career. *To my lab colleagues at OSU and MU: Allen Carpenter, Patrick Hecht,

Katherine Higgins, Karen Jones, Namhee Kim, Ananth Narayanan, Ryan Smith, and

Catherine White, thanks for making the Beversdorf Lab a great place to work and learn.

*To Dr. Shawn Christ at MU and his lab team, thank you for sharing your labspace, equipment, training and assistance. *To the faculty, staff, and students of the OSU MSP and IBGP, as well as the College of Medicine at OSU, thank you for providing comprehensive training and support. *To the OSU Neurology Department, the departments of Neurology, Radiology, and Psychology at MU, as well as the Thompson

Center for the use of their resources, thank you. *To Dr. Menachem Shoham and Dr.

Elizabeth Pehek of Case Western Reserve University, thank you for introducing me to research. *To all of my family and friends, I extend deepest thanks for helping me along the way - particularly to Kyle Schneider; my siblings, Sabir, Samiya, and Sabera; my mother, Syeda Saklayen, and father, Dr. Mohammad Saklayen. *To all, gratitude.

Vita

May 1999……..…….Centerville High School, Centerville, OH

2001-2003……………Research Assistant, Department of Psychiatry, Case Western

2003…………………….B.A. Biochemistry and Psychology, with honors, Case Western

2003 to present ……Graduate Associate and Fellow (2006-2009), The Ohio State University

Publications

Saklayen SS, Mabrouk O, Pehek EA. (2003) “Negative Feedback Regulation of Nigrostriatal

Dopamine Release: Mediation by Striatal D1 Receptors.” JPET 311 (1); 342-348.

Cao R, Lee B, Cho HY, Saklayen SS, Obrietan K. (2008). “Photic Regulation of the mTOR

Signaling Pathway in the Suprachiasmatic Circadian Clock.” Mol Cell Neurosci 38

(3); 312-324.

Fields of Study

Major Field: Integrated Biomedical Sciences Graduate Program

Areas of Interest: Biology of Neurological Disorders and Translational Research

Minor Field: Neuroscience

Table of Contents

Abstract…………………………………………………………………………………………………………………………….ii

Dedication………………………………………………………………………………………………………………………..iv

Acknowledgements…………………………………………………………………………………………………………..v

Vita…………………………………………………………………………………………………………………………………..vi

List of Tables….…………………………………………………………………………………………………………...... viii

List of Figures……………………………………………………………………………………………………………...... ix

Chapter 1: Introduction….…………………………………………………………………………………………………1

Chapter 2: Effect of Propranolol on Verbal Fluency in ASD……………………………………………..29

Chapter 3: Effect of Propranolol on Eye Contact in ASD………………………………………………… 54

Chapter 4: Discussion……………………………………………………………………………………………………..84

Chapter 5: Future Directions……….……………………………………………………………………………….…94

References……………………………………………………………………………………………………………………..99

Appendix A: Verbal Fluency data …….…………………………………………………………………...... 123

Appendix B: Eye Contact data……..………………………………………………………………………………..126

vii

List of Tables

Table 1. Demographic and diagnostic data for ASD group………………………………………….....34

Table 2. Mean words generated…..………………………………….…………………………………………….41

Table 3. Demographic and diagnostic data for ASD group………………….…………………………..58

Table 4. Terminology……………………..……………………………………………………………………………….68

Table 5. Proportionate time on the eyes and Total time on AOIs…………………………………….75

viii

List of Figures

Figure 1. Norepinephrine…..……………………………………………………………………………………………..8

Figure 2. Norepinephrine synthesis…………………………………………………………………………………..9

Figure 3. Semantic network restriction in ASD………………………………………………………………..18

Figure 4. Remote eyetracker…………..………………………………………………………………………………20

Figure 5. Propranolol……………………………………………………………………………………………………..24

Figure 6. Blood pressure……………….…………………………………………………………………………….….38

Figure 7. Heart rate…………………….…………………………………………………………………………………..39

Figure 8. Semantic verbal fluency……………………………………………………………………………………42

Figure 9. Individual scores on semantic fluency task………………………………………………………43

Figure 10. Individuals with Asperger’s syndrome vs. individuals with autism…………………44

Figure 11. Phonemic verbal fluency………………………………………………………………………………..45

Figure 12. Individual scores on phonemic fluency task…………………………………………...... 46

Figure 13. Eyetracker calibration……….……………………………………………………………………………60

Figure 14. Areas of interest…………………………………………………………………………………………….63

Figure 15. Sample .eyd file………………………………………………………………………………………………65

Figure 16. Visualizing eye data……………………………………………………………………………………….66

Figure 17. Blood pressure……………………………………………………………………………………………….71

Figure 18. Heart rate……………………………………………………………………………………………………….72

Figure 19. Amount of fixation on individual facial features by group and drug ……………….76

Figure 20. Proportionate time spent on the eyes……………………………………………………….……77

CHAPTER 1: INTRODUCTION

Dr. Temple Grandin designs plans for facilities that handle livestock. Her purpose is to alleviate animal stress and to help develop more humane methods of slaughter in the meat industry [1]. She has designed major facilities of this type that are located around the world and at this time, the majority of beef cattle in the United States are handled in a restraining system of her design [1]. Grandin also teaches about livestock behavior and facility design at the university level and she is currently the author of over three-hundred scholarly articles and seven books.

Despite her current professional success, when she was a child, Grandin did not learn to speak until she was beyond the age of three [2]. Unable to verbalize her needs, her early attempts to communicate were primarily through screams and humming noises

[2-3]. When she was diagnosed with autism in 1950, her doctors recommended that she be institutionalized for her condition [2-3]. Several years later, in her book, entitled

Emergence: Labeled Autistic , Grandin described her experiences in rising above societal and medical expectations to becoming the successful scientist that she is today [3-4].

Grandin’s book provided personal insight into the minds of autistic individuals and a voice

to the struggles that many of them face for the first time since child psychiatrist Leo

Kanner described autism in 1943 [5]. It also increased hope among many in the medical community and families that their patients and loved ones with autism could also live fulfilling and successful lives.

Initially, the word “autism” was coined by Eugene Bleuler in 1910 to refer to the loss of contact with reality often experienced by patients with adult schizophrenia [6].

After a study of eleven children who exhibited signs of social withdrawal, Kanner described “infantile autism” as "lack of affective contact, fascination with objects, desire for sameness and non-communicative language before 30 months of age" [5]. Afterwards though, autism was still considered a form of childhood schizophrenia for several decades. In 1967, the World Health Organization (WHO) included autism in their

International Statistical Classification of Diseases and Related Health Problems (ICD-8) under the heading of schizophrenia [7]. Thirteen years later, autism was reclassified as an independent diagnosis in the Diagnostic and Statistical Manual of Mental Disorders (DSM-

III) in 1980 [8]. A related condition to autism, Asperger syndrome, was characterized in

1944, shortly after Kanner’s work was first published, and was added to the DSM-IV in

1994 [9].

According to the DSM-IV, autism is characterized by social and communication impairments as well as stereotypical behaviors that begin prior to age three [9]. Asperger syndrome is also described by these characteristics, with the exception that language is preserved [9]. The term “Autism Spectrum Disorder (ASD)” refers to both types of conditions, grouping them together as on a continuum [10-12]

Social impairments in ASD can be grouped as a triad of specific deficiencies in social recognition, social communication, and social understanding that are observed in early childhood [13-16]. Social recognition is defined as the ability to identify and appreciate other human beings as worthy of one’s interest, and impairments can range from mild (e.g. a limited intellectual understanding of social behavior) to severe (e.g. total indifference) [17]. Social communication includes the interchange of nonverbal and verbal social cues as well as exchange of ideas from person to person [17]. An autistic individual with mild impairment of social communication may be quite verbal but lack appropriate reciprocity in conversations; in contrast, severe impairment may produce a complete lack of desire for communication with others [17]. Finally, individuals with ASD lack a degree of social understanding, or “theory of mind,” the ability to internally recognize that other individuals have minds of their own [18]. Individuals with autism have difficulty with tasks that require understanding others’ perspectives [18]. Typically, young children develop theory of mind at an early age, but children with autism develop theory of mind later, if at all [19]. For example, children with autism had difficulty attributing situation-appropriate beliefs to puppets in a play and to use this information to predict the subsequent behavior of the puppet character in the scene [18]. In a higher functioning individual, the lack of theory of mind may manifest as an intellectual understanding that other human beings have internal thoughts and feelings, however, this limited awareness is accompanied by an inability to recognize and respond to others in this context [17].

Another primary diagnostic criteria for ASD is the presence of repetitive or stereotypical activity [9]. In a higher functioning individual, one may observe an intense absorption with accumulation of factual information or repetitive viewing, listening or reading of particular media, to an extent that exceeds common interest [17]. Patients affected more severely may exhibit physical movements such as repetitive arm-flapping or rocking behavior, whereas the most severely affected individuals may lack spontaneous physical behavior entirely [17].

Individuals with ASD may also demonstrate decreased language comprehension as well as impairments in social use of language [20-23]. Language deficiencies in less severely affected patients may range from inability to comprehend others’ responses in conversation to difficulty with generating or using specific components of language such as semantics, vocabulary and syntax. More affected individuals may exhibit echolalia, verbal repetition of nonsensical words or sounds, and some patients may be entirely mute [17].

Two of the most common tests clinically used when ASD is being considered in a diagnostic differential include the Autism Diagnostic Interview-Revised (ADI-R) [24] and the Autism Diagnostic Observation Schedule (ADOS) [25-26], which are often used in conjunction. An examiner administering the ADI-R conducts a detailed interview with the parents of the patient being referred. Retrospective parental reports of the patient’s birth and early development are used to determine whether developmental characteristics of ASD are present, and the timeframe in which they occur. Four subscores are calculated for each patient in the areas of communication, repetitive

behaviors and stereotyped patterns, qualitative impairments in reciprocal social interaction, and age of abnormality in development [24]. Scores in each domain must be higher than a predetermined cutoff to indicate a diagnosis of autism. On the other hand, the ADOS involves direct patient-examiner interaction while the patient is asked to perform a series of tasks. The examiner observes and scores by pre-determined criteria.

Subscores measure communication skills, social skills, and the interaction between the two[26].

Statistics suggest that the diagnosis of autism has increased significantly over the past decade [27]. Some report that from 3 to 6 children of every 1,000 will develop autism [28]. Others have suggested rates as high as 1 in 150 children may be at risk [29].

This increase is suggested to be due to better clinical judgment among medical professionals, the development of broader diagnostic criteria, and increased public awareness [30-31].

Despite the growing number of affected children and adults, the etiology of ASD remains unknown. Many have suggested a strong link between genetics and autism [32-

38]. Autism researchers have proposed a number of candidates for genes contributing to

ASD, including those for the synaptic cell-adhesion proteins, neurexins and neuroligins

[39-45]. In addition to genetics, other important factors such as prenatal stress [46-47], timing of birth [48-49] or early developmental environment [50], have been thought to play a role in the development of autism. Unfortunately, while many continue to explore possibilities, at this time a preventable cause, or cure, for ASD remains unknown [51].

Until an etiology is discovered, exploration of therapeutic strategies will be the critical factor toward improving lives of patients with ASD.

The primary goals of ASD treatment are to improve quality of life and functional independence [52]. Standard therapeutic regimens for ASD patients often include pharmacologic treatments. Drugs that are currently thought to benefit autism patients include atypical antipsychotics [53-56], selective-serotinin reuptake inhibitors (SSRIs)[57], psychostimulants [58-59], and alpha-adrenergic agonists [60]. The most researched psychotropic drug in ASD is risperidone, a neuroleptic that has demonstrated benefit for behavioral symptoms of autism [61-62]. Risperidone is also the only drug approved by the Federal Drug Administration in the United States for the treatment of autism, although other classes of drugs are prescribed off-label as needed for treatment [63].

One class of these, SSRIs, are thought to improve impairments of global function in ASD

[57, 64-69]. Alpha-adrenergic agonists, such as clonidine, are used to reduce impulsivity and inattention and also benefit sleep [70]. Most literature suggests that drug treatment that is currently available for ASD simply alleviates problematic symptoms, such as aggression, and does not address core impairments that are inherent to the condition [57,

71].

Social and behavioral support and training such as applied behavioral analysis

(ABA) [72] or occupational therapy [73] have been suggested to provide more benefit of core symptoms of autism such as social and language skills. Parent education [52, 74], academic support and behavioral interventions in school systems [75-77] as well as peer support through structured patient support groups [78-79] are also beneficial. Many

studies suggest that a tandem approach, involving both pharmacological and behavioral therapy, results in overall behavioral improvement in some ASD patients [80]. However, it is agreed upon that early intervention strategies in ASD may be the most vital factor for effective treatment. While some methods for early interventions have been proposed

[81-82], further study is still needed.

Thus, since most drugs used in the current treatment of ASD do not affect core symptoms, novel approaches to treatment of those symptoms will also be critical to developing effective interventions in ASD. Indeed, beta-adrenergic blockade has shown benefits in language and social behaviors in ASD [83]. Similar findings were observed with alpha-2 adrenergic agonists [84-86]. Both beta-adrenergic antagonists and alpha-2- adrenergic agonists act to produce blockade of the noradrenergic system [87-91].

The noradrenergic system refers to all body systems that carry norepinephrine, a catecholaminergic chemical messenger that acts as both a hormone in the blood and a neurotransmitter in the brain and autonomic nervous system [92]. The chemical structure of norepinephrine, also called noradrenaline, consists of a benzene ring linked to two hydroxyl groups and an aminohydroxyethyl chain, yielding the formal chemical name 4-(2-amino-1-hydroxyethyl)benzene-1,2-diol (see Figure 1).

Figure 1. Norepinephrine. Chemical structure of norepinephrine [93].

Norepinephrine is synthesized via the same synthetic pathway that yields all other catecholamines (see Figure 2). Biosynthesis of norepinephrine begins with the amino acid tyrosine, which is hydroxylated to levo-dihydroxyphenylalanine (L-DOPA) by tyrosine hydroxylase. L-DOPA in turn is decarboxylated to dopamine by DOPA decarboxylase. Dopamine-beta-hydroxylase converts dopamine to norepinephrine in a beta-oxidation step. Also, epinephrine, which differs from norepinephrine by the addition of a methyl group, can be synthesized from norepinephrine through the action of phenylethanolamine N-methyltransferase (PNMT). Major metabolites of norepinephrine are normatenephrine (synthesized through catechol–O-methyltransferase, or COMT) as well as 3,4-dihydoxymandelic acid(via monoamine oxidase, MAO), 3-methoxy-4- hydroxymandelic acid (VMA, also made by MAO), and 3-methoxy-4- hydroxyphenylethylene glycol (MHPG, also by MAO) [94].

Catecholamine biosynthesis

L-tyrosine

Tyrosine hydroxylase (rate-limiting)

L-dihydroxyphenylalanine (L-DOPA)

DOPA decarboxylase Aromatic L-amino acid decarboxylase

Dopamine

Dopamine beta-hydroxylase

Norepinephrine

Phenylethanolamine N-methyltransferase

Epinephrine

Figure 2. Norepinephrine synthesis . Synthesis of norepinephrine occurs via the catecholamine biosynthetic pathway [95].

Norepinephrine is synthesized in chromaffin cells of the adrenal medulla, located in the adrenal gland above the kidney, as well as postganglionic neurons of the sympathetic nervous system, whose cell bodies are found in the dorsal root ganglia of the sympathetic chain along the spinal column and project axons to effector organs of the

autonomic nervous system. Norepinephrine is released from the adrenal medulla, along with epinephrine, during activation of the sympathetic nervous system [96], involved in fight-or-flight reactions [97]. Physiologic changes that result from increased sympathetic activity include increased blood pressure, increased heart rate, constriction of peripheral blood vessels, dilation of lung bronchi, activation of sweat glands, pupillary dilation, and decreased gut motility [98].

Nine receptor subtypes have been identified that bind norepinephrine: alpha-1a, alpha-1b, alpha-1c, alpha 1d, alpha-2a, alpha-2b, alpha-2c, beta-1, beta-2, and beta-3

[99]. Of these, alpha-1-class agonism by endogenous norepinephrine mediates most aspects of sympathetic activation [98]. However, beta-2 adrenergic activation also plays a major role. Binding of beta-2 receptors by endogenous norepinephrine contributes to increased heart rate and force, vasoconstriction of peripheral vasculature, bronchodilation of lungs, increased glucose production (glycogenolysis and gluconeogenesis) in the liver, increased saliva production, and decreased gut motility [98].

Both alpha-class and beta-class adrenergic receptors are G-protein-coupled receptors. Alpha-class receptors, both alpha-1 and alpha-2, activate a Gi-protein which in turn activates phospholipase C. Phospholipase C then affects inositol triphosphate (IP3) and calcium ion [100]. Also activated in the process, phosphate kinase C works to increase protein phosphorylation throughout the cell [100]. Alpha-1 receptors are located postsynaptically whereas alpha-2 receptors are located presynaptically as autoreceptors [94]. Well known agents that specifically activate alpha-1-class receptors include phenylephrine, methoxamine, and cirazoline, while specific alpha-1-class blocking

drugs include prazosin and doxazosin [101]. Beta-2 adrenergic receptors activate a Gs- protein which upregulates adenylyl cyclase, and by extension, protein kinase A to also increase cellular protein activation [100]. Beta-2 –adrenergic agonists are used in the treatment of obstructive respiratory diseases, such as asthma and chronic obstructive pulmonary disease (COPD) due to their dilatory effect on lung bronchi. Common clinically used beta-2-agonists include albuterol, isoproterenol, and terbutaline. Beta-2-specific antagonists, such as propranolol, nadolol, and metoprolol, are commonly used in the clinical treatment of hypertension, by lowering blood pressure and heart rate through noradrenergic blockade, and angina, by decreasing contractile force of the heart [101].

Although noradrenergic system regulation plays an important role in physiologic responses to the autonomic nervous system, norepinephrine also has a significant role as a neurotransmitter in the central nervous system. In the brain, the majority of noradrenergic cell bodies are contained in the nucleus known as the locus coeruleus, which is found adjacent to the fourth ventricle along the dorsal pons [99]. Relative to the trillions of neurons in the human nervous system, there are quite few noradrenergic cell bodies (15,000) located in the locus coeruleus [99]. From this small nucleus, however, project countless distributions of axons that travel the entire central nervous system

[102-104]. Projections from the locus coeruleus travel to many other brain areas, including the frontal and prefrontal cortices, the motor cortex, and the visual cortex; yet, it is specific areas within these regions that ultimately receive innervation from the locus coeruleus [99]. Thus, while there is global reach of noradrenergic innervation through the

widespread distribution of axons, only particular local areas are likely to be densely innervated, such as particular cortical layers in primates [103, 105].

Hasselmo et al. examined the effects of exogenous norepinephrine on excitatory synaptic potentials in rat piriform cortex, involved in olfaction. Two cortical layers were examined, Ia and Ib. Cortical layer Ia contained afferent fibers from the olfactory bulb, and cortical layer Ib contained pyramidal cells that synapsed among themselves and also synapsed with inhibitory interneurons to the hippocampus. Perfusion of the cortical brain slices with norepinephrine produced a selective decrease in excitatory synaptic potentials in a laminar distribution. Layer Ib showed more suppression than layer Ia. These effects were dose-dependent. Altogether, norepinephrine application enhanced the effect of selective suppression between the cortical layers, so layer Ib became even more suppressed while layer Ia was even less suppressed. Hasselmo’s findings supported a cellular mechanism for the signal: noise ratio previously observed for norepinephrine in neurophysiologic studies [106]. Thus, cortical signal: noise ratio Is thought to be increased by norepinephrine, while noradrenergic blockade by a beta-blocker like propranolol decreases the signal : noise ratio.

Studies have found important roles for norepinephrine in arousal [107], attention

[107], working memory [108], executive function [109], stress [96] and cognition [110] in typical individuals. Noradrenergic effects on arousal are mediated through action at cortical and thalamic neurons [99]. Increased tonic activation, or sustained, low- frequency discharge, is thought to be produce distractibility or inattention, whereas increased phasic activation, or short bursts of episodic discharge is thought to be involved

in sustained attention [99, 111]. Research also points to a role for noradrenaline in working memory due to its release from locus coeruleus neurons to the prefrontal cortex, which plays a role in suppressing attention to irrelevant stimuli [99, 112-113]. Executive functions such as inhibition and motivation are also thought to be modulated by norepinephrine [114].

Substantial evidence shows that the stress response is also mediated by the noradrenergic system [96-97, 115-121]. The stress response is well known to involve both the sympatho-adreno-medullary system (SAM) and the hypothalamic-pituitary-adrenal axis (HPA axis). Each pathway modulates physiologic adaptations when activated by stressors. In brief, the effect of the SAM system is to release norepinephrine and epinephrine from the adrenal medulla, as referred to previously. The HPA axis, on the other hand, increases cortisol production from the adrenal cortex through a targeted cascade of hormones primarily located in the hypothalamus and pituitary gland. These two neuroendocrine pathways are both known to respond to major metabolic stressors.

However, for other stressors, they are thought to differ by timecourse and thresholds for activation [122-124].

Some animal studies suggest that fibers containing one of the HPA axis hormones, corticotrophin-releasing hormone (CRH), project to the locus coeruleus from the central amygdalaee, involved in emotional learning and emotional memory [125-131]. These

CRH-containing fibers have been proposed to excite locus coeruleus neurons, thereby possibly linking elements of the SAM and HPA axis [131-132]. However, other reports suggest that the SAM and HPA axis paths are distinct in actual physiological stress [122].

Kvetnansky et al. and Pacak et al. have, respectively, suggested stressor-specific activation of the SAM and HPA axis pathways of stress response [133-134]. Further examination of the functional relationship between these two pathways of stress response is needed.

For the purposes of this work, the effects of stress on the noradrenergically-mediated

SAM pathway will be discussed in further detail.

In general, noradrenergic upregulation as a function of the stress response can be adaptive or maladaptive depending on the situation [123-124]. For example, if one is walking through the woods and encounters an angry bear, noradrenergic upregulation would be adaptive and likely to promote survival instincts through triggering appropriate physiological changes, increasing arousal and attention, and as will soon be explored, decreasing cognitive flexibility until the danger is gone [135]. By contrast, in a typical classroom, a student who has severe test anxiety may have maladaptive noradrenergic upregulation (e.g. anxiety, distraction), simply while trying to think of the answers to their exam questions. In sum, the effects of the stress response, particularly the role of the noradrenergic system, have been described above in terms of physiologic relevance. Yet, to fully understand the effects of the stress response on cognitive performance, such as in the example of the student taking their exam, a thorough exploration of the concept of cognitive flexibility will be required.

Cognitive flexibility is defined as the ability to search through a broad array of options to determine a solution and is the basis for creative thought, decision-making, and insight, among other functions [136]. Anatomically, cognitive flexibility suggests greater means of access to neural networks of the brain which contain information. Thus,

flexibility represents the ease and fluidity by which this search for answers to cognitive problems is completed [137].

Neural network models help to describe the typical anatomical substrate for cognitive flexibility. In neural network modeling, neuronal groups are represented by nodes in the network. In this framework, the nodes, or neuronal groups, are each more strongly activated the more closely they are associated with the primary node that is activated [138-139]. The strength of association between neurons is dependent on the processes of potentiation and depression. The degree of modulation of neural network associations is also dependent on the original state of the network. Thus, increased cognitive flexibility depends on stronger associations that allow for broader spread of activation through the network and access to further-reaching nodes, whereas the opposite would be expected in decreased flexibility.

Several neural networks have been studied with regards to cognitive flexibility, including visual processing, attention and memory networks [137, 140]. The semantic and associative networks however have been suggested to be noradrenergically modulated [110, 141].

Noradrenergic modulation of semantic networks has been examined in typical individuals undergoing stress[141] as well as non-stressed typical individuals [136]. These studies have revealed: a benefit of propranolol on anagram task performance relative to ephedrine in non-stressed typical individuals [136], impaired performance on cognitive flexibility tasks in healthy individuals exposed to stress [141], and decreased stress- related impairment of cognitive flexibility task performance with propranolol in healthy

individuals [110]. Individuals experiencing acute cocaine withdrawal also benefit on cognitive flexibility tasks from drugs that block noradrenergic activation [142].

Evidence suggests that there is decreased functional connectivity, the synchronous activation of different brain regions as measured by fMRI or fcMRI, in autism

[143], a finding which has been supported by several other studies[144-148]. Others have hypothesized that there is a decrease in “central coherence,” the ability to use context to understand one’s environment, which may help to explain cognitive flexibility impairments in ASD [149]. Altogether, the latter findings support restriction of the semantic and associative networks [150], in ASD-affected individuals relative to typical individuals [151].

Results of false memory recognition studies, performed by Beversdorf et al., continue to be consistent with ASD network-restriction [152-153]. In a false memory task using words, participants listened to a list of words that were all semantically related to a lure word which was not listed. Afterwards, subjects were asked to confirm whether or not particular words were on the list. The words presented for recognition included words on the list as well as the lure word. Typical individuals usually recognized a false memory for the lure word due to increased cognitive flexibility of semantic networks.

However, individuals with ASD were shown to be particularly accurate at discriminating the lure word from others on the list [152], likely due to restriction of their semantic networks. Adults with ASD also discriminated false lures in a related study examining a series of associated shapes and symbols [153], instead of words.

This finding, like others noted above, supports hyper-restriction of semantic networks [151], which means the spread of activation from primary nodes outward to further-reaching nodes is decreased due to weaker associations between further-reaching nodes (see Figure 3).

______

Typical activation of a semantic network:

Word presented Primary network node Associated nodes

SKY SKY BLUE

CLOUDS

SUN

MOON

STARS

Activation of a hyper-restricted semantic network in ASD:

Word presented Primary network node Associated nodes

SKY SKY BLUE

CLOUDS

SUN

MOON

STARS

Figure 3. Semantic network restriction in ASD. A schematic example of the semantic network associated with the word “Sky” in typical individuals compared to ASD affected individuals. Presentation of the word triggers activation of the primary network node in both typical individuals and ASD affected individuals who have hyper-restriction of the semantic network. However, the degree of subsequent activation of other associated nodes, is decreased in ASD affected individuals. Thicker lines indicate greater activation. Figure adapted from Beversdorf et al. [151].

In addition to having decreased cognitive flexibility, autism-affected individuals are well known to exhibit poor eye contact with others [9], as a manifestation of their difficulty with social relationships. Eye contact, here defined as mutual gaze, plays an important role in establishing social connections and also in communicating socially relevant emotions [154-157]. Research has shown that eye contact plays an important role in the development of interpersonal relationships [158].

Eye contact is known to be important in mother-infant bonding [159]. Typical infants look intently at, or fixate, other faces within moments of birth [160] and follow, or visually pursue, face-like stimuli when only two days old [161]. Since ASD diagnosis is usually confirmed after a child is about three years old, attempts to prospectively study characteristics of ASD-affected infants, including eye contact, have primarily involved high-risk infants who have an older sibling with ASD [162-163]. However, retrospective blinded reviews of home videotapes from early childhood and infancy [164-166] have observed decreased eye contact in infants and young children that were later diagnosed with ASD, irrespective of their initial risk factors. Thus, measures of eye contact hold both diagnostic and predictive value in ASD.

Some techniques that have been used to measure eye contact, both in typical individuals and in ASD, have included two-way mirror observation [167], physiologic studies [168-169], retrospective scoring of video footage [164, 166, 170], head-mounted eyetracking where the eyetracking camera is connected to a supportive base that is placed on the subject’s head [171-172], and remote eyetracking where the eyetracking camera is placed at a short, fixed distance from the subject (see Figure 4) [173-175]. The

most objective measures of eye contact have been determined by eyetracking technology, also called oculography.

Figure 4. Remote eyetracker. The illuminator (left) emits infrared light which is reflected from the subject’s eye and detected by the eye camera (right). The eye camera and illuminator are located at a fixed distance from subjects near a display screen .Image from ASL (Applied Science Laboratories, Bedford, MA).

Eyetrackers consist of two parts, the illuminator and the eye camera. The illuminator emits a small beam of infrared light directly toward the subject’s pupil. The beam then reflects off of the retina. The eye camera detects this retinal reflection to determine the location of the pupil. In addition to pupil detection, the eye camera then 20

detects the degree of corneal reflection that results as the beam exits the eye. By sampling the positions of these two points as often as hundreds of times per second, the eye movement monitor calculates the point of gaze of a subject to within a predetermined number of pixels on a display screen. When the number of consecutive sample values at a specific location reach an assigned threshold value, a fixation, also called ocular fixation, is recorded in the data log [176].

Numerous studies conducted using eyetracking technology in ASD-affected populations have reliably indicated decreased eye contact [163, 170-171, 177-189].

These findings are consistent with the findings of observational and video-rating studies reviewed earlier [164, 167, 190]. However, in contrast to earlier methods, the added benefit of eyetracking technology is that the eye contact can be objectively measured.

A study by Klin et al. used a head-mounted eye camera to measure the point of gaze of ASD subjects and controls when viewing naturalistic social situations on short clips from the movie, “Who’s Afraid of Virginia Woolf?” (1967) [191]. In that study, adolescent participants with ASD exhibited decreased fixation, or gaze, in the eyes regions of characters’ faces and increased fixation on the mouth regions and body regions [191].

This result was in contrast to controls who, as may be predicted, spent most of their time fixating the eyes region, relative to other regions on the videotape [191]. Similar findings that autistic individuals look to the mouth area of faces instead of the eyes support the work of Klin et al. However, Pelphrey et al. demonstrated increased eye fixations relative to the mouth within their ASD population, but found decreased eye and increased mouth

fixations in ASD relative to controls [177, 191-193]. Overall, both studies, along with many others have found decreased eye contact in autism [163, 170, 178-189].

Many possible explanations have been proposed for diminished eye contact in patients with ASD, including abnormal face processing as mediated by inappropriate methods of face processing [194], abnormal functional connectivity [148, 156], abnormality of the amygdalae [195-196], abnormality of the ventral temporal cortex

[197] and abnormality of the fusiform gyrus [198-199], which is thought to be involved in normal face processing [200]. Other problems that affect face viewing and processing in

ASD include inappropriate expression and recognition of emotions on others’ faces [201-

202]and possible increased salience of the mouth area [171, 203].

Recent evidence suggests that diminished eye contact in ASD is a result of physiological stress experienced when observing direct gaze [169]. In a Finnish study, twelve individuals with autism were connected to a skin conductance response monitor, which records changes in electrical conductance of the skin due to sweat, to measure physiological stress response [204]. Each subject was presented with twelve video images of people gazing directly toward the camera or averting their gaze [169]. While controls did not show a difference in physiological response to either type of video image, individuals with autism had stronger SCR to videos of straight gaze than to those of averted gaze [169].

This and other related studies suggest that diminished eye contact in ASD may be due to associated stress [169, 181, 205-206]. Stress is well-known to activate the

noradrenergic system, as has been discussed [96]. A putative agent that could decrease noradrenergic activation could reduce stress associated with eye contact in ASD.

Norepinephrine is also known to increase the signal: noise ratio as previously described [106]. Stress-associated noradrenergic activation then would increase activity of primary network nodes and decrease that of further-reaching nodes in cortical networks. It is thought that some neural networks in ASD are already hyper-restricted

[150], which suggests that even nodes that are closely associated with the primary node, are not activated as strongly. Thus, reducing noradrenergic activation, to increase spread of network activation, may also benefit cognitive flexibility in this population.

As noted, noradrenergic blockade through beta-adrenergic antagonists have shown some benefit for language and social areas in ASD [83]. An approach to pharmacologic treatment of ASD that addresses the core symptoms would be a novel one. Thus, in addition to examining cognitive flexibility task performance in typical individuals and ASD-affected individuals, previous work in this lab has also begun to characterize some benefits of propranolol in ASD.

Propranolol (see Figure 5) is a nonselective beta antagonist with both peripheral and central activity. Under the brand name, Inderal, propranolol is clinically prescribed for hypertension. It also has off-label use in essential tremor and situational anxiety.

Several studies have demonstrated a beneficial effect of propranolol on cognitive flexibility tasks in typical individuals [110] as well as in ASD [207].

______

Figure 5. Propranolol. Chemical structure of propranolol.

In an earlier study, performance of eighteen healthy subjects was measured for a series of cognitive tasks 45 minutes after administration of 40 mg propranolol or 25 mg ephedrine, a noradrenergic agonist, with a placebo condition as a control [136]. Of the three cognitive tasks measured – number series, shape manipulation, and anagrams – solution time for anagrams improved for the best anagram-solvers when they took propranolol as compared to the ephedrine condition [136]. This research suggested that performance on cognitive flexibility tasks may be modulated by the noradrenergic system.

Soon after, a related study was completed that measured anagram performance in typical individuals when administered propranolol, a central and peripheral beta-blocker; nadolol, a peripheral-only beta-blocker; or placebo [208], to examine whether effects on cognitive flexibility were related to peripheral effects. They found that propranolol

improved anagram performance, a task requiring cognitive flexibility, relative to nadolol

[208]. Further examination of the specificity of this novel finding was conducted. It was found that alpha-1-agonists [209] , benzodiazepine anxiolytics [210], and dopaminergic agonists [211], showed no effect on cognitive flexibility task performance.

Later studies found that the effects of propranolol were dependent on stress level, ability to solve problems when unstressed, and on the difficulty of the task [110,

141, 212]. First, Alexander et al. exposed typical participants to psychosocial stress via the Trier Social Stress Test (TSST). The TSST requires participants to sit in front of an audience panel who are all wearing white coats and who are instructed to avoid pleasantries with the participant. The panel spokesman then instructs the participant to give a public speech for five minutes on an assigned topic with less than five minutes to prepare. After five minutes, the subject is interrupted and asked to perform mental arithmetic, such as sequential two-digit subtraction, for five additional minutes [141].

Subjects performed the TSST twice, once with propranolol and once with placebo in a crossover double-blind design and were also administered cognitive flexibility tasks during brief interruptions in the course of their stressor tasks. Overall, in a sample of sixteen healthy individuals, Alexander et al. determined not only that psychosocial stress impairs a typical individual’s performance on cognitive flexibility tasks, but also that propranolol administration reversed this impairment [141].

Later, Kelley et al. showed that cognitive flexibility task performance was impaired in individuals undergoing acute cocaine withdrawal (1-7 days after last use)

[213] and that propranolol improved cognitive performance in this population [142]. It is

thought the propranolol benefited this group because they have upregulation of noradrenergic activation [142]. More recently, Campbell et al. established that healthy subjects benefited from propranolol primarily on tasks that were more difficult, requiring a broader search of semantic and associative networks. It was also observed in that study that baseline degree of performance on a task, in the absence of stress, played a role in determining whether propranolol would benefit a subject [110]. Campbell et al. also examined semantic and phonemic verbal fluency, reporting an improvement in phonemic verbal fluency but not semantic.

Kelley et al. also studied effects of propranolol on verbal fluency in individuals undergoing acute cocaine withdrawal. Verbal fluency refers to the generation of words in response to a cue. Two types of verbal fluency, semantic (words that are related to a category cue) and phonemic (words that start with the same letter as the cue) are usually used. Verbal fluency may be a measure of cognitive flexibility as well and in this study, phonemic verbal fluency increased with propranolol while semantic did not. However, this study only used one letter and one category as cues, whereas the task usually requires three trials of each type [142]. Campbell et al. found similar results using all three trials of the task with typical individuals given propranolol; namely that phonemic fluency increased but semantic fluency did not. However, they also found that for the individuals who struggled (the lower 1/3) a greater improvement in semantic fluency was observed [110]. This further supported their finding that individuals that have difficulty with a cognitive flexibility task benefit more with propranolol.

Thus, both typical individuals under stress or tackling difficult cognitive problems and individuals undergoing cocaine withdrawal demonstrate improved performance on some cognitive flexibility tasks, such as anagrams, with propranolol [110, 141-142]. A recent study demonstrated that individuals with ASD also benefit from propranolol when performing anagrams [207]. This finding is supported by unpublished work from this lab which demonstrated an effect of propranolol on functional connectivity in individuals with ASD. Functional connectivity has been previously shown to be decreased in ASD as mentioned above [143, 214]. The work by Narayanan et al. using fMRI, showed that propranolol increases functional connectivity in ASD [215]. Taken together, evidence suggests a possible role for propranolol in improving flexibility of the semantic and associative networks.

As individuals with ASD exhibit decreased eye contact from as early as infancy [9], much research has been devoted to examining face processing in autism, and many theories have been proposed to explain mechanisms for decreased eye contact in ASD

[169, 182, 199, 216-217]. Behavioral interventions to increase eye contact have been proposed [218-220] and some attempts at pharmacologic interventions in ASD have included measures of eye contact in the design of the study. However, to our knowledge, no study has demonstrated a pharmacologic method of increasing eye contact [221].

Moreover, performance on verbal problem solving tasks has been found to improve with propranolol in ASD but verbal fluency tasks have not been examined in this population. While phonemic verbal fluency appears to benefit from propranolol in typical individuals and cocaine withdrawal patients, semantic verbal fluency did not show the

same result in those groups. Limited data suggest a possible benefit of propranolol on semantic verbal fluency in typical individuals who have difficulty with the task. Struggling with a task is thought to be due to weaker associations between nodes of the semantic network, or decreased “semantic network flexibility” [110]. Since individuals with ASD are thought to have hyper-restricted semantic networks [151], they may possibly benefit on verbal fluency tasks from propranolol, in a similar manner as did the struggling typical individuals, through decreasing the signal: noise ratio and performing a broader network search.

In sum, a need exists for further examination of the effects of noradrenergic blockade by propranolol in autism. This is particularly important in areas of impairment such as language and social behavior, as these two core symptoms of ASD cannot currently be treated with pharmacologic agents. Thus, for the above reasons, we asked whether individuals with ASD, when administered propranolol, would improve verbal fluency and eye contact relative to placebo, and also whether any detected benefits were specific to ASD. We expected to see a specific improvement from propranolol in both verbal fluency task performance as well as eye contact for individuals with ASD.

CHAPTER 2: EFFECT OF PROPRANOLOL ON VERBAL FLUENCY IN AUTISM

SPECTRUM DISORDER (ASD)

INTRODUCTION

Originally described by Leo Kanner in 1943, autism is characterized by social impairments, communication impairments and stereotyped or repetitive behavior, which begin before age three [5, 9]. Asperger syndrome, also described by these characteristics, also retains language ability [9]. The term “Autism Spectrum Disorder (ASD)” refers to both of these conditions as well as others including Pervasive Developmental Disorder

(PDD) and PDD Not Otherwise Specified (PDD NOS), and groups them together [11].

Many theories have been put forth to define and explain the cognitive impairment observed in ASD. One of these suggests that ASD-affected individuals lack

“central coherence,” the ability to use context when interpreting one’s surrounding environment [149]. In keeping with this theory, researchers have also proposed that individuals with ASD have restriction of flexibility of access to semantic and associative

networks of the brain [151]. Other studies have supported this finding, such as a report that demonstrated that ASD-affected individuals do not increase word recall when contextual information is added [222] and that they have decreased clustering of semantic information in verbal memory [150]. Moreover, Beversdorf et al. found that

ASD-affected individuals performed better than typical individuals in a “false memory” task [152]. In this task, participants listened to a list of words that were all semantically related to a lure word that was not presented. Afterwards, when asked whether particular words were present on the list, typical individuals usually recognized a false memory for the lure word, while participants with ASD were more able to discriminate the lure word from the original list. This result was in spite of a typical performance on other cognitive tasks administered to the same ASD-affected participants [11]. This further suggests a decrease in semantic network flexibility in ASD.

Interestingly, fMRI research examining functional connectivity (fcMRI), which measures the synchronicity of function of associated brain regions, has recently suggested a possible neural substrate for the decrease in network flexibility. A study by

Just et al. has shown that functional connectivity between active regions of the brain is decreased during sentence comprehension tasks in ASD [223].

Some other studies suggest that individuals with ASD do use semantic and associative information to a degree, particularly during memory tasks, including another

“false memory” task [224]. However, performance on these tasks by ASD participants still lags behind typical performance. In light of these findings, a putative agent that could

impact both the semantic and associative networks might indeed prove beneficial to individuals with ASD.

Studies exploring the effects of noradrenergic agents on network flexibility with verbal problem solving have used anagram (word unscrambling) tasks. Anagram performance in typical individuals was improved with propranolol relative to ephedrine or nadolol; however, it was not significantly different from placebo [136, 208]. Later, it was found that propranolol benefited typical individuals primarily on more difficult tasks, likely due to the difficult problems requiring a wider network search and hence more network flexibility [110] In the same study, it was observed that propranolol impaired typical individuals performance when the task was easy [110]. In patients with presumed upregulated noradrenergic activity due to cocaine withdrawal, however, Kelley et al. found cognitive benefit even on the performance of easy tasks [142]

A report by Alexander et al. also showed that typical individuals without anxiety- related disorders who were presented with a psychosocial stressor had impaired performance on the anagram task, and also other cognitive tasks that require semantic network flexibility [141]. With propranolol, this stress-induced cognitive impairment was reversed [141].

Drugs that decrease noradrenergic system activity are often utilized to relieve anxiety and also for behavioral treatment in autism. In one case series, beta-adrenergic antagonists also provided benefits in areas of language and social behavior in ASD [83].

While some studies have proposed that there is increased noradrenergic activity in autism

[225-226], there may be numerous other explanations for these results [227].

Furthermore, there is no known pathology in the locus coeruleus, the brain region which contains noradrenergic neuronal cell bodies, in autism [228]. Regardless of how the noradrenergic system functions in ASD however, based on previous pertinent findings, it may be worthwhile to determine what types of network flexibility tasks benefit from noradrenergic blockade in this population [207].

In previous work, Beversdorf et al. tested individuals with ASD and controls with an anagram task, having given a single dose of propranolol or placebo prior to the task

[207]. With propranolol, participants with ASD were found to improve simple anagram task performance, while controls were impaired on the same simple anagram task [207].

These results suggested a benefit from propranolol in ASD that affected semantic and associative network flexibility. Previous work by Campbell et al. suggests benefits for verbal fluency in typical individuals given propranolol [110].

Therefore, we chose to further explore this benefit on network flexibility by examining the effect of propranolol on verbal fluency in ASD using both a phonemic verbal fluency task (where subjects have 30 seconds to make a list of words that all start with the same letter) and a semantic verbal fluency task (where subjects have 30 seconds to make a list of words that all fit in the same category). We expected that, relative to placebo, propranolol would improve performance on both types of verbal fluency tasks in

ASD subjects by affecting the flexibility of the semantic and associative networks.

METHODS

Subjects

Twenty-eight subjects participated in two test sessions that were between 24 hours and 2 weeks apart. At one visit, the subject was given 40 mg propranolol and at the other they were given placebo. Drug order was balanced in a crossover design and the drugs were given in a double-blinded manner. Fourteen subjects (10 males, 4 females) were high-functioning individuals with ASD (see Table 1). Subjects were deemed to be high-functioning based on average IQ scores over 80. All ASD subjects were clinically diagnosed and were also confirmed by the Autism Diagnostic Interview-Revised [229].

Fourteen other subjects were age and IQ-matched controls without any neurodevelopmental diagnoses (10 males, 4 females). All subjects had a Wechsler

Abbreviated Scales of Intelligence (WASI) IQ of at least 80 and were native English speakers, without other diagnosed language or learning disorders, and no subjects were taking any noradrenergic agents. All subjects were consented in accordance with the

Institutional Review Board at University of Missouri-Columbia.

Gender N

Female 4

Male 10

Age in years 18 .9 ± 2.9 (range 15 -23 years)

Clinical diagnosis N

Autism 7

Asperger’s 7

PDD or PDD -NOS 0

ADI -R Subscores mean score ± SD

Social Impairment 21. 14 ± 4. 45

(diagnostic cutoff = 10)

Communication impairment 15.57 ± 4. 57

(diagnostic cutoff = 8)

Repetitive and stereotypical activity 6.29 ± 2. 61

(diagnostic cutoff = 3)

Age of abnormality of development 3.43 ± 1.16

(diagnostic cutoff = 1)

TABLE 1. Demographic and diagnostic data for ASD group. Proportionately more male individuals with ASD were represented in the study than female individuals, in accordance with data that shows a male preponderance for ASD prevalence. Also, ASD group clinical diagnosis and mean subscores on the Autism Diagnostic Interview – Revised [24] for ASD subject group. All subjects were tested and met diagnostic cutoffs for ASD.

Procedures

Baseline heart rate and blood pressure were compared between groups using t- tests. Heart rate and blood pressure were also compared within subject before (time zero) and after propranolol (60 minutes and 150 minutes post-administration) as well as placebo using paired t-tests. All testing was completed after 60 minutes post-drug which is the timeframe previously utilized to yield cognitive effects in other studies of propranolol [110].

For the phonemic verbal fluency task, the subject was given an alphabetical letter verbally and told to generate words that started with that letter. For the semantic verbal fluency task, the subject was given a short topic verbally and told to generate words that fit into that topic. For both the phonemic and semantic verbal fluency tasks, each word generated in 30 seconds was recorded and the number of words was tabulated. Subjects were given three letters and three categories per visit, presented one at a time for every person. The drug order was balanced in a crossover design between drugs so that half of the group were given first visit tests with propranolol first, and the other half received the first visit tests with placebo first. During the second visit, each subject received the drug they had not taken before and completed the second visit tests. During the first visit, the letters were F, A, and S, and the categories were “Animals,” “Things to Wear,” and

“Vegetables.” At the second visit, the letters were P, R, and W, and the categories were

“Drinks,” “Things in the Kitchen,” and “Hobbies.” No letter or category was repeated for any subject. The average number of words over three trials in each task was calculated as the raw scores. Statistical analysis using two-way ANOVA to determine drug effect was

carried out using SPSS (SPSS, Inc., Chicago, IL). SPSS was also used to calculate post-hoc t- tests comparing semantic and phonemic verbal fluency task performance between propranolol and placebo.

RESULTS

Sample Characteristics

Average subject age (years + SD ) was 18.9 ± 2.9 (range 15-23 years) for ASD subjects, and 19.4 ± 2.0 (range 18-25 years) for controls. Average subject age was not significantly different between ASD subjects and controls ( p = .60).

WASI IQ scores [230] (score + SD ) were 103.9 ± 12.3 (range 84-127) for ASD subjects, and 108.1 ± 7.9 (range 93-121) for controls. There were no significant differences in WASI IQ scores [230] between groups (p = .30).

Hemodynamic response analysis

Heart rate and blood pressure (systolic as well as diastolic) at baseline did not significantly differ between subject groups at time zero or at 60-minutes or 150-minutes post-administration of drug. As expected, propranolol led to significant decreases between baseline and the time of testing in heart rate as well as blood pressure: systolic blood pressure decreased from 126.6 ± 13.1 to 115.4 ± 15.4 ( t(27) = 3.06, p = .005), diastolic blood pressure remained the same across all timepoints, and heart rate

decreased from 78.1 ± 12.4 to 68.1 ± 9.7 ( t(27) = 3.69, p = .001). The significant decreases in systolic blood pressure and heart rate continued to be present at the 150-minute post- administration timepoint: systolic blood pressure decreased further to 111.1 ± 16.6 ( t(27)

= 4.15, p = .0003) and heart rate decreased further to 60.8 ± 9.7 ( t(27) = 6.68, p =

.00000036).

No change occurred in blood pressure or heart rate (see Figures 6 and 7) with placebo during the first 60-minutes post-administration, although at the 150-minute timepoint post-administration of drug, a significant decrease in heart rate from 79.7 ±

15.3 to 70.6 ± 12.8 ( t(27) = 2.50 , p = .019) and a trend for decreased systolic blood pressure from 125.6 ± 11.5 to 120.0 ± 11.5 ( t(27) = 1.88 , p = .071) were observed, suggesting possible habituation to the testing environment during the placebo session.

Comparison in blood pressure and heart rate between subjects receiving placebo during the first-session and those receiving placebo during the second-session conditions revealed no significant differences, suggesting no habituation to the testing environment across sessions.

FIGURE 6. Blood pressure. The effects of propranolol on systolic and diastolic blood pressure over three timepoints in the experimental session using all participants. Error bars indicate standard deviation. Relative to placebo, propranolol significantly decreased systolic blood pressure from time zero to 60-minutes post-drug. This decrease continued until at least 150- minutes post-drug when the experimental session was concluded. No differences were found between propranolol and placebo on diastolic blood pressure. Diagnostic groups did not differ in hemodynamic response to propranolol.

FIGURE 7. Heart rate. Effect of propranolol on mean heart rate over three timepoints during the experimental session using all subjects. Error bars indicate standard deviation. Relative to placebo, propranolol produced a significant drop in heart rate from time zero through the course of the experimental session, as noted by measurements taken at 60-minutes post-drug and 150- minutes post-drug. Diagnostic groups did not differ in hemodynamic response to propranolol.

Behavioral analysis

The words verbally generated for each of the three phonemic and three semantic verbal fluency task trials were recorded for each participant at each visit. The average number of words for the phonemic verbal fluency set and the semantic verbal fluency set

for each participant’s two visits were tabulated (Table 1). Raw scores on these tasks were normally distributed (skewness ≤ 1.0 for all conditions). A 2 x 2 ANOVA revealed a trend for an interaction effect between drug and group (F(1, 26) =2.86, p = .10) as well as a significant main effect of drug ( F(1, 26) = 5.25, p = .03). There was no main effect of group.

Individual t-tests were then performed. Paired t-tests revealed an improvement in semantic verbal fluency task performance in the ASD group when given propranolol, relative to their placebo visit ( t(13) = 2.12, p = .027). No differences were observed for phonemic verbal fluency task performance in the ASD group between the propranolol and placebo condition. The control group showed no differences between drug groups in either task.

Comparison of the placebo condition between subject groups further revealed that the ASD subjects showed impairment in both phonemic ( t(13) = 2.92, p = .012) and semantic ( t(13) = 2.68, p = .019) types of verbal fluency, relative to the controls in the placebo condition (see Table 2 and Figures 8 and 11). While the ASD subjects continued to exhibit impaired phonemic verbal fluency even with propranolol ( t(13) = 2.15, p =

.051), their semantic verbal fluency impairment was increased to similar levels as controls in the propranolol condition ( t(13) = .069, p = .505). As noted, the controls showed no difference between drug conditions.

Further analysis showed that the improvement in semantic verbal fluency was driven primarily by the subjects who had a primary clinical diagnosis of Asperger’s syndrome as opposed to those with classic autism (see Figure 10). Subjects with

Asperger’s had a trend to increase number of words generated when under propranolol versus placebo (t(6)=1.824, p=.059). Individual score plotting for the semantic (see Figure

9) and phonemic (see Figure 12) verbal fluency tasks is also shown for both drugs and both ASD and controls. Score values can be found in Appendix A.

AUTISM CONTROL

Phonemic Semantic Phonemic Semantic

PLACEBO 7.5 ± 2.3 9.2 ± 2.6 9.6 ± 1.9 11.2 ± 1.6

PROPRANOLOL 7.8 ± 2.1 10.7 ± 3.3 9.4 ± 2.2 11.4 ± 2.4

TABLE 2. Mean number of words generated. Number of words generated (mean ± SD ) per verbal fluency task, distributed by drug and group (n = 14).

FIGURE 8. Semantic verbal fluency. Mean number of words generated in the semantic verbal fluency task by drug and group. Error bars indicate standard deviation. ASD subjects (n = 14) had impaired performance on the semantic verbal fluency task, relative to controls during the placebo condition. Propranolol administration improved the ASD group’s performance on the semantic verbal fluency task to comparable to controls. No difference in semantic fluency task performance was seen between drugs for the control group (n = 14).

(A)

(B) Figure 9. Individual scores on semantic fluency task. Average number of words generated in the semantic fluency task over three trials of 30 seconds each for A) autism (n = 14 )and B) controls (n = 14). Half of all subjects received the categories “Animals” “Things in the Kitchen” and “Vegetables” at the placebo visit while the other half of all subjects received the categories “Drinks” “Things to wear” and “Hobbies” at the placebo visit. During the propranolol visit the other three categories were used. Drug visits were in a crossover design.

Figure 10. Individuals with Asperger’s syndrome versus individuals with autism. Indivduals with a clinical diagnosis of Asperger’s syndrome ( n= 7) had a greater increase in the semantic fluency task than individuals with a clinical diagnosis of classic autism ( n = 7) ( t(6)=1.824, p=.059). Error bars indicate standard deviation.

FIGURE 11. Phonemic verbal fluency. Mean number of words generated in the phonemic verbal fluency task by drug and group. Error bars indicate standard deviation. ASD subjects (n = 14) had impaired performance on the phonemic verbal fluency task, relative to controls during the placebo condition. Propranolol administration did not improve the ASD group’s performance on the phonemic verbal fluency task. No difference in phonemic fluency task performance was seen between drugs for the control group (n= 14) .

(A)

(B) Figure 12. Individual scores on phonemic fluency task. Average number of words generated in the semantic fluency task over three trials of 30 seconds each for A) autism (n = 14) and B) controls (n = 14). Half of all subjects received the letters “F” “A” and “S” at the placebo visit while the other half of all subjects received the categories “P” “R” and “W” at the placebo visit. During the propranolol visit the other three letters were used. Drug visits were in a crossover design. 46

DISCUSSION

Previous work examining noradrenergic influences on verbal fluency in typical individuals found evidence that individuals who struggled the most with verbal fluency tasks showed benefit of propranolol, while others did not improve [110]. Subsequently, propranolol was shown to improve verbal problem solving in ASD even on simple anagram problems on which typical individuals did not benefit, by increasing flexibility of hyper-restricted networks in ASD [151, 207]. Based on these findings, we chose to examine whether verbal fluency would improve similarly in an ASD population. We tested fourteen individuals with ASD using a common verbal fluency task. We found that both types of verbal fluency, phonemic and semantic, were impaired in the ASD group in the placebo condition, relative to controls in the placebo condition. We also found that propranolol administration reversed this impairment only for semantic verbal fluency, but not for phonemic verbal fluency.

The ASD subjects had a greater improvement with propranolol on the semantic verbal fluency task than the controls did with propranolol. There was also a main effect of drug, driven primarily by the ASD subject group, as the control group showed no significant differences in performance on the tasks. This was also supported by a trend for an interaction effect between drug and group.

Verbal fluency performance in ASD has been examined in several other studies

[231-234]. Results from these studies generally agree that aspects of verbal fluency are impaired in ASD patients relative to typical individuals [231-234], which we also found. In

the current findings, semantic verbal fluency performance was impaired during the placebo condition and was increased in the propranolol condition to typical levels. Kelley et al. found similar results using cocaine withdrawal patients [142, 213].

While propranolol is thought to affect verbal fluency in cocaine withdrawal patients by suppressing ambient noradrenergic activation [142] it remains unclear whether similar increases in noradrenergic activation are found in ASD [225, 227-228].

Research in functional connectivity, defined as the synchronicity of activation of disparate brain regions [143], has suggested global underconnectivity with local overconnectivity in individuals with ASD [144]. Further, network models suggest that neural networks are hyper-restrictive in ASD [235-236]. Whether the impairment in verbal fluency performance reported in ASD is due to an abnormality of neurochemistry, functional connectivity, neural network structure, or a combination of these remains unknown.

However, a benefit of propranolol in cognition in ASD, possibly from the drug improving performance of hyper-restrictive neural networks [151] has been demonstrated in this study as well as in other studies [207]. Thus, these findings warrant further exploration of the benefits of noradrenergic blockade in ASD.

We also report a differential effect of propranolol on the two types of verbal fluency, phonemic and semantic in our ASD subpopulation. Specifically, semantic verbal fluency improved with propranolol while phonemic verbal fluency did not. Earlier work in this lab found improvement for phonemic verbal fluency but not semantic verbal fluency for typical individuals [110] as well as individuals experiencing cocaine withdrawal [142].

Patients experiencing cocaine withdrawal are thought to have restricted flexibility of

access to networks due to noradrenergic upregulation [142], which has not yet been clearly demonstrated in ASD [225, 227]. However, healthy individuals without neuropsychiatric diagnoses are known to benefit from propranolol when performing more difficult tasks of cognitive flexibility [110], whereas patients with ASD are known to benefit from propranolol when performing simple tasks of cognitive flexibility [207].

Campbell et al. suggests that observed benefits of noradrenergic blockade on cognitive flexibility task performance is dependent on the ability of the individual to solve the problem in the absence of stress as well as the difficulty of the task [110].

Noradrenergic activation increases the signal: noise ratio of cortical neurons while noradrenergic blockade decreases this ratio [135]. Increased signal with noradrenergic activation can be beneficial for learning and memory [110], however, noise that is present with decreased noradrenergic activation is thought to allow improved access to associative and semantic network nodes through increases in intrinsic associative activity of cortical neurons [135]. Altogether noise may actually allow further-reaching exploration of semantic networks when searching for a solution [141]. Both the semantic and phonemic verbal fluency tasks are known to require access to associative networks and thus may benefit from suppression of the signal: noise ratio through the action of propranolol.

In this study, ASD subjects specifically improved performance on the semantic verbal fluency task with propranolol and not the phonemic verbal fluency tasks. As noted, individuals with ASD are known to have decreased cognitive flexibility due to possible hyper-restriction of neural networks [151]. Thus, at baseline, broad searches of

the semantic and associative networks are limited in ASD individuals who are presented with cognitive flexibility tasks. This is supported by our findings that autism subjects given placebo demonstrated impaired verbal fluency performance relative to control subjects given placebo. If hyper-restriction of networks in ASD is relieved by propranolol, allowing expanded searches through further-reaching network nodes [110, 151], cognitive flexibility task performance will improve, which is also what was found in this study.

The differential effect on the two types of verbal fluency may be explained similarly. Individuals with ASD are thought to have hyper-restricted semantic and associative networks [207]. Norepinephrine increases signal: noise ratio of cortical neurons [106], which restricts the network. In contrast, propranolol may increase cognitive flexibility by reducing the signal: noise ratio of cortical neurons through noradrenergic blockade [110], which relieves the network. However the extent of the network involved in semantic verbal fluency may be broader than that of the network for phonemic verbal fluency due to the specific network content. As a result, propranolol may increase accessibility of the much broader semantic fluency network (word meanings) but may not affect the narrower phonemic fluency network (same letter) in this population with already restricted networks. Thus, even with relief of the phonemic fluency network, there is little improvement with propranolol, while an effect is seen with the semantic fluency network.

In addition, the effect of propranolol on semantic verbal fluency appears to be driven by the Asperger’s syndrome subgroup of our ASD participants, more so than the autism subgroup. Individuals with Asperger’s syndrome by definition have normal early

language development whereas individuals with autism by definition usually have some abnormality or delay of language development. Propranolol may have preferably increased accessibility for the subgroup that already has a broader semantic fluency network due to normal language development. As many current studies about ASD and autism often include individuals with Asperger’s syndrome as participants, further characterization of these and other ASD subgroups is needed to delineate specific differences between the subgroups, particularly in the area of language.

Norepinephrine is one of the primary components of the arousal mechanism. It is also known to impact cognition in several areas, including probabilistic learning, response inhibition, and extradimensional set-shifting (through action on the alpha-1 receptor)

[109]. The beta-adrenergic receptors, however, have been reported to affect flexibility of access to the semantic and associative networks [141]. In earlier work, effects of propranolol on neural networks during verbal problem solving were difficult to detect in typically developing individuals [110, 136, 208]. However, previous reports have also demonstrated a benefit of propranolol in typical adults who are exposed to a psychosocial stressor [141], cocaine withdrawal (which may be due to upregulation of the noradrenergic system) [142], healthy non-stressed individuals who are attempting to solve difficult verbal problems [110], and also ASD-affected individuals who are solving simple verbal problems [207]. A number of mechanisms have been proposed to explain cognitive impairment in ASD [144, 225, 227-228]. However, previous work from this lab indicates benefit of propranolol on hyper-restrictive neural networks in ASD [151].

In this study baseline heart rate and blood pressure did not differ between groups. This suggests that these results are less likely to be attributable to a higher baseline stress level in the ASD group relative to controls. However, other measures of stress response will need to be utilized to get a clearer picture of how stress regulation works in ASD. Age and baseline intelligence, as determined by the WASI IQ test [230] were well matched between the two groups and IQ scores for both groups reflected average intelligence. Therefore, the observed impairment of the ASD group on the verbal fluency tasks is unlikely to be due to an intelligence or age discrepancy between the two groups. Had intelligence or age been different between the subject groups, the amount of struggle on the tasks would differ and the effects of propranolol on cognitive performance might have differed between groups, since previous work has shown that the effects of propranolol on cognition are dependent on subject ability [110].

Future work should examine the effects of propranolol on a wider range of tasks assessing flexibility of access to semantic and associative networks. Characterization of multiple benefits of propranolol in cognition will provide further rationale for appropriate clinical examination of its putative role in conditions like ASD, cocaine withdrawal, and situational anxiety. Also future studies of this nature will need to include assessments of baseline attention and working memory in order to determine if attention levels and memory function influence the findings seen here. Also, as this study worked with higher functioning individuals with ASD, it remains unclear how or whether these results may apply to other subpopulations with ASD, such as low-functioning patients and young children. Since such groups are often difficult to examine in a controlled study, methods

to do so will need to be developed and tested so that more individuals with ASD can possibly benefit from these findings. At this time, clinical implications of these results cannot be fully determined, although evidence of propranolol’s benefits in ASD is accumulating. Indeed, some clinical benefit of beta-blocking drugs in ASD, primarily affecting language and social behavior, has been suggested through case studies already

[237] Large controlled clinical studies will need to determine whether these results relate to the putative clinical benefits of noradrenergic agents that have been previously reported in ASD [207, 237]. Future studies examining clinical relevance of this work should also involve thorough neuropsychological testing as well as exploration of other clinical benefits in ASD. Altogether, while further work is certainly needed, beta-blockade through propranolol shows promise as a possible future treatment in ASD.

CHAPTER 3: EFFECT OF PROPRANOLOL ON EYE CONTACT IN AUTISM SPECTRUM

DISORDER (ASD)

INTRODUCTION

Autism spectrum disorders (ASD) are characterized by social and communication impairments as well as stereotypical or repetitive behaviors [9]. Statistics suggest that the diagnosis of autism has increased significantly over the past decade [27]. It is currently suggested that from 3 to 6 children of every 1,000 will develop autism [28]. In fact, some have suggested rates as high as 1 in 150 children may be at risk [29]. While symptoms vary widely, autism can be a very disabling condition affecting everything from social life and employment to day to day life skills.

Characteristically, individuals with autism exhibit poor eye contact with others from an early age [9]. Recent evidence suggests that eye contact may actually be stressful to those affected by autism [169, 181]. Stress is well known to activate the noradrenergic system [96, 133]. Therefore, an agent that could reliably decrease the stress related to

eye contact by acting to block noradrenergic activation could be beneficial to those affected with ASD.

Propranolol produces noradrenergic blockade with central and peripheral nervous system effects. Through its action in the brain, propranolol has been found to decrease the impact of emotional stimuli in typically developed adults [238]. Propranolol was also found to decrease test anxiety and improve Scholastic Aptitude Test (SAT) scores in students confirmed to have stress-related cognitive-dysfunction [239]. Moreover, in a study of anxiety-prone individuals, propranolol was found to produce decreased anxiety during public speaking [240-241]. In a similar study of typical adults, propranolol reversed social stressor-associated cognitive impairment [141]. Finally, our lab has recently reported benefit of propranolol on cognitive flexibility tasks in ASD [207]. It is thought that propranolol improves performance of hyper-restrictive neural networks in ASD [151], possibly by decreasing the signal: noise ratio of cortical neurons [135]. This result is in accordance with other reports of beta-blockers improving language as well as social behavior in autism [237].

In addition to the cognitive effects of propranolol, since decreased eye contact in autistic individuals may be linked to stress and propranolol is known to decrease social stress, we proposed to conduct a pilot study to determine whether autism-affected individuals increased their social eye contact when given propranolol. Eye contact was measured using a common method of eye movement monitoring that detected consecutive points of gaze, or fixations, while the subject viewed short video clips [178,

191, 242-244]. This method reliably maps all of the locations on a display toward which

the subject gazes by measuring visual angle from detection of corneal reflection [176].

Klin et al. used this method to confirm that individuals with ASD had decreased fixation on the eyes region of movies depicting social situations, and increased fixation on the mouth, relative to controls [191]. We hypothesized that propranolol administration, through its action of decreasing the stress response, would lead patients with ASD to spend more proportionate time making eye contact, as measured by proportionate amount of fixations in the eyes region, compared to placebo administration. Controls were examined, in the same manner, to determine whether differences between propranolol and placebo groups within the autism subpopulation are specific to autism.

METHODS

Subjects

Twenty-eight subjects participated in two test sessions that were between 24 hours and 2 weeks apart. At one visit, the subject was given 40 mg propranolol and at the other they were given placebo. Drug order was balanced in a crossover design and the drugs were given in a double-blinded manner. Fourteen subjects (10 males, 4 females) were high-functioning individuals with ASD (see Table 3). Subjects were deemed to be high-functioning based on average IQ scores over 80. All ASD subjects were clinically diagnosed and were also confirmed by the Autism Diagnostic Interview-Revised [229].

Fourteen other subjects were age and IQ-matched controls without any

neurodevelopmental diagnoses (10 males, 4 females). All subjects had a Wechsler

Institutional Review Board at University of Missouri-Columbia. None of the subjects had any eye conditions that were uncorrectable with eyeglasses.

Procedures

Baseline heart rate and blood pressure were compared between groups using t- tests (see Figures 14 and 15). Heart rate and blood pressure were also compared within subject before (time zero) and after propranolol (60 minutes and 150 minutes post- administration) as well as placebo using paired t-tests. All testing was completed after 60 minutes post-drug which is the timeframe previously utilized to yield cognitive effects in other studies of propranolol [110].

Gender N

Female 4

Male 10

Age in years 18 .9 ± 2.9 (range 15 -23 years)

Clinical diagnosis N

Autism 7

Asperger’s 7

PDD or P DD -NOS 0

ADI -R Subscores mean score ± SD

Social Impairment 21. 14 ± 4. 45

(diagnostic cutoff = 10)

Communication impairment 15.57 ± 4. 57

(diagnostic cutoff = 8)

Repetitive and stereotypical activity 6.29 ± 2. 61

(diagnostic cutoff = 3)

Age of abnormali ty of development 3.43 ± 1.16

(diagnostic cutoff = 1)

Subjects were seated in a dark room facing a computer display screen with their head resting on a basic, unobtrusive chinrest. Seat height was adjusted to allow for maximum comfort.

Eye movements were recorded using an Eye-Trac R6 remote eye movement monitor with video head tracking (Applied Sciences Laboratories, Bedford, MA). Prior to the experimental session, the eye movement monitor was calibrated by having the participant fixate at nine points, representing the points of intersection on an equally-spaced 3 x 3 grid (e.g., upper-left, middle-center, bottom-right) (see Figure 13). A piecewise linear interpolation of the calibration points was then used to compute eye position. Stimuli consisted of 84 ten-second video-clips distributed over the two experimental sessions. Video clips included brief footage of 32 neutral human faces

(16M: 16F), 32 inanimate objects, and 20 dogs presented in pseudorandom order.

Specifically, the video order for each set of videos was generated initially using a computerized random number generator and then that generated order was preserved for experimental sessions. Video models were instructed to look directly into the camera and to maintain a neutral facial expression. Video models were also asked to cover their clothing with a solid blue drape and to remove distracting accessories. Filmed dogs were obtained from local obedience schools and were videotaped such that they were gazing at the camera. All video subject matter was filmed in front of a solid white backdrop. All video footage included a slow zoom-in effect to better capture attention of study participants [245]. Stimuli preparation methods were modified from those used by

Kylliainen and Hietanen [246]. An equal number of videos from each category were available at each experimental session and no videos were repeated between the two visits.

After calibration and before beginning video stimulus-presentation, subjects were provided written on-screen instructions to “View the images that follow.”

______

(A) (B)

FIGURE 13. Eyetracker calibration.. A) Calibration screen. Subject was asked to look at each circle while the computer recorded the location of the gaze. Based on the information obtained during calibration, the remainder of eye movement recording was mapped to an imaginary coordinate grid adjusted for the individual subject’s location and height. B) Center calibration screen. This screen was displayed between each video clip and the experimenter monitored the accuracy of the target as the subject looked at the center. If center fixation was off, the experimenter was able to switch to calibration mode directly and then resume the video stimuli.

Immediately before each video stimulus, the participant’s eye position was also sampled using the Eye-Trac monitor to ensure that they were fixated at the center of the display (see Figure 16). Fixation was grossly detected by the experimenter on a smaller, separate viewing screen to the side of the experimental display which showed the video stimulus with a target crosshair (an X) in the location where the subject was fixating.

Participants that failed to fixate correctly (within 2º of the fixation point) were recalibrated as above. With fixation confirmed, trial presentation proceeded. This method was used previously by Christ et al. [173-174]. A target fixation was recorded whenever the participant attended to a given location on the display for at least 50 ms.

Shorter fixations were excluded because they typically reflect simple saccade-related eye movements.

Areas of interest corresponding to the eyes, nose, and mouth of each face were designated a priori and fixation data was mapped onto the videos using EyeNal and

FixPlot programs (Applied Sciences Laboratories, Bedford, MA). Each video clip was converted to twenty .jpeg image frames, taken at time zero and every 500 ms thereafter, using a frame capture program (Video Snapshot Wizard), and these images were subsequently converted to bitmap form using a simple batch conversion program, for use with FixPlot.

To draw an area of interest (AOI), the bitmap file was opened in FixPlot, and attachment points for the upper-left (P01) and lower-right corners (P02) of the coordinate grid were provided (h1 = 9, v1 = 24, h2 = 241, v2 = 216). Then a rectangular select tool was used to draw the area of interest directly onto the bitmap. Defined boundaries for

the eyes region, nose region, and mouth region were established using earlier work by

Baron –Cohen et al. [247]. For example, the eyes region upper left and right corners were the model’s temples, just above the eyebrow, while the lower left and right corners rested just under the orbit of the eye. The rectangle defining the nose region included the bridge of the nose, the tip of the nose, and the nostrils. Finally the mouth region was defined at its upper border by the top of the vertical indentation above the upper lip called the philtrum, and at the lower border, the horizontal indentation where the chin begins to protrude (see Figure 14).

______

(A)

(B)

FIGURE 14. Areas of interest. A) An example of a female face image and B) an example of a male face image, both with the three AOIs drawn (rectangles) and the attachment points (PO1 and PO2) designated.

Once all three AOIs were drawn on all of the 640 bitmap images that resulted from the frame capture process, eyetracker data in the form of an .eyd file was processed. As mentioned, fixations were recorded only if they met criteria for a fixation

and the location of the fixation was dependent on accurate calibration of the eye movement monitor. A tagged code called an XDAT value was associated with each video clip and electronically sent to the eye movement monitor computer when that video was on display. Thus, in a single .eyd file (see Figure 15), 16 XDATS corresponding to different face videos were found. Since the two drug visits occurred on different days, two separate .eyd recordings were made for each subject.

The first step in eye movement data processing was to create fixation files, which reduced and organized overall raw eye data to a series of recorded fixations. Fixation files were created independently for each of the individual bitmap images that had AOIs drawn on them. This was to account for the slight differences in face region location due to the zooming motion of the video clip over the course of its timeframe. Once the fixation data was extracted into fixation files for each of the subsections of each video, batch fix sequence files were created, which compiled fixation data and AOI images to calculate which fixations were located in particular AOI regions and for how long they remained there. This was visualized using FixPlot to confirm accuracy (see Figure 16).

video Pupil CR Horiz_gaze Vert_gaze field# total_secs recogn Recogn XDAT Coordinate Coordinate 7934 62813.53235 TRUE TRUE 130 131.699997 134.8000 031 7935 62813.54903 TRUE TRUE 130 131 131.9000092 7936 62813.56572 TRUE TRUE 130 131 127.5 7937 62813.5824 TRUE TRUE 130 130.800003 126.7000046 7938 62813.59908 TRUE TRUE 130 130.800003 126.5 7939 62813.61577 TRUE TRUE 130 131.100006 125.7000046 7940 62813.63245 TRUE TRUE 130 131.100006 125.7000046 7941 62813.64913 TRUE TRUE 130 131.300003 124.5999985 7942 62813.66582 TRUE TRUE 130 131.199997 124.2000046 7943 62813.6825 TRUE TRUE 130 131.5 124.0999985 7944 62813.69918 TRUE TRUE 130 131. 100006 123.8000031 7945 62813.71587 TRUE TRUE 130 130.900009 124.5 FIGURE 15. Sample .eyd file. A 200-ms section of .eyd file. Video field refers to the computer-designated code for each sampling attempt for eye location. Total seconds refers to the time of day, designated in total number of seconds since midnight. Pupil and corneal reflection (CR) recognition columns state “TRUE” if the monitor is able to detect them and “FALSE” if they are undetectable. XDAT is the code number given to the video clips to identify when in the .eyd file they were displayed. Horizontal gaze coordinate and vertical gaze coordinate indicate pupil position as calculated by corneal reflection.

(1A) (2A)

(1B) (2B)

(1C) (2C) FIGURE 16. Visualizing eye data. Examples of visualized datapoints for a video of a female face (1a, b, c) and a video of a male face (2a, b, c). Example image frames shown here were taken from the A) beginning (0ms-500ms), B) middle (5000ms-5500ms), and C) end (9500ms-10000ms) of each respective video although in actual analysis all twenty frames were used for each video’s data. Rectangles delineate the predetermined areas of interest (AOIs) of the eye region, nose region, and mouth region. Circles indicate recorded fixations and the size of the circle represents the duration of that particular fixation. Dashed lines show transitions between fixation points. Circles and lines together map the points of gaze and transitions between them for all ASD subjects who viewed these videos. Gray lines and circles indicate subjects who took placebo when they viewed these images, and black lines and circles indicate subjects who were on 40mg propranolol.

Finally, the numerical data for each fixation sequence file was compiled using a standard spreadsheet and, for each drug group, a series of mean values was calculated.

SPSS (SPSS, Inc., Chicago, IL) was used to calculate post-hoc t-tests comparing semantic and phonemic verbal fluency task performance between propranolol and placebo.

Further parameters were calculated based on these raw data, including total time on

AOIs, amount of time spent off of the AOIs and proportionate time spent on the eyes (see

Table 4). Control data analysis was conducted using these same methods. Once ASD and control groups were analyzed, 2x2 ANOVA was carried out to determine whether there was a drug by group interaction. Sub-analysis of Asperger’s and autism groups was not conducted as raw eyetracker data was grouped per each video stimulus after analysis and not by individual subject.

Mean parameter calculated Value represented Amount of time spent on the eyes Average amount of time in seconds, per video clip, that subjects looked at the eyes region Amount of time spent on the nose Aver age amount of time in seconds, per video clip, that subjects looked at the nose region Amount of time spent on the mouth Average amount of time in seconds, per video clip, that subjects looked at the mouth region Amount of time spent off of The length of the video (10s) minus the designated AOIs sum of the amounts of time spent on the eyes, nose and mouth in seconds. Total time spent on AOIs The average amount of time in seconds spent, as a sum, on the eyes, mouth and nose. Proportionate fixation time o n the eyes Percentage of recorded fixations devoted to the eyes region. This value differs from percentage of total time spent on the eyes, because total recorded fixation time is less than the full length of the videos due to exclusion of gazes that do not meet minimum fixation criteria and blinking. TABLE 4. Terminology. Values represented by calculated parameters for eye contact data.

RESULTS

Sample Characteristics

Hemodynamic response analysis

Heart rate and blood pressure (systolic as well as diastolic) at baseline did not significantly differ between subject groups at time zero or at 60-minutes or 150-minutes post-administration of drug (see Figures 17 and 18). As expected, propranolol led to significant decreases between baseline and the time of testing (60-minutes post-drug) in heart rate as well as blood pressure: systolic blood pressure decreased from 126.6 ± 13.1 to 115.4 ± 15.4 ( t(27) = 3.06, p = .005), diastolic blood pressure remained the same across all timepoints and heart rate decreased from 78.1 ± 12.4 to 68.1 ± 9.7 ( t(27) = 3.69, p =

.001). The significant decreases in systolic blood pressure and heart rate continued to be present at the 150-minute post-administration time-point: systolic blood pressure

decreased further to 111.1 ± 16.6 ( t(27) = 4.15, p = .0003) and heart rate decreased further to 60.8 ± 9.7 ( t(27) = 6.68, p = .00000036).

No change occurred in blood pressure or heart rate with placebo during the first

60-minutes post-administration, although at the 150-minute time-point post- administration of drug, a significant decrease in heart rate from 79.7 ± 15.3 to 70.6 ± 12.8

(t(27) = 2.50 , p = .019) and a trend for decreased systolic blood pressure from 125.6 ±

11.5 to 120.0 ± 11.5 ( t(27) = 1.88 , p = .071) were observed, suggesting possible habituation to the testing environment during the placebo session. Comparison of blood pressure and heart rate between subjects receiving placebo during the first-session and those receiving placebo during the second-session conditions revealed no significant differences, suggesting no habituation to the testing environment across sessions.

FIGURE 17. Blood pressure. The effects of propranolol on systolic and diastolic blood pressure over three timepoints in the experimental session using all participants. Error bars indicate standard deviation. Relative to placebo, propranolol significantly decreased systolic blood pressure from time zero to 60-minutes post-drug. This decrease continued until at least 150- minutes post-drug when the experimental session was concluded. No differences were found between propranolol and placebo on diastolic blood pressure. Diagnostic groups did not differ in hemodynamic response to propranolol.

FIGURE 18. Heart rate. Effect of propranolol on mean heart rate over three timepoints during the experimental session using all subjects. Error bars indicate standard deviation. Relative to placebo, propranolol produced a significant drop in heart rate from time zero through the course of the experimental session, as noted by measurements taken at 60-minutes post-drug and 150- minutes post-drug. Diagnostic groups did not differ in hemodynamic response to propranolol.

Eyetracking analysis

Mean amount of time spent on each region of the videos (eyes, nose, mouth, and time off of AOIs) and proportionate time spent on the eyes were calculated from fixation data. All fixation data values were normally distributed (skewness ≤ 1.0 for all 72

conditions). A 2 x 2 ANOVA revealed a trend for an interaction effect between drug and group (F(1, 124) =3.88, p = .051) as well as a significant main effect of drug (F(1, 124) =

21.88, p = .0001). There was no main effect of group.

Individual t-tests were then performed to further examine the observed drug effects. Between the propranolol and placebo condition, no differences were observed in the total amount of time blinking for either group. Further, no differences were observed in time off the AOIs where subjects were recorded looking at display areas outside of the three salient face regions mentioned. The ASD and control group spent similar time fixating the eyes in the placebo group, ( t(26) = .984, p = .167). Under propranolol, the

ASD ( t( 13) = 2.472, p = .014) and control ( t( 13) = 3.278, p = .003) groups both increased their amount of time on the eyes significantly from placebo, although controls had a trend for more time on the eyes than ASD in the propranolol condition (t(26) = 1.327, p =

.098). The time that both groups spent looking at the nose appeared to decrease with propranolol, with controls reaching significance ( t(13) = 2.512, p = .013). The time that both groups spent looking at the mouth appeared to decrease with propranolol as well, with ASD reaching significance ( t(13) = 3.852, p = .001) (see Figure 19).

The ASD and control groups were similar in their total amount of time fixating facial features (the sum of eyes, nose, and mouth) in the placebo condition (t(26) = 1.187, p = .123). The control group improved in their total amount of time fixating facial features, as defined by the AOIs, with propranolol ( t( 13) = 2.837, p = .007). The ASD group given propranolol did not differ from the placebo condition in this parameter. In the placebo condition, the proportion of recorded fixations on the eyes was significantly

greater in the ASD group than in the control group ( t( 26) = 2.992 , p = .006). Both groups had significantly increased proportion of recorded fixations on the eyes with propranolol relative to the placebo condition (Autism: t(13)= 2.767, p = .008; Controls: (t(13)= 2.767, p=.008) (See Table 5).

AUTISM CONTROL

Percentage Total time on Percentage Total time

time on eyes facial time on eyes on facial

(%) features (%) features

(seconds) (seconds)

PLACEBO 67.54 ± 11.77 4.47 ± 0.54 56.52 ± 17.70 4.68 ± 0.66

PROPRANOLOL 73.56 ± 7.93 4.57 ± 0.62 70.06 ± 12.47 5.27 ± 0.85

Table 5. Proportionate time on eyes and Total time on AOIs by drug and group. Proportionate time on the eyes indicates the percentage of total fixations (eyes, mouth, nose, and off of AOIs) for each video that were on the eyes region. Total time on AOIs indicates the average total fixation time for all three facial features per video (eyes, mouth, and nose regions) and does not include fixation time off of AOIs. The ASD and control groups were similar in their amount of time fixating total salient facial features with placebo (t(26) = 1.187, p = .1229). The control group improved in their amount of time fixating salient facial features with propranolol ( t( 13) = 2.837, p = .007). In the placebo condition, the proportion of recorded fixations on the eyes was significantly greater in the ASD group than in the control group ( t( 26) = 2.992 , p = .006). Both groups had significantly increased percent of recorded fixations on the eyes with propranolol (Autism: t(13)= 2.767, p = .008; Controls: ( t(13)= 2.767, p=.008).

Figure 19. Amount of fixation on individual facial features by group and drug. Comparing the effects of propranolol and placebo on the amount of time ASD and control subjects spent per 10-second video fixating on each region. The ASD and control group spent similar time looking at the eyes in the placebo group, ( t(26) = .984, p = .167). Under propranolol, the ASD ( t( 13) = 2.472, p = .014) and control ( t( 13) = 3.278, p = .003) groups both increased their eye contact significantly from placebo, although controls increased more than ASD. The time that both groups spent looking at the nose appeared to decrease with propranolol, with controls reaching significance ( t(13) = 2.512, p = .013). The time that both groups spent looking at the mouth appeared to decrease with propranolol as well, with ASD reaching significance ( t(13) = 3.852, p = .001).

Figure 20. Proportionate time spent on the eyes. Mean percentage of all fixations (eyes, mouth, nose, and off AOIs) in which eyes were fixated. In the placebo condition, the proportion of recorded fixations on the eyes was significantly greater in the ASD group than in the control group (t( 26) = 2.992 , p = .006). Both groups had significantly increased percent of recorded fixations on the eyes with propranolol (Autism: t(13)= 2.767, p = .008; Controls: ( t(13)= 2.767, p=.008).

DISCUSSION

Previous literature has found that individuals with autism viewing videoclips of social situations while being monitored by an eyetracking device spent significantly less time on the eyes region of faces than controls and significantly more time on the mouth

[191]. Other research has suggested that decreased eye contact in ASD is due to increased physiological stress [246], which is well known to be associated with activation of the noradrenergic system. Our data did not detect impairment of eye contact among our ASD group versus controls. However both groups appeared to have increased eye contact with propranolol versus placebo suggesting a nonspecific effect of the drug.

The autism group appeared to spend less total time fixating the AOIs than the controls in the propranolol condition. This appears to be driven by the increase in eye fixation by controls given propranolol. Using a larger sample size in future studies of this nature may further clarify whether there is impairment of fixation on salient facial features in ASD. Another possible reason for this result may be due to increased saccadic movements making fixations less detectable in the ASD group [249]. Moreover, the ASD group may have increased fixations outside of AOI regions due to decreased attention

[250].

Both groups had increased time spent on the eyes with the propranolol versus placebo, with the controls showing a greater improvement in time spent on the eyes. The autism group, however, showed a greater percentage of fixation time spent on the eyes region with propranolol. A potential reason for this finding includes that the young adults

with ASD whom we tested have learned over time to look at other people “in the eyes.”

This may be due to persistent reminding, ABA therapy or other social skills therapy experiences as a youth, experiences which we did not exclude during subject recruitment.

Possibly due to these circumstances of growing up with ASD there may be a learned, and possibly exaggerated, emphasis on eye contact in the subpopulation we studied [251-

255].

Both controls and ASD subjects improved in eye fixation with propranolol. Part of our rationale for this study was that individuals with autism may relieve their stress associated with eye contact when given propranolol, and thus have better fixation on the eyes. However, participating in a research study may present with its own stressors for a control subject, including concerns about study performance.

Individuals with ASD are thought to spend more time fixating the mouth than controls [171, 203, 217]. Although other studies have found increased mouth-gaze in ASD

[191, 202], it remains unclear whether our autism group had a true difference in their mouth-gaze versus controls due to our small sample size. Further studies of this kind should use larger groups when attempting to detect a difference in mouth-gaze between these groups.

In our ASD results we showed a greater amount of eye region fixation relative to the nose or mouth regions, even with placebo. Klin et al. showed opposite baseline amounts of eye contact in their study employing a common movie to depict naturalistic social situations. Namely, mouth fixation was greatly increased compared to eye fixation.

However this may be a function of added mouth salience due to the presence of speech

depicted in their social stimuli. Mouth fixation amount in that study may also be different from our data due to a possible difference in the severity of autism symptoms between the two studies. Klin et al. reported that increased fixation on the mouth was correlated with increased social competence scores, while amount of eye fixation was found to be unrelated to degree of social competence [171]. Others have found similar results [306].

We did not use a measure of social competence in our study. Further examination of the effects of propranolol on social competence in conjunction with fixation on salient facial features, such as the eyes and mouth in ASD, will be necessary to determine whether propranolol affects mouth fixation in ASD and whether social competence is a factor in such an effect.

Several eyetracking studies which have found decreased eye contact in ASD have used photographs [177, 248] or video images [169, 171] of human faces as stimuli. The question remains, are the responses these stimuli elicit an accurate reflection of eye contact during a face-to-face encounter? With current technology, eyetracking remains the least obtrusive and most objective method of following eye movements. Head- mounted eyetracking allows the subject to wear a headpiece that tracks gaze in more natural settings. Further study of eye contact in ASD should include studies using head- mounted eyetracking to assess naturalistic levels of gaze.

Our stimuli, videos of neutral faces, incorporated a degree of motion through a zoom effect to create a more realistic representation of social encounters [245-246].

Motion in film, as well as in real life, attracts attention from the viewer [245, 256-258].

Making eye contact with video stimuli might be considered less stressful than in actual

social situations. However, this may not be the case in ASD, as is suggested by the results obtained in previous work using videoclips with skin conductance response in autism

[169]

Cognitive benefits of propranolol have been previously reported in ASD populations as well [207, 237]. Earlier work has demonstrated a benefit of propranolol in typical adults who are exposed to a psychosocial stressor [141], cocaine withdrawal

(which may be due to upregulation of the noradrenergic system) [142], healthy non- stressed individuals who are attempting to solve difficult verbal problems [110], and also

ASD-affected individuals who are solving simple verbal problems [207]. Recent work in this lab has demonstrated a benefit of propranolol in verbal fluency, particularly semantic fluency (Ch. 2). Here we report that propranolol benefits eye contact in ASD as well as in controls. In keeping with the findings of cognitive data and the suggestion that propranolol may benefit cognitive flexibility in individuals with ASD by decreasing restriction in semantic and associative networks [151], perhaps a role for the noradrenergic system in visual processing and social networks should be explored. With several benefits suggested for propranolol in ASD, further characterization of these benefits will be needed.

In this study, baseline heart rate and blood pressure did not differ between groups. This suggests that our results are less likely to be attributable to a differing baseline stress level between the ASD group and controls. Although we did not include any other measure of physiologic stress, noradrenergic blockade was effective at improving eye contact. Future ASD research in this area should incorporate other

measures of stress response to obtain a clearer picture of how stress regulation works in eye contact.

As noted, reports have consistently confirmed decreased eye contact in ASD relative to typical individuals [164, 171, 259-260]. Some behavioral approaches to increase eye contact in autism have been examined, particularly as potential early interventions [170, 218-220, 252], and some pharmacologic agents have been examined without effect on eye contact [221]. However, propranolol may be among the first pharmacologic agents to be studied in this manner and also to show a putative benefit to eye contact, although the effect is nonspecific to autism.

Due to the increasing number of people affected by autism [28-29], and the social impairments inherent in this condition [9], it is becoming more important to study possible ways improve any aspect of social interaction in autism-affected individuals. Eye contact is an important aspect of social function. With further examination, propranolol may be one of the first treatment strategies developed in ASD that may improve eye contact, and by extension social behavior, as well as possess benefits for other areas of core impairment in autism.

Pertinent next steps will include attempting similar eyetracking studies in a larger group of ASD subjects, including analysis of subgroups of ASD participants (i.e. Asperger’s syndrome vs. autism). If further positive results are found, it will be important to conduct larger clinical studies of propranolol examining its accumulated benefits to determine if it is an option for core symptom treatment in autism. If so, a thorough pre-clinical examination of effects of sustained dosing in ASD will need to be carried out. Studies of

safety and efficacy will also be important in ASD populations. Further, examination of propranolol’s effects on eye contact in low-functioning individuals as well as determination of early intervention strategies involving propranolol will be needed. It will be interesting to see how the effect of propranolol varies with degrees of baseline eye contact.

CHAPTER 4: DISCUSSION

Core symptoms in ASD include communication and language impairments as well as stereotypical behavior [9]. In this work we have suggested some effects of beta- adrenergic blockade in ASD. Propranolol improved semantic verbal fluency in ASD specifically, but not in controls. Propranolol also had an effect of increasing eye contact in both ASD and controls. When considered in combination with other benefits of propranolol in ASD, these results may suggest an approach to treating core symptoms of autism, including those of language and social behavior reported in a short series of case studies [237]. Indeed, the findings that propranolol increases semantic verbal fluency and also improves verbal problem solving are related to aspects of language, while increased eye contact may have implications for social behavior.

Altogether, the results obtained here contribute to growing evidence for propranolol to benefit key areas of impairment in ASD. In Chapter 2, where propranolol improved semantic verbal fluency in ASD, a main effect of drug, driven by significant differences in the ASD group relative to unchanged controls, was observed as well as a trend for an interaction effect of drug and group. In Chapter 3, where propranolol

increased eye contact in ASD, although nonspecifically, a main effect of drug was also observed. These results are consistent with other findings from this lab suggesting a range of benefits of propranolol in ASD [207, 215].

The proposed cellular mechanism for improvement in semantic verbal fluency in response to propranolol is thought to be a resulting decrease of the signal: noise ratio in cortical neurons [261]. By increasing relative noise, which represents background associative activity, more diffuse activation of the semantic network is possible [110]. The mechanism by which propranolol is thought to produce an increase in eye contact may occur by decreasing associated stress, through blockade of the noradrenergic system.

Eye-contact associated stress has been measured in other studies of gaze in ASD [169,

181]. Our physiologic measures of stress (blood pressure, heart rate) did not indicate baseline differences between our autism and control groups and no other assessment of stress was used. Therefore, further exploration of stress regulation in ASD will be needed to determine how stress actually affects eye contact in this population. Other stress regulation measures to consider for an experiment of this nature include more frequent or continuous heart rate and blood pressure monitoring, skin conductance response [246,

262], salivary cortisol levels [263-267], urinary catecholamines and metabolites [226, 268-

270], pupillary diameter [185], and electromyogram (EMG) of the corrugator facial muscle

[271].

Another important consideration for future work of this kind is the influence of previous treatment and therapy in a young adult ASD population and the strong impact of intervention on outcome in ASD [253, 272-274]. We recruited our subjects as young

adults from an autism treatment center. As a result, it is likely that our ASD subjects had been treated for some time and may have been far from their baseline at diagnosis when they participated in our study. The ADI-R score which was used to confirm clinical diagnosis would primarily reflect a potential subject’s early developmental conditions and may not factor in the latter influences of outside treatment or therapies. However, the

ADOS which measures current presentation of autism symptoms was not used in this study [275]. Previous behavioral therapy may improve eye contact from baseline [253-

255]. Therefore, other studies of this kind should also incorporate both the ADI-R as well as the ADOS in addition to clinical diagnosis to fully characterize subjects with ASD. While the ADI-R would provide information about baseline function and diagnosis, the ADOS may indicate the degree of recovered function through therapy in a young adult population with ASD, such as the one used for these studies.

Functional connectivity in ASD, measured with fMRI, has been shown to be decreased at baseline [143] and increased by propranolol [215]. Studies of this type in autism have also found abnormal functional connectivity involving the fusiform gyrus during face processing in ASD [148]. Face processing is an important factor in eye contact. Further, abnormal activation of the fusiform gyrus, also studied with fMRI, is one of the anatomic mechanisms thought to underlie decreased eye contact in ASD [146, 199,

276]. Some report that decreased eye contact in autism is associated with abnormal activation of the fusiform gyrus while increased activation of the amygdalaee is associated with increased eye fixations, suggesting a neural mechanism for heightened physiological arousal during eye contact in autism [277]. The fusiform gyrus is thought to

be where face processing normally occurs in typically developing individuals as well as in autism [198, 200, 248]. However, some reports suggest face processing in autism occurs outside of the fusiform gyrus [199]. However, despite multiple studies examining autism and even ASD [197, 278-279], few studies have examined activation of the fusiform gyrus in face processing using distinct high functioning autism or Asperger’s syndrome subpopulations [195].

Some reports show that ASD individuals who have Asperger’s syndrome (AS) or have high-functioning autism (HFA) have differential neural activation than controls when watching fearful faces, which does not produce the expected activation of the amygdalaee [195]. In a similar finding, Baron-Cohen et al. found that individuals with autism did not activate the amygdalaee in response to “mentalistic references to the eyes” while controls did have amygdalaee activation [280]. One possible neuroanatomic reason for such abnormalities may be the presence of small densely packed nuclei in the medial aspects of the amygdalaee which are a consistent pathologic finding in individuals with autism [281].

How might the noradrenergic system become involved in these structural differences? One fMRI study describes a mutation (val158met) in the gene for COMT

(Catechol-O-methyltransferase) which is one of the two proteins involved in the degradation of catecholamines, including norepinephrine. An increase in the number of val158met alleles was positively associated with an increase in face processing of aversive stimuli, particularly in the bilateral fusiform gyri [282]. Further, Cahill et al. have reported

that propranolol decreases the impact of emotional memories, a function that is mediated by the amygdalaee, in typical individuals [283].

The differences in fusiform gyrus and amygdalaee activation in autism are still not fully characterized. However, it is clear that these structures are not functioning normally in individuals with autism. Also, they may function in a differential manner between subgroups of ASD, such as autism versus Asperger’s syndrome. Therefore, it is possible that we had recruited an ASD subpopulation that did not produce adequate physiological arousal and resulting impairment of eye contact due to structural differences. There is also the possibility that some aspect of our video stimuli was perceived as aversive or uninteresting and influenced our eye contact findings. However, this is less likely since our stimuli were carefully modeled after those used in a previous study of skin conductance response with direct and averted gaze [246]. In this latter study though, eyetracking was not used, so there is potential for incompatibility between the stimuli and the procedure to affect our results.

While follow up imaging studies will need to be done to characterize the amygdalaee function in each subgroup of ASD, the data demonstrated here suggest a strong need for an understanding of the autistic brain. In addition to the neuropathology and functional neuroanatomy of this condition, having a thorough understanding of the neurochemistry of ASD will be important to future treatment development. As noted in the introduction, currently one drug in the United States is FDA approved for autism treatment, Risperidone, and it predominantly targets behavioral aggression symptoms through its effects as a dopamine antagonist with some anticholinergic side effects [284-

285]. In our studies, noradrenergic blockade is thought to mediate the effects seen. Yet, the inherent status of the noradrenergic system in ASD is not fully understood. Although some have suggested that individuals with ASD have increased arousal, thought to be noradrenergically-mediated [286], current evidence does not clearly suggest noradrenergic dysregulation in ASD [227-228, 287-288].

Irrespective of baseline levels of norepinephrine in ASD, blocking norepinephrine release through action at a postsynaptic receptor appears to show some benefits as we expected as in the cognition study, and other benefits that were slight variations on our expectations, as in the eyetracker study. Further, in experimental sessions for both studies performed here, the expected hemodynamic response to propranolol was observed. As noted above, propranolol decreased systolic blood pressure and heart rate over the course of the 150-minute testing session significantly more than did placebo.

The presence of the hemodynamic response confirms that the administered propranolol was active during the experimental session and that beta-blockade was successful in these studies. Therefore, beta-blockade in this study is sufficient to produce hemodynamic changes in addition to its observed effects on cognitive flexibility and eye contact. However, further examination of noradrenergic system function in ASD is needed.

Other neurotransmitters may play a role in our results than simply norepinephrine. One particular example is serotonin. Many individuals with autism have hyperserotonemia, suggested to be due to increased platelet release of serotonin [289].

Platelet serotonin does not cross the blood brain barrier to influence brain function,

however [290]. Propranolol, which does have central activity, is a partial agonist of some serotonin receptors [285]. Although the role of serotonergic binding of propranolol remains unclear, the potential exists for serotonin to have influenced our findings in part.

Thus it will be important to further investigate the serotonergic milieu in ASD as well as inherent noradrenergic system function.

In addition to the neurochemistry of ASD, much is being learned today about the genetics of ASD. Some genes to consider exploring in light of our findings include the beta-2 adrenergic receptor gene (ARB2) which has also been associated with autism

[291], the serotonin transporter gene (including the associated polymorphism 5HTTLPR) which pilot work in our lab suggests a possible association with prenatal stress in autism,

[289, 292], and the genes for the 5HT1D [293] and 5HT-2 receptors [294] which are serotonin receptors that are both thought to bind propranolol.

However, other genes to examine further based on current literature are the genes for neuroligins and neurexins which have been implicated in autism [40-42, 45,

295-297]. These genes for synaptic adhesion proteins of neurons are critical to the proper “hardwiring” of the brain during development. Neuroligins and neurexins have been linked to autism and a small percentage of individuals with ASD may carry a mutation for a neuroligin according to associative studies [298-299]. One study by Cheng et al. explored the association of cholinergic nicotinic receptor subunits with neurexin 1 beta [295]. As research in autism genetics progresses, studies should continue to explore the gene-gene interactions of neuroligin and neurexin genes with the genes and gene products for neurotransmitter synthesis, metabolism, receptors and transporters.

Moreover, defective synaptic adhesion may in part account for the presence of restricted networks in ASD. Understanding the relationship of beta-adrenergic genes with the genes for neuroligins and neurexins may lend further evidence for the putative effect of propranolol decreasing the signal: noise ratio in cortical networks [135, 300].

We recommend that some alterations to the protocol be made if these studies are repeated. First, order of task could influence outcome of findings, including how attractive a task is to the subject. The eyetracker task was completed first in a dark room, which by chance was generally cold and small. Subjects were asked to keep their head still in a headrest for close to 15 minutes. Those few minutes may have been stressful for a participant who spent part of their time focusing on keeping their head still in the headrest, however the task was not difficult, as the subject was instructed to just view the screen. After concluding that portion of the testing session, the cognitive testing was completed. After the eyetracker task experience, it is possible that the cognitive task was perceived as more difficult due to the nature of the task, which required timed response, however the study environment was more comfortable given that there was no headrest and the task was completed in a shorter amount of time. Thus, both portions of the experimental session contained elements that could influence the stress level of the subject. Previous work from this lab has suggested a benefit of propranolol in typical individuals who are given difficult tasks [301], and a benefit for ASD subjects who are given easy tasks [207]. However, it is not clear based on our study design whether one experimental task was more stressful than the other for either group. Thus, although

drug order was counterbalanced in these studies, if repeated, task order should be counterbalanced as well.

Also, it is important to remember that there is some probability of head motion during the eyetracker study, despite clear instructions to hold the head still. Even a slight tilt forward or backward could have skewed the results of the study, detecting an eye fixation as a forehead fixation or as a nose fixation. Future research like this should include head motion correction.

Finally, if these studies are repeated, we recommend introducing measures of anxiety, attention, and measure of eye contact into the pretesting phase. Measures of anxiety such as the Hamilton Anxiety Scale could provide valuable insight to the subjective stress levels experienced by participants [307]. Measures of attention will provide a method of determining baseline noradrenergic function, as attention is thought to be peaked with a moderate degree of noradrenergic tone [99, 302]. Lastly, an objective measure of baseline eye contact might provide general insight on whether a potential study participant has impaired or normal eye contact.

In previous work, we have found that propranolol has a specific effect in ASD on cognitive flexibility tasks, which it improves in ASD but not in controls. Our findings support the previous work and together they highlight the possible role for noradrenergic blockade in ASD. In addition, the data from the eyetracker study which showed no impairment of ASD relative to controls suggests that the physiologic stress response to eye contact may change over time in ASD based on previous treatment or therapy.

Much is left to learn about ASD. It may likely be found that ASD as we currently know it includes an even greater myriad of independent conditions, both through clinical definition and by pathologic criterion, than what we currently understand. Some areas we have discussed include neurobehavioral, neuroimaging, neuropathologic, neurochemical, and genetic/pharmacogenomic perspectives of current autism research.

Each of these points of view continues to address aspects of pharmacological systems in autism and highlights a need for further research in this field.

Altogether, the information gleaned from these and related studies should be used to affect the future of ASD diagnosis and treatment. One way to apply these findings is to further examine our understanding of noradrenergic system function in ASD.

Another way to explore the use of propranolol in ASD further is to further examine the possible improvement of eye contact with propranolol in a lower functioning ASD population. Finally, as we learn more about propranolol in ASD, it will be wise to apply our knowledge of beta-blockade in ASD to the development of early interventional methods.

CHAPTER 5: FUTURE DIRECTIONS

Here we have suggested some further benefit of noradrenergic blockade through propranolol for individuals on the autism spectrum, particularly in the areas of cognitive flexibility and eye contact. Other benefits of propranolol in ASD that have been examined include shorter latency to solve verbal problems requiring cognitive flexibility, increased functional connectivity, and some benefits to social behavior and language [83, 207, 215].

Thus, we have reported here that propranolol may benefit core impairments of ASD, namely language –through improved function of the semantic network, and non-verbal social behavior, eye contact.

An immediate next step will be to further examine our findings of increased eye contact in ASD with a larger ASD population, preferably with a wider age range to accommodate for the acquisition of social skills with age. It will also be important to accumulate further evidence of benefit of propranolol in ASD, particularly for other core symptoms, such as repetitive behavior.

Moreover, most studies of propranolol in autism mentioned here have used a single low dose of propranolol (40mg) to determine effects. Thus, dosing trials to

examine the effectiveness, safety, efficacy, and risks of sustained propranolol use in high- functioning ASD will be the next critical step. Propranolol has side effects of decreased heart rate and blood pressure, bronchoconstriction, and sedation, among others, which will require careful examination in sustained dosing trials [285]. With successful dosing trials, large controlled clinical studies should be carried out to determine whether the results found here, as well as any future benefits of propranolol in ASD that are identified, relate to the putative clinical benefits of noradrenergic agents that have been previously reported in ASD [207, 237].

These latter studies of propranolol will likely involve high-functioning individuals on the autism spectrum. However, once the effects of propranolol in ASD are better characterized, the next step will be to determine a viable method to test the effects of these drugs on lower functioning individuals with ASD as well. This will help to determine whether the effects of propranolol and usage guidelines characterized in the high- functioning individuals are also present in the low-functioning population.

It must be noted that even if propranolol is found to be unsuitable as a sustained- dose drug, or is ineffective in certain populations, the benefit of new knowledge developed here may still make strides toward improving ASD treatment through application of these principles in novel drug design. Since propranolol may improve core symptoms of ASD, it is possible that a drug with the combined effects of propranolol with other drugs that may improve aspects of ASD - such as risperidone - may provide added benefit to individuals with ASD. Of course, it will be important to examine other classes

of drugs affecting neurotransmitter function, such as SSRIs, to further explore their effects on core symptoms of ASD [57, 67-69].

In sum, a prime goal of ASD research continues to be the study and development of early interventional regimens targeting core symptoms of the condition. Both behavioral and pharmacologic treatment options should be pursued. This will be particularly important in the near future as diagnostic accuracy and screening methods in early childhood, even as young as infancy, is improving in ASD [303-304], and treatment options for that age group remain very limited. As reviewed in chapter 4, the development of future ASD treatment options should also be guided by forthcoming ASD research involving neuropathology, neuroimaging, and genetics in addition to behavioral and pharmacologic areas.

For example, noradrenergic blockade with propranolol is thought to decrease emotional memory, which is mediated by the amygdalaee [283]. Further, some suggest that decreased eye contact in ASD is due to abnormal face processing, involving both the fusiform gyrus and the amygdalaee [202]. Future neuropathology and neuroimaging studies should continue to characterize the development and function of the fusiform gyrus and the amygdalaee in ASD-affected individuals versus healthy controls. Further study of these brain areas may clarify how face processing and other functions, such as emotion processing by the amygdalaee, may differ between ASD and typically developing individuals, both at baseline and with drug intervention. Further, fMRI studies comparing

ASD subjects and control subjects may help characterize functional connectivity among these areas in response to drugs other than propranolol [215], such as SSRIs.

In addition, studies of gene-gene and gene-protein interactions may highlight important linkages and associations in ASD, including that of neurotransmitter systems and synaptic development. These types of studies should include examination of the genes for neurexins and neuroligins, synaptic adhesion proteins thought to be linked to

ASD [297], and the genes and gene-products for neurotransmitter receptors, transporters, and synthetic/degradative enzymes in order to determine how each neurotransmitter milieu, including the noradrenergic system affected by propranolol, may be associated with synaptic development in ASD.

Moreover, future work should examine the influence of other neurotransmitters to eye contact and cognition using pharmacologic studies that involve other classes of drugs. For instance, propranolol may have a nominal effect on serotonin receptors [285], in which case serotonin may play a role in cognition and eye contact findings. A randomized, controlled study of eye contact and cognitive flexibility using an SSRI, such as fluoxetine, may help characterize the role of serotonin in these functions. Further, while benzodiazepine anxiolytics show no effect on cognitive flexibility [210], it is unknown whether they influence eye contact. Anxiolytics, like propranolol, may relieve stress associated with eye contact in individuals with ASD. Thus, further examination of the role of anxiolytics in eye contact in ASD is warranted. Finally, risperidone is thought to decrease aggression in ASD, but its effects on core symptoms remain unclear[305] .

Further characterization of risperidone in ASD will be important to determine the influence of the dopaminergic system in ASD, particularly with regards to ASD core symptoms.

Overall, to better examine propranolol or other future putative treatment options in ASD will require a multidisciplinary approach. One example is to use behavioral approaches and fMRI to determine treatment response to a drug, such as propranolol, in

ASD subjects with different genotypes for a specific target gene, such as the beta-2 adrenergic receptor gene. Another example is to use eyetracking to measure eye contact concurrent with fMRI and SSRI administration in ASD subjects. Using similar methods in further subanalysis of autism subjects versus Asperger’s syndrome subjects may help differentiate the effects of specific treatment options within these subgroups of ASD.

Similarly, lower functioning individuals with ASD, who may have decreased baseline eye contact [171], may be compared to higher functioning individuals with ASD in the above manner, using propranolol or SSRIs, to determine whether level of functioning impacts eye contact with each drug.

In summary, the results of this study demonstrate that there are many other opportunities for further development and examination of putative ASD treatment options. Ultimately, integration of different perspectives in ASD research through collaboration and a comprehensive approach will improve the future for patients with

ASD.

REFERENCES

1. Grandin, T. Dr. Temple Grandin's Webpage . 2009 July 25, 2009]; www.grandin.com/index] .

2. Future Horizons, I. Dr. Temple Grandin . 2005 July 25, 2009]; http://www.templegrandin.com/templehome.html] .

3. Ratey, J.J., T. Grandin, and A. Miller, Defense behavior and coping in an autistic savant: the story of Temple Grandin, PhD. Psychiatry, 1992. 55 (4): p. 382-91.

4. Grandin, T. and M. Scariano, Emergence: Labeled Autistic . 1986, New York: Grand Central Publishing. 200.

5. Kanner, L., Autistic disturbances of affective contact. Acta Paedopsychiatr, 1968. 35 (4): p. 100-36.

6. Bleuler, E., Textbook of Psychiatry . English translation ed. 1924, New York: Macmillan.

7. Organization, W.H., International Statistical Classification of Diseases and Related Health Problems, 8th Revision , Geneva: World Health Assembly.

8. Johnson, B. Some Key Dates in Autism History . Washington Post 2008 July 25, 2009]; http://www.washingtonpost.com/wp- dyn/content/article/2008/06/27/AR2008062703062.html] .

9. Association, A.P., Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) . 1995, Washington, DC: American Psychological Association.

10. Frith, U. and F. Happe, Autism spectrum disorder. Curr Biol, 2005. 15 (19): p. R786- 90.

11. Beversdorf, D.Q., et al., The effect of semantic and emotional context on written recall for verbal language in high functioning adults with autism spectrum disorder. J Neurol Neurosurg Psychiatry, 1998. 65 (5): p. 685-92.

12. Singhania, R., Autistic spectrum disorders. Indian J Pediatr, 2005. 72 (4): p. 343-51.

13. Gould, J., The Lowe and Costello Symbolic Play Test in socially impaired children. J Autism Dev Disord, 1986. 16 (2): p. 199-213.

14. Wing, L., Language, social, and cognitive impairments in autism and severe mental retardation. J Autism Dev Disord, 1981. 11 (1): p. 31-44.

15. Wing, L. and J. Gould, Severe impairments of social interaction and associated abnormalities in children: epidemiology and classification. J Autism Dev Disord, 1979. 9(1): p. 11-29.

16. Shah, A., N. Holmes, and L. Wing, Prevalence of autism and related conditions in adults in a mental handicap hospital. Appl Res Ment Retard, 1982. 3(3): p. 303-17.

17. Wing, L., The continuum of autistic characteristics , in Diagnosis and assessment in autism , E. Schopler and G. Mesibov, Editors. 1988. p. 91-113.

18. Baron-Cohen, S., A.M. Leslie, and U. Frith, Does the autistic child have a "theory of mind"? Cognition, 1985. 21 (1): p. 37-46.

19. Baron-Cohen, S., The autistic child's theory of mind: a case of specific developmental delay. J Child Psychol Psychiatry, 1989. 30 (2): p. 285-97.

20. Baltaxe, C.A. and J.Q. Simmons, Bedtime soliloquies and linguistic competence in autism. J Speech Hear Disord, 1977. 42 (3): p. 376-93.

21. Cromer, R.F., Developmental language disorders: cognitive processes, semantics, pragmatics, phonology, and syntax. J Autism Dev Disord, 1981. 11 (1): p. 57-74.

22. Frith, U. and F. Happe, Language and communication in autistic disorders. Philos Trans R Soc Lond B Biol Sci, 1994. 346 (1315): p. 97-104.

23. Tager-Flusberg, H., On the nature of linguistic functioning in early infantile autism. J Autism Dev Disord, 1981. 11 (1): p. 45-56.

100

24. Lord, C., et al., Diagnosing autism: analyses of data from the Autism Diagnostic Interview. J Autism Dev Disord, 1997. 27 (5): p. 501-17.

25. Lord, C., et al., The autism diagnostic observation schedule-generic: a standard measure of social and communication deficits associated with the spectrum of autism. J Autism Dev Disord, 2000. 30 (3): p. 205-23.

26. Lord, C., et al., Autism diagnostic observation schedule: a standardized observation of communicative and social behavior. J Autism Dev Disord, 1989. 19 (2): p. 185-212.

27. Kippes, C. and C.B. Garrison, Are we in the midst of an autism epidemic? A review of prevalence data. Mo Med, 2006. 103 (1): p. 65-8.

28. NINDS. "Autism Fact Sheet." NIH Publication No. 06-1877 2006.

29. Rice, C., Prevalence of Autism Spectrum Disorders - Autism and Developmental Disabilities, Monitoring Network, 14 Sites, United States, 2002. (CDC) Morbidity and Mortality Weekly Report, 2007. 56 : p. 12-28.

30. Rutter, M., Incidence of autism spectrum disorders: changes over time and their meaning. Acta Paediatr, 2005. 94 (1): p. 2-15.

31. Wing, L. and D. Potter, The epidemiology of autistic spectrum disorders: is the prevalence rising? Ment Retard Dev Disabil Res Rev, 2002. 8(3): p. 151-61.

32. Folstein, S. and M. Rutter, Infantile autism: a genetic study of 21 twin pairs. J Child Psychol Psychiatry, 1977. 18 (4): p. 297-321.

33. Maestrini, E., et al., Identifying autism susceptibility genes. Neuron, 2000. 28 (1): p. 19-24.

34. Ritvo, E.R., et al., Concordance for the syndrome of autism in 40 pairs of afflicted twins. Am J Psychiatry, 1985. 142 (1): p. 74-7.

35. Trottier, G., L. Srivastava, and C.D. Walker, Etiology of infantile autism: a review of recent advances in genetic and neurobiological research. J Psychiatry Neurosci, 1999. 24 (2): p. 103-15.

36. Consortium, T.A.G.P., Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nature Genetics, 2007. 39 : p. 319-328.

37. Kumar, R.A. and S.L. Christian, Genetics of autism spectrum disorders. Curr Neurol Neurosci Rep, 2009. 9(3): p. 188-97. 101

38. Cantor, R.M., Molecular genetics of autism. Curr Psychiatry Rep, 2009. 11 (2): p. 137-42.

39. Buxbaum, J.D., Multiple rare variants in the etiology of autism spectrum disorders. Dialogues Clin Neurosci, 2009. 11 (1): p. 35-43.

40. Chen, X., et al., Structural basis for synaptic adhesion mediated by neuroligin- neurexin interactions. Nat Struct Mol Biol, 2008. 15 (1): p. 50-6.

41. Comoletti, D., et al., Synaptic arrangement of the neuroligin/beta-neurexin complex revealed by X-ray and neutron scattering. Structure, 2007. 15 (6): p. 693- 705.

42. Lise, M.F. and A. El-Husseini, The neuroligin and neurexin families: from structure to function at the synapse. Cell Mol Life Sci, 2006. 63 (16): p. 1833-49.

43. Nguyen, T. and T.C. Sudhof, Binding properties of neuroligin 1 and neurexin 1beta reveal function as heterophilic cell adhesion molecules. J Biol Chem, 1997. 272 (41): p. 26032-9.

44. Pardo, C.A. and C.G. Eberhart, The neurobiology of autism. Brain Pathol, 2007. 17 (4): p. 434-47.

45. Szatmari, P., et al., Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nat Genet, 2007. 39 (3): p. 319-28.

46. Beversdorf, D.Q., et al., Timing of prenatal stressors and autism. J Autism Dev Disord, 2005. 35 (4): p. 471-8.

47. Kinney, D.K., et al., Prenatal stress and risk for autism. Neurosci Biobehav Rev, 2008. 32 (8): p. 1519-32.

48. Stevens, M.C., D.H. Fein, and L.H. Waterhouse, Season of birth effects in autism. J Clin Exp Neuropsychol, 2000. 22 (3): p. 399-407.

49. Reichenberg, A., et al., Birth order effects on autism symptom domains. Psychiatry Res, 2007. 150 (2): p. 199-204.

50. London, E.A., The environment as an etiologic factor in autism: a new direction for research. Environ Health Perspect, 2000. 108 Suppl 3 : p. 401-4.

51. Gerlai, J. and R. Gerlai, Autism: a large unmet medical need and a complex research problem. Physiol Behav, 2003. 79 (3): p. 461-70.

102

52. Beversdorf, D., Therapeutic interventions in autism: a review for primary care physicians. Mo Med, 2008. 105 (5): p. 390-5.

53. McDougle, C.J., et al., Atypical antipsychotics in children and adolescents with autistic and other pervasive developmental disorders. J Clin Psychiatry, 2008. 69 Suppl 4 : p. 15-20.

54. Hellings, J.A., et al., Weight gain in a controlled study of risperidone in children, adolescents and adults with mental retardation and autism. J Child Adolesc Psychopharmacol, 2001. 11 (3): p. 229-38.

55. Malone, R.P. and A. Waheed, The role of antipsychotics in the management of behavioural symptoms in children and adolescents with autism. Drugs, 2009. 69 (5): p. 535-48.

56. Masi, G., et al., Aripiprazole monotherapy in children and young adolescents with pervasive developmental disorders: a retrospective study. CNS Drugs, 2009. 23 (6): p. 511-21.

57. West, L., S.H. Brunssen, and J. Waldrop, Review of the evidence for treatment of children with autism with selective serotonin reuptake inhibitors. J Spec Pediatr Nurs, 2009. 14 (3): p. 183-91.

58. Nickels, K., et al., Stimulant medication treatment of target behaviors in children with autism: a population-based study. J Dev Behav Pediatr, 2008. 29 (2): p. 75-81.

59. Jahromi, L.B., et al., Positive effects of methylphenidate on social communication and self-regulation in children with pervasive developmental disorders and hyperactivity. J Autism Dev Disord, 2009. 39 (3): p. 395-404.

60. Aman, M.G., et al., Treatment of inattention, overactivity, and impulsiveness in autism spectrum disorders. Child Adolesc Psychiatr Clin N Am, 2008. 17 (4): p. 713- 38, vii.

61. Adetunji, B., et al., Risperidone for the core symptom domains of autism. Am J Psychiatry, 2006. 163 (3): p. 551; author reply 551-2.

62. Bryson, S.E., S.J. Rogers, and E. Fombonne, Autism spectrum disorders: early detection, intervention, education, and psychopharmacological management. Can J Psychiatry, 2003. 48 (8): p. 506-16.

63. West, L., J. Waldrop, and S. Brunssen, Pharmacologic treatment for the core deficits and associated symptoms of autism in children. J Pediatr Health Care, 2009. 23 (2): p. 75-89. 103

64. Hollander, E., Treatment of obsessive-compulsive spectrum disorders with SSRIs. Br J Psychiatry Suppl, 1998(35): p. 7-12.

65. Santosh, P.J. and G. Baird, Pharmacotherapy of target symptoms in autistic spectrum disorders. Indian J Pediatr, 2001. 68 (5): p. 427-31.

66. Lindsay, R.L. and M.G. Aman, Pharmacologic therapies aid treatment for autism. Pediatr Ann, 2003. 32 (10): p. 671-6.

67. Moore, M.L., S.F. Eichner, and J.R. Jones, Treating functional impairment of autism with selective serotonin-reuptake inhibitors. Ann Pharmacother, 2004. 38 (9): p. 1515-9.

68. Kolevzon, A., K.A. Mathewson, and E. Hollander, Selective serotonin reuptake inhibitors in autism: a review of efficacy and tolerability. J Clin Psychiatry, 2006. 67 (3): p. 407-14.

69. Posey, D.J., et al., The use of selective serotonin reuptake inhibitors in autism and related disorders. J Child Adolesc Psychopharmacol, 2006. 16 (1-2): p. 181-6.

70. Ming, X., et al., Use of clonidine in children with autism spectrum disorders. Brain Dev, 2008. 30 (7): p. 454-60.

71. Volkmar, F.R., Pharmacological interventions in autism: theoretical and practical issues. J Clin Child Psychol, 2001. 30 (1): p. 80-7.

72. Foxx, R.M., Applied behavior analysis treatment of autism: the state of the art. Child Adolesc Psychiatr Clin N Am, 2008. 17 (4): p. 821-34, ix.

73. Watling, R., S. Tomchek, and P. LaVesser, The scope of occupational therapy services for individuals with autism spectrum disorders across the lifespan. Am J Occup Ther, 2005. 59 (6): p. 680-3.

74. Seung, H.K., et al., Verbal communication outcomes in children with autism after in-home father training. J Intellect Disabil Res, 2006. 50 (Pt 2): p. 139-50.

75. Kern, P., M. Wolery, and D. Aldridge, Use of songs to promote independence in morning greeting routines for young children with autism. J Autism Dev Disord, 2007. 37 (7): p. 1264-71.

76. Jordan, R., Managing autism and Asperger's syndrome in current educational provision. Pediatr Rehabil, 2005. 8(2): p. 104-12.

104

77. Bledsoe, R., B.S. Myles, and R.L. Simpson, Use of a Social Story intervention to improve mealtime skills of an adolescent with Asperger syndrome. Autism, 2003. 7(3): p. 289-95.

78. Kohler, F.W., et al., Using a group-oriented contingency to increase social interactions between children with autism and their peers. A preliminary analysis of corollary supportive behaviors. Behav Modif, 1995. 19 (1): p. 10-32.

79. Krasny, L., et al., Social skills interventions for the autism spectrum: essential ingredients and a model curriculum. Child Adolesc Psychiatr Clin N Am, 2003. 12 (1): p. 107-22.

80. Wachtel, L.E. and L.P. Hagopian, Psychopharmacology and applied behavioral analysis: tandem treatment of severe problem behaviors in intellectual disability and a case series. Isr J Psychiatry Relat Sci, 2006. 43 (4): p. 265-74.

81. Charman, T., et al., Research into early intervention for children with autism and related disorders: methodological and design issues. Report on a workshop funded by the Wellcome Trust, Institute of Child Health, London, UK, November 2001. Autism, 2003. 7(2): p. 217-25.

82. Dawson, G., Early behavioral intervention, brain plasticity, and the prevention of autism spectrum disorder. Dev Psychopathol, 2008. 20 (3): p. 775-803.

83. Ratey, J.J., et al., Open trial effects of beta-blockers on speech and social behaviors in 8 autistic adults. J Autism Dev Disord, 1987. 17 (3): p. 439-46.

84. Koshes, R.J. and N.L. Rock, Use of clonidine for behavioral control in an adult patient with autism. Am J Psychiatry, 1994. 151 (11): p. 1714.

85. Jaselskis, C.A., et al., Clonidine treatment of hyperactive and impulsive children with autistic disorder. J Clin Psychopharmacol, 1992. 12 (5): p. 322-7.

86. Fankhauser, M.P., et al., A double-blind, placebo-controlled study of the efficacy of transdermal clonidine in autism. J Clin Psychiatry, 1992. 53 (3): p. 77-82.

87. Dehner, R., et al., Adrenaline in cardiovascular diseases--effect of beta- adrenoceptor antagonists. Z Kardiol, 1990. 79 Suppl 3 : p. 79-88.

88. Misu, Y. and T. Kubo, Presynaptic beta-adrenoceptors. Med Res Rev, 1986. 6(2): p. 197-225.

89. Majewski, H., Modulation of noradrenaline release through activation of presynaptic beta-adrenoreceptors. J Auton Pharmacol, 1983. 3(1): p. 47-60. 105

90. Invernizzi, R.W. and S. Garattini, Role of presynaptic alpha2-adrenoceptors in antidepressant action: recent findings from microdialysis studies. Prog Neuropsychopharmacol Biol Psychiatry, 2004. 28 (5): p. 819-27.

91. Hertting, G., S. Wurster, and C. Allgaier, Regulatory proteins in presynaptic function. Ann N Y Acad Sci, 1990. 604 : p. 289-304.

92. Silverberg, A.B., et al., Norepinephrine: hormone and neurotransmitter in man. Am J Physiol, 1978. 234 (3): p. E252-6.

93. SciStore. Norepinephrine_IUPAC . July 26, 2009]; http://images.cambridgesoft.com/chemstore/CBN/norepinephrine_IUPAC.jpg] .

94. Cooper, J., F. Bloom, and R. Roth, The biochemical basis of neuropharmacology . 8th ed. 2003: Oxford University Press. 400.

95. Unknown. Budapeststudent . July 26, 2009]; http://www.budapeststudent.com/notes/biochemistry/images/1/1B30Epinephri ne.png] .

96. Ward, M.M., et al., Epinephrine and norepinephrine responses in continuously collected human plasma to a series of stressors. Psychosom Med, 1983. 45 (6): p. 471-86.

97. Laborit, H., On the mechanism of activation of the hypothalamo--pituitary-- adrenal reaction to changes in the environment (the 'alarm reaction'). Resuscitation, 1976. 5(1): p. 19-30.

98. McCorry, L.K., Physiology of the autonomic nervous system. Am J Pharm Educ, 2007. 71 (4): p. 78.

99. Berridge, C.W. and B.D. Waterhouse, The locus coeruleus-noradrenergic system: modulation of behavioral state and state-dependent cognitive processes. Brain Res Brain Res Rev, 2003. 42 (1): p. 33-84.

100. Dohlman, H.G., et al., Model systems for the study of seven-transmembrane- segment receptors. Annu Rev Biochem, 1991. 60 : p. 653-88.

101. Rang, H., et al., Rang and Dale's Pharmacology, 6th edition . 2007: Churchill Livingstone. 844.

102. Foote, S.L., F.E. Bloom, and G. Aston-Jones, Nucleus locus ceruleus: new evidence of anatomical and physiological specificity. Physiol Rev, 1983. 63 (3): p. 844-914.

106

103. Lewis, D.A., The catecholamine innervation of primate cerebral cortex , in Stimulant drugs and ADHD: Basic and Clinical Neuroscience , M.V. Solanto, A.F.T. Arnsten, and F.X. Castellanos, Editors. 2001, Oxford University Press: New York. p. 77-103.

104. Swanson, L.W. and B.K. Hartman, The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-beta-hydroxylase as a marker. J Comp Neurol, 1975. 163 (4): p. 467-505.

105. Foote, S.L. and J.H. Morrison, Extrathalamic modulation of cortical function. Annu Rev Neurosci, 1987. 10 : p. 67-95.

106. Hasselmo, M.E., et al., Noradrenergic suppression of synaptic transmission may influence cortical signal-to-noise ratio. J Neurophysiol, 1997. 77 (6): p. 3326-39.

107. Coull, J.T., et al., The neural correlates of the noradrenergic modulation of human attention, arousal and learning. Eur J Neurosci, 1997. 9(3): p. 589-98.

108. Chamberlain, S.R., et al., Noradrenergic modulation of working memory and emotional memory in humans. Psychopharmacology (Berl), 2006. 188 (4): p. 397- 407.

109. Robbins, T.W., Shifting and stopping: fronto-striatal substrates, neurochemical modulation and clinical implications. Philos Trans R Soc Lond B Biol Sci, 2007. 362 (1481): p. 917-32.

110. Campbell, H.L., et al., Increased task difficulty results in greater impact of noradrenergic modulation of cognitive flexibility. Pharmacol Biochem Behav, 2008. 88 (3): p. 222-9.

111. Aston-Jones, G. and J.D. Cohen, Adaptive gain and the role of the locus coeruleus- norepinephrine system in optimal performance. J Comp Neurol, 2005. 493 (1): p. 99-110.

112. Knight, R.T., et al., The effects of frontal cortex lesions on event-related potentials during auditory selective attention. Electroencephalogr Clin Neurophysiol, 1981. 52 (6): p. 571-82.

113. Woods, D.L. and R.T. Knight, Electrophysiologic evidence of increased distractibility after dorsolateral prefrontal lesions. Neurology, 1986. 36 (2): p. 212- 6.

107

114. Papazian, O., I. Alfonso, and R.J. Luzondo, [Executive function disorders]. Rev Neurol, 2006. 42 Suppl 3 : p. S45-50.

115. Morilak, D.A., et al., Role of brain norepinephrine in the behavioral response to stress. Prog Neuropsychopharmacol Biol Psychiatry, 2005. 29 (8): p. 1214-24.

116. Frankenhaeuser, M., et al., Catecholamine excretion as related to cognitive and emotional reaction patterns. Psychosom Med, 1968. 30 (1): p. 109-24.

117. Mefford, I.N., et al., Determination of plasma catecholamines and free 3,4- dihydroxyphenylacetic acid in continuously collected human plasma by high performance liquid chromatography with electrochemical detection. Life Sci, 1981. 28 (5): p. 477-83.

118. Anisman, H., M. Ritch, and L.S. Sklar, Noradrenergic and dopaminergic interactions in escape behavior: analysis of uncontrollable stress effects. Psychopharmacology (Berl), 1981. 74 (3): p. 263-8.

119. Boulenger, J.P. and T.W. Uhde, Biological peripheral correlates of anxiety. Encephale, 1982. 8(2): p. 119-30.

120. Stone, E.A., Adaptation to stress and brain noradrenergic receptors. Neurosci Biobehav Rev, 1983. 7(4): p. 503-9.

121. Glavin, G.B., Stress and brain noradrenaline: a review. Neurosci Biobehav Rev, 1985. 9(2): p. 233-43.

122. Malarkey, W.B., I.M. Lipkus, and J.T. Cacioppo, The dissociation of catecholamine and hypothalamic-pituitary-adrenal responses to daily stressors using dexamethasone. J Clin Endocrinol Metab, 1995. 80 (8): p. 2458-63.

123. Stratakis, C.A. and G.P. Chrousos, Neuroendocrinology and pathophysiology of the stress system. Ann N Y Acad Sci, 1995. 771 : p. 1-18.

124. Pacak, K. and M. Palkovits, Stressor specificity of central neuroendocrine responses: implications for stress-related disorders. Endocr Rev, 2001. 22 (4): p. 502-48.

125. Valentino, R.J., S.L. Foote, and G. Aston-Jones, Corticotropin-releasing factor activates noradrenergic neurons of the locus coeruleus. Brain Res, 1983. 270 (2): p. 363-7.

126. Valentino, R.J., CRH effects on central noradrenergic neurons: relationship to stress. Adv Exp Med Biol, 1988. 245 : p. 47-64. 108

127. Valentino, R.J. and S.L. Foote, Corticotropin-releasing hormone increases tonic but not sensory-evoked activity of noradrenergic locus coeruleus neurons in unanesthetized rats. J Neurosci, 1988. 8(3): p. 1016-25.

128. Valentino, R.J., M.E. Page, and A.L. Curtis, Activation of noradrenergic locus coeruleus neurons by hemodynamic stress is due to local release of corticotropin- releasing factor. Brain Res, 1991. 555 (1): p. 25-34.

129. Valentino, R.J., S.L. Foote, and M.E. Page, The locus coeruleus as a site for integrating corticotropin-releasing factor and noradrenergic mediation of stress responses. Ann N Y Acad Sci, 1993. 697 : p. 173-88.

130. Cahill, L., et al., The amygdalae and emotional memory. Nature, 1995. 377 (6547): p. 295-6.

131. Bouret, S., et al., Phasic activation of locus ceruleus neurons by the central nucleus of the amygdalae. J Neurosci, 2003. 23 (8): p. 3491-7.

132. Jedema, H.P. and A.A. Grace, Corticotropin-releasing hormone directly activates noradrenergic neurons of the locus ceruleus recorded in vitro. J Neurosci, 2004. 24 (43): p. 9703-13.

133. Kvetnansky, R., et al., Stressor specificity of peripheral catecholaminergic activation. Adv Pharmacol, 1998. 42 : p. 556-60.

134. Pacak, K., Stressor-specific activation of the hypothalamic-pituitary-adrenocortical axis. Physiol Res, 2000. 49 Suppl 1 : p. S11-7.

135. Hasselmo, M., et al., Noradrenergic Suppression of Synaptic Transmission May Influence Cortical Signal-to-Noise Ratio. J. Neurophysiol., 1997. 77 : p. 3326-3339.

136. Beversdorf, D.Q., et al., Noradrenergic modulation of cognitive flexibility in problem solving. NeuroReport, 1999. 10 (13): p. 2763-7.

137. Mesulam, M.M., Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Ann Neurol, 1990. 28 (5): p. 597-613.

138. Frith, U., Cognitive explanations of autism. Acta Paediatr Suppl, 1996. 416 : p. 63- 8.

139. Lopez, B., S.R. Leekam, and G.R. Arts, How central is central coherence? Preliminary evidence on the link between conceptual and perceptual processing in children with autism. Autism, 2008. 12 (2): p. 159-71.

109

140. Loh, M. and G. Deco, Cognitive flexibility and decision-making in a model of conditional visuomotor associations. Eur J Neurosci, 2005. 22 (11): p. 2927-36.

141. Alexander, J.K., et al., Beta-adrenergic modulation of cognitive flexibility during stress. J Cogn Neurosci, 2007. 19 (3): p. 468-78.

142. Kelley, B.J., et al., The effect of propranolol on cognitive flexibility and memory in acute cocaine withdrawal. Neurocase, 2007. 13 (5): p. 320-7.

143. Just, M.A., et al., Functional and anatomical cortical underconnectivity in autism: evidence from an FMRI study of an executive function task and corpus callosum morphometry. Cereb Cortex, 2007. 17 (4): p. 951-61.

144. Belmonte, M.K., et al., Autism and abnormal development of brain connectivity. J Neurosci, 2004. 24 (42): p. 9228-31.

145. Deeley, Q. and D. Murphy, Pathophysiology of autism: evidence from brain imaging. Br J Hosp Med (Lond), 2009. 70 (3): p. 138-42.

146. Frith, C., What do imaging studies tell us about the neural basis of autism? Novartis Found Symp, 2003. 251 : p. 149-66; discussion 166-76, 281-97.

147. Kalia, M., Brain development: anatomy, connectivity, adaptive plasticity, and toxicity. Metabolism, 2008. 57 Suppl 2 : p. S2-5.

148. Kleinhans, N.M., et al., Abnormal functional connectivity in autism spectrum disorders during face processing. Brain, 2008. 131 (Pt 4): p. 1000-12.

149. Frith, U. and F. Happe, Autism: beyond "theory of mind". Cognition, 1994. 50 (1-3): p. 115-32.

150. Minshew, N.J. and G. Goldstein, The pattern of intact and impaired memory functions in autism. J Child Psychol Psychiatry, 2001. 42 (8): p. 1095-101.

151. Beversdorf, D.Q., et al., Network model of decreased context utilization in autism spectrum disorder. J Autism Dev Disord, 2007. 37 (6): p. 1040-8.

152. Beversdorf, D.Q., et al., Increased discrimination of "false memories" in autism spectrum disorder. Proc Natl Acad Sci U S A, 2000. 97 (15): p. 8734-7.

153. Hillier, A., et al., Decreased false memory for visually presented shapes and symbols among adults on the autism spectrum. J Clin Exp Neuropsychol, 2007. 29 (6): p. 610-6.

110

154. Boraston, Z.L., et al., Brief report: perception of genuine and posed smiles by individuals with autism. J Autism Dev Disord, 2008. 38 (3): p. 574-80.

155. Bregman, J.D., J.F. Leckman, and S.I. Ort, Fragile X syndrome: genetic predisposition to psychopathology. J Autism Dev Disord, 1988. 18 (3): p. 343-54.

156. Skuse, D., Genetic influences on the neural basis of social cognition. Philos Trans R Soc Lond B Biol Sci, 2006. 361 (1476): p. 2129-41.

157. Ploog, D., Psychobiology of partnership behaviour. Psychol Med, 1975. 5(4): p. 327-39.

158. Grumet, G.W., Eye contact: the core of interpersonal relatedness. Psychiatry, 1983. 46 (2): p. 172-80.

159. White-Traut, R., Providing a nurturing environment for infants in adverse situations: multisensory strategies for newborn care. J Midwifery Womens Health, 2004. 49 (4 Suppl 1): p. 36-41.

160. Greenman, G.W., Visual behavior of newborn infants . Modern perspectives in child development, ed. A.J. Solnit and S.E. Provence. 1963, New York: International University Press.

161. Barten, S., B. Birns, and J. Ronch, Individual differences in the visual pursuit behavior of neonates. Child Dev, 1971. 42 (1): p. 313-9.

162. Zwaigenbaum, L., et al., Studying the emergence of autism spectrum disorders in high-risk infants: methodological and practical issues. J Autism Dev Disord, 2007. 37 (3): p. 466-80.

163. Baron-Cohen, S., et al., Can Asperger syndrome be diagnosed at 26 months old? A genetic high-risk single-case study. J Child Neurol, 2006. 21 (4): p. 351-6.

164. Adrien, J.L., et al., Autism and family home movies: preliminary findings. J Autism Dev Disord, 1991. 21 (1): p. 43-9.

165. Adrien, J.L., et al., Blind ratings of early symptoms of autism based upon family home movies. J Am Acad Child Adolesc Psychiatry, 1993. 32 (3): p. 617-26.

166. Adrien, J.L., et al., Early symptoms in autism from family home movies. Evaluation and comparison between 1st and 2nd year of life using I.B.S.E. scale. Acta Paedopsychiatr, 1992. 55 (2): p. 71-5.

111

167. Hutt, C. and C. Ounsted, The biological significance of gaze aversion with particular reference to the syndrome of infantile autism. Behav Sci, 1966. 11 (5): p. 346-56.

168. Donovan, W.L. and L.A. Leavitt, Physiologic correlates of direct and averted gaze. Biol Psychol, 1980. 10 (3): p. 189-99.

169. Kylliainen, A. and J.K. Hietanen, Skin conductance responses to another person's gaze in children with autism. J Autism Dev Disord, 2006. 36(4): p. 517-25.

170. Jones, W., K. Carr, and A. Klin, Absence of preferential looking to the eyes of approaching adults predicts level of social disability in 2-year-old toddlers with autism spectrum disorder. Arch Gen Psychiatry, 2008. 65 (8): p. 946-54.

171. Klin, A., et al., Visual fixation patterns during viewing of naturalistic social situations as predictors of social competence in individuals with autism. Arch Gen Psychiatry, 2002. 59 (9): p. 809-16.

172. Hua, H., P. Krishnaswamy, and J.P. Rolland, Video-based eyetracking methods and algorithms in head-mounted displays. Opt Express, 2006. 14 (10): p. 4328-50.

173. Christ, S.E., et al., Inhibitory control in children with phenylketonuria. Dev Neuropsychol, 2006. 30 (3): p. 845-64.

174. Christ, S.E., et al., Inhibitory control in children with autism spectrum disorder. J Autism Dev Disord, 2007. 37 (6): p. 1155-65.

175. Herdman, A.T. and J.D. Ryan, Spatio-temporal brain dynamics underlying saccade execution, suppression, and error-related feedback. J Cogn Neurosci, 2007. 19 (3): p. 420-32.

176. Guestrin, E.D. and M. Eizenman, General theory of remote gaze estimation using the pupil center and corneal reflections. IEEE Trans Biomed Eng, 2006. 53 (6): p. 1124-33.

177. Pelphrey, K.A., et al., Visual scanning of faces in autism. J Autism Dev Disord, 2002. 32 (4): p. 249-61.

178. Riby, D. and P.J. Hancock, Looking at movies and cartoons: eye-tracking evidence from Williams syndrome and autism. J Intellect Disabil Res, 2009. 53 (2): p. 169-81.

179. Norbury, C.F., et al., Eye-movement patterns are associated with communicative competence in autistic spectrum disorders. J Child Psychol Psychiatry, 2009. 50 (7): p. 834-42. 112

180. Ashwin, C., P. Ricciardelli, and S. Baron-Cohen, Positive and negative gaze perception in autism spectrum conditions. Soc Neurosci, 2009. 4(2): p. 153-64.

181. Joseph, R.M., et al., Affective response to eye contact and face recognition ability in children with ASD. J Int Neuropsychol Soc, 2008. 14 (6): p. 947-55.

182. Senju, A., et al., Is anyone looking at me? Direct gaze detection in children with and without autism. Brain Cogn, 2008. 67 (2): p. 127-39.

183. Bird, G., et al., Attention does not modulate neural responses to social stimuli in autism spectrum disorders. NeuroImage, 2006. 31 (4): p. 1614-24.

184. Emery, N.J., The eyes have it: the neuroethology, function and evolution of social gaze. Neurosci Biobehav Rev, 2000. 24 (6): p. 581-604.

185. Falck-Ytter, T., Face inversion effects in autism: a combined looking time and pupillometric study. Autism Res, 2008. 1(5): p. 297-306.

186. Gepner, B., Autism, movement, and facial processing. Am J Psychiatry, 2004. 161 (9): p. 1719; author reply 1719-20.

187. Nation, K. and S. Penny, Sensitivity to eye gaze in autism: is it normal? Is it automatic? Is it social? Dev Psychopathol, 2008. 20 (1): p. 79-97.

188. Wallace, S., et al., A study of impaired judgment of eye-gaze direction and related face-processing deficits in autism spectrum disorders. Perception, 2006. 35 (12): p. 1651-64.

189. Dawson, G., S.J. Webb, and J. McPartland, Understanding the nature of face processing impairment in autism: insights from behavioral and electrophysiological studies. Dev Neuropsychol, 2005. 27 (3): p. 403-24.

190. Richer, J.M. and R.G. Coss, Gaze aversion in autistic and normal children. Acta Psychiatr Scand, 1976. 53 (3): p. 193-210.

191. Klin, A., et al., Visual Fixation Patterns during Viewing of Naturalistic Social Situations as Predictors of Social Competence in Individuals with Autism. Arch. Gen. Psychiat., 2002. 59 : p. 809-816.

192. Joseph, R.M. and J. Tanaka, Holistic and part-based face recognition in children with autism. J Child Psychol Psychiatry, 2003. 44 (4): p. 529-42.

113

193. Joseph, R.M., H. Tager-Flusberg, and C. Lord, Cognitive profiles and social- communicative functioning in children with autism spectrum disorder. J Child Psychol Psychiatry, 2002. 43 (6): p. 807-21.

194. Carey, S. and R. Diamond, From piecemeal to configurational representation of faces. Science, 1977. 195 (4275): p. 312-4.

195. Ashwin, C., et al., Differential activation of the amygdalae and the 'social brain' during fearful face-processing in Asperger Syndrome. Neuropsychologia, 2007. 45 (1): p. 2-14.

196. Spezio, M.L., et al., Amygdalae damage impairs eye contact during conversations with real people. J Neurosci, 2007. 27 (15): p. 3994-7.

197. Schultz, R.T., et al., Abnormal ventral temporal cortical activity during face discrimination among individuals with autism and Asperger syndrome. Arch Gen Psychiatry, 2000. 57 (4): p. 331-40.

198. Hadjikhani, N., et al., Activation of the fusiform gyrus when individuals with autism spectrum disorder view faces. NeuroImage, 2004. 22 (3): p. 1141-50.

199. Pierce, K., et al., Face processing occurs outside the fusiform 'face area' in autism: evidence from functional MRI. Brain, 2001. 124 (Pt 10): p. 2059-73.

200. Kanwisher, N., J. McDermott, and M.M. Chun, The fusiform face area: a module in human extrastriate cortex specialized for face perception. J Neurosci, 1997. 17 (11): p. 4302-11.

201. Hariri, A., et al., The Amygdalae Response to Emotional Stimuli: A Comparison of Faces and Scenes. NeuroImage, 2002. 17 : p. 317-323.

202. Dalton, K.M., et al., Brain function and gaze fixation during facial-emotion processing in fragile X and autism. Autism Res, 2008. 1(4): p. 231-9.

203. Riby, D.M., G. Doherty-Sneddon, and V. Bruce, The eyes or the mouth? Feature salience and unfamiliar face processing in Williams syndrome and autism. Q J Exp Psychol (Colchester), 2009. 62 (1): p. 189-203.

204. Taylor, S.P. and S. Epstein, The measurement of autonomic arousal. Some basic issues illustrated by the covariation of heart rate and skin conductance. Psychosom Med, 1967. 29 (5): p. 514-25.

114

205. Kleinke, C. and P. Pohlen, Affective and Emotional Responses as a Function of Other Person's Gaze and Cooperativeness in a Two-Person Game. J. Personality and Soc. Psychol., 1971. 17 (3): p. 308-313.

206. Hutt, C. and C. Ounsted, The Biological Significance of Gaze Aversion with Particular Reference to the Syndrome of Infantile Autism. Syndrome of Infantile Autism, Behav. Sci., 1966. 11 (5): p. 346-356.

207. Beversdorf, D.Q., et al., Effect of propranolol on verbal problem solving in autism spectrum disorder. Neurocase, 2008. 14 (4): p. 378-83.

208. Beversdorf, D.Q., et al., Central beta-adrenergic modulation of cognitive flexibility. NeuroReport, 2002. 13 (18): p. 2505-7.

209. Choi, Y., et al., The effect of alpha-2 adrenergic agonists on memory and cognitive flexibility. Cogn Behav Neurol, 2006. 19 (4): p. 204-7.

210. Silver, J.A., et al., Effect of anxiolytics on cognitive flexibility in problem solving. Cogn Behav Neurol, 2004. 17 (2): p. 93-7.

211. Smyth, S.F. and D.Q. Beversdorf, Lack of dopaminergic modulation of cognitive flexibility. Cogn Behav Neurol, 2007. 20 (4): p. 225-9.

212. Hillier, A., J.K. Alexander, and D.Q. Beversdorf, The effect of auditory stressors on cognitive flexibility. Neurocase, 2006. 12 (4): p. 228-31.

213. Kelley, B.J., et al., Cognitive impairment in acute cocaine withdrawal. Cogn Behav Neurol, 2005. 18 (2): p. 108-12.

214. Belmonte, M.K. and D.A. Yurgelun-Todd, Functional anatomy of impaired selective attention and compensatory processing in autism. Brain Res Cogn Brain Res, 2003. 17 (3): p. 651-64.

215. Narayanan, A., et al., Effect of Propranolol on Functional Connectivity in Autism. [in press], 2009.

216. Senju, A. and M.H. Johnson, Atypical eye contact in autism: Models, mechanisms and development. Neurosci Biobehav Rev, 2009.

217. Neumann, D., et al., Looking you in the mouth: abnormal gaze in autism resulting from impaired top-down modulation of visual attention. Soc Cogn Affect Neurosci, 2006. 1(3): p. 194-202.

115

218. Baharav, E. and R. Darling, Case report: Using an auditory trainer with caregiver video modeling to enhance communication and socialization behaviors in autism. J Autism Dev Disord, 2008. 38 (4): p. 771-5.

219. Donnelly, J.L., P.D. Luyben, and C.S. Zan, Increasing eye contact toward learning materials in a toddler with autism. J Prev Interv Community, 2009. 37 (3): p. 170-6.

220. Koegel, R.L., T.W. Vernon, and L.K. Koegel, Improving Social Initiations in Young Children with Autism Using Reinforcers with Embedded Social Interactions. J Autism Dev Disord, 2009.

221. Niederhofer, H., W. Staffen, and A. Mair, Tianeptine: a novel strategy of psychopharmacological treatment of children with autistic disorder. Hum Psychopharmacol, 2003. 18 (5): p. 389-93.

222. Hermelin, B. and N. O'Connor, Psychological Experiments with Autistic Children . 1970, Oxford: Oxford University Press.

223. Just, M., et al., Cortical Activation and Synchronization during Sentence Comprehension in High-Functioning Autism: Evidence of Underconnectivity. Brain, 2004. 127 (1811-1821).

224. Bowler, D.M., et al., Memory illusions: false recall and recognition in adults with Asperger's syndrome. J Abnorm Psychol, 2000. 109 (4): p. 663-72.

225. Martineau, J., et al., Monoamines (serotonin and catecholamines) and their derivatives in infantile autism: age-related changes and drug effects. Dev Med Child Neurol, 1992. 34 (7): p. 593-603.

226. Barthelemy, C., et al., Urinary free and conjugated catecholamines and metabolites in autistic children. J Autism Dev Disord, 1988. 18 (4): p. 583-91.

227. Minderaa, R.B., et al., Noradrenergic and adrenergic functioning in autism. Biol Psychiatry, 1994. 36 (4): p. 237-41.

228. Martchek, M., et al., Lack of Evidence of Neuropathology in the Locus Coeruleus in Autism. Acta Neuropath., 2006. 111 : p. 497-499.

229. Lord, C., N. Rutter, and A. LeCouteur, Autism Diagnostic Interview-Revised: A Revised Version of a Diagnostic Interview for Caregivers of Individuals with Possible Pervasive Developmental Disorders. JADD, 1994. 24 : p. 659-685.

230. Doppelt, J.E., Estimating the full scale score on the Wechsler adult intelligence scale from scores on four subtests. J Consult Psychol, 1956. 20 (1): p. 63-6. 116

231. Hertzman, M., Galantamine in the treatment of adult autism: a report of three clinical cases. Int J Psychiatry Med, 2003. 33 (4): p. 395-8.

232. Kleinhans, N.M., et al., Atypical functional lateralization of language in autism spectrum disorders. Brain Res, 2008. 1221 : p. 115-25.

233. Spek, A., et al., Verbal fluency in adults with high functioning autism or Asperger syndrome. Neuropsychologia, 2009. 47 (3): p. 652-6.

234. Turner, M.A., Generating novel ideas: fluency performance in high-functioning and learning disabled individuals with autism. J Child Psychol Psychiatry, 1999. 40 (2): p. 189-201.

235. McClelland, J.L., The basis of hyperspecificity in autism: a preliminary suggestion based on properties of neural nets. J Autism Dev Disord, 2000. 30 (5): p. 497-502.

236. Cohen, I.L., An artificial neural network analogue of learning in autism. Biol Psychiatry, 1994. 36 (1): p. 5-20.

237. Ratey, J., et al., Brief Report: Open Trial Effects of Beta-Blockers on Speech and Social Behaviors in 8 Autistic Adults. JADD, 1987. 17 (3): p. 439-446.

238. Cahill, L., et al., Beta-adrenergic Activation and Memory for Emotional Events. Nature, 1994. 371 : p. 702-704.

239. Faigel, H., The Effect of Beta-blockade on Stress-induced Cognitive Dysfunction in Adolescents. Clin. Pediatrics., 1991. 30 : p. 441-445.

240. Lader, M., Beta-adrenoceptor antagonists in neuropsychiatry: an update. J Clin Psychiatry, 1988. 49 (6): p. 213-23.

241. Laverdue, B. and J.P. Boulenger, Medications Beta-Bloquantes et Anxiete. Un Interet Therapeutique Certain. [Beta-blocking Drugs and Anxiety: A Proven Therapeutic Value]. L'Encephale, 1991. 17 : p. 481-492.

242. Weiss, A.H., et al., Eye movement Abnormalities in Joubert Syndrome. Invest Ophthalmol Vis Sci, 2009.

243. Wieser, M.J., et al., Is eye to eye contact really threatening and avoided in social anxiety?--An eye-tracking and psychophysiology study. J Anxiety Disord, 2009. 23 (1): p. 93-103.

117

244. Goldstein, R.B., R.L. Woods, and E. Peli, Where people look when watching movies: do all viewers look at the same place? Comput Biol Med, 2007. 37 (7): p. 957-64.

245. Detenber, B., R. Simons, and G. Bennett Jr., Roll 'em!: The Effects of Picture Motion on Emotional Responses. J Broadcasting and Electronic Media, 1998. 42 (1): p. 112-126.

246. Kylliainen, A. and J. Hietanen, Skin Conductance Responses to Another Person's Gaze in Children with Autism. JADD, 2006. 36 (4): p. 517-525.

247. Baron-Cohen, S., et al., Reading the Mind in the Face: A Cross-cultural and Developmental Study. Visual Cognition, 1996. 3(1): p. 39-59.

248. Dalton, K.M., et al., Gaze fixation and the neural circuitry of face processing in autism. Nat Neurosci, 2005. 8(4): p. 519-26.

249. Kemner, C., et al., Abnormal saccadic eye movements in autistic children. J Autism Dev Disord, 1998. 28 (1): p. 61-7.

250. van der Geest, J.N., et al., Eye movements, visual attention, and autism: a saccadic reaction time study using the gap and overlap paradigm. Biol Psychiatry, 2001. 50 (8): p. 614-9.

251. Roeyers, H., The influence of nonhandicapped peers on the social interactions of children with a pervasive development disorder. J Autism Dev Disord, 1996. 26 (3): p. 303-20.

252. Hwang, B. and C. Hughes, The effects of social interactive training on early social communicative skills of children with autism. J Autism Dev Disord, 2000. 30 (4): p. 331-43.

253. Bauminger, N., The facilitation of social-emotional understanding and social interaction in high-functioning children with autism: intervention outcomes. J Autism Dev Disord, 2002. 32 (4): p. 283-98.

254. Koegel, R.L., T.W. Vernon, and L.K. Koegel, Improving social initiations in young children with autism using reinforcers with embedded social interactions. J Autism Dev Disord, 2009. 39 (9): p. 1240-51.

255. Wong, V.C. and Q.K. Kwan, Randomized Controlled Trial for Early Intervention for Autism: A Pilot Study of the Autism 1-2-3 Project. J Autism Dev Disord, 2009.

118

256. Abrams, R.A. and S.E. Christ, Motion onset captures attention: a rejoinder to Franconeri and Simons (2005). Percept Psychophys, 2006. 68 (1): p. 114-7.

257. Abrams, R.A. and S.E. Christ, Motion onset captures attention. Psychol Sci, 2003. 14 (5): p. 427-32.

258. Abrams, R.A. and S.E. Christ, The onset of receding motion captures attention: comment on Franconeri and Simons (2003). Percept Psychophys, 2005. 67 (2): p. 219-23.

259. Zwaigenbaum, L., et al., Behavioral manifestations of autism in the first year of life. Int J Dev Neurosci, 2005. 23 (2-3): p. 143-52.

260. Osterling, J. and G. Dawson, Early recognition of children with autism: a study of first birthday home videotapes. J Autism Dev Disord, 1994. 24 (3): p. 247-57.

261. Beversdorf, D., The Role of the Noradrenergic System in Autism Spectrum Disorders , in The Neurochemical Basis of Autism: Molecules to Minicolumns , G.J. Blatt, Editor. 2008, Springer.

262. Lang, P., et al., Looking at Pictures: Affective, Facial, Visceral, and Behavioral Reactions. Psychophysiol., 1993. 30 : p. 261-273.

263. Corbett, B.A., et al., Cortisol circadian rhythms and response to stress in children with autism. Psychoneuroendocrinology, 2006. 31 (1): p. 59-68.

264. Corbett, B.A., et al., Comparing cortisol, stress, and sensory sensitivity in children with autism. Autism Res, 2009. 2(1): p. 39-49.

265. Marinovic-Curin, J., et al., Slower cortisol response during ACTH stimulation test in autistic children. Eur Child Adolesc Psychiatry, 2008. 17 (1): p. 39-43.

266. Richdale, A.L. and M.R. Prior, Urinary cortisol circadian rhythm in a group of high- functioning children with autism. J Autism Dev Disord, 1992. 22 (3): p. 433-47.

267. Tani, P., et al., Higher plasma ACTH levels in adults with Asperger syndrome. J Psychosom Res, 2005. 58 (6): p. 533-6.

268. Adrien, J.L., et al., Use of bioclinical markers for the assessment and treatment of children with pervasive developmental disorders. Neuropsychobiology, 1989. 22 (3): p. 117-24.

269. Launay, J.M., et al., Catecholamines metabolism in infantile autism: a controlled study of 22 autistic children. J Autism Dev Disord, 1987. 17 (3): p. 333-47. 119

270. Young, J.G., et al., Decreased urinary free catecholamines in childhood autism. J Am Acad Child Psychiatry, 1978. 17 (4): p. 671-8.

271. Brunet, A., et al., Effect of post-retrieval propranolol on psychophysiologic responding during subsequent script-driven traumatic imagery in post-traumatic stress disorder. J Psychiatr Res, 2008. 42 (6): p. 503-6.

272. Boyd, R.D. and M.J. Corley, Outcome survey of early intensive behavioral intervention for young children with autism in a community setting. Autism, 2001. 5(4): p. 430-41.

273. Chung, K.M., et al., Peer-mediated social skills training program for young children with high-functioning autism. Res Dev Disabil, 2007. 28 (4): p. 423-36.

274. McDougle, C.J., et al., A double-blind, placebo-controlled study of risperidone in adults with autistic disorder and other pervasive developmental disorders. Arch Gen Psychiatry, 1998. 55 (7): p. 633-41.

275. Kamp-Becker, I., et al., Dimensional structure of the autism phenotype: relations between early development and current presentation. J Autism Dev Disord, 2009. 39 (4): p. 557-71.

276. Hadjikhani, N., et al., Abnormal activation of the social brain during face perception in autism. Hum Brain Mapp, 2007. 28 (5): p. 441-9.

277. Dalton, K., et al., Gaze Fixation and the Neural Circuitry of Face Processing in Autism. Nature Neurosci., 2005. 8(4): p. 519-526.

278. Noonan, S.K., F. Haist, and R.A. Muller, Aberrant functional connectivity in autism: evidence from low-frequency BOLD signal fluctuations. Brain Res, 2009. 1262 : p. 48-63.

279. Silani, G., et al., Levels of emotional awareness and autism: an fMRI study. Soc Neurosci, 2008. 3(2): p. 97-112.

280. Baron-Cohen, S., et al., The amygdalae theory of autism. Neurosci Biobehav Rev, 2000. 24 (3): p. 355-64.

281. Bauman, M.L. and T.L. Kemper, The neuropathology of the autism spectrum disorders: what have we learned? Novartis Found Symp, 2003. 251 : p. 112-22; discussion 122-8, 281-97.

120

282. Smolka, M.N., et al., Catechol-O-methyltransferase val158met genotype affects processing of emotional stimuli in the amygdalae and prefrontal cortex. J Neurosci, 2005. 25 (4): p. 836-42.

283. Cahill, L., et al., Beta-adrenergic activation and memory for emotional events. Nature, 1994. 371 (6499): p. 702-4.

284. Parikh, M.S., A. Kolevzon, and E. Hollander, Psychopharmacology of aggression in children and adolescents with autism: a critical review of efficacy and tolerability. J Child Adolesc Psychopharmacol, 2008. 18 (2): p. 157-78.

285. Katzung, B.G., ed. Basic and Clinical Pharmacology . 8 ed. 2001, Lange Medical Books/McGraw-Hill: New York.

286. Hutt, C., et al., Arousal and Childhood Autism. Nature, 1964. 204 : p. 908-9.

287. Young, J.G., et al., Decreased 24-hour urinary MHPG in childhood autism. Am J Psychiatry, 1979. 136 (8): p. 1055-7.

288. Martineau, J., et al., Catecholaminergic metabolism and autism. Dev Med Child Neurol, 1994. 36 (8): p. 688-97.

289. Hranilovic, D., et al., Hyperserotonemia in autism: the potential role of 5HT- related gene variants. Coll Antropol, 2008. 32 Suppl 1 : p. 75-80.

290. Janusonis, S., Origin of the blood hyperserotonemia of autism. Theor Biol Med Model, 2008. 5: p. 10.

291. Connors, S.L., et al., beta2-adrenergic receptor activation and genetic polymorphisms in autism: data from dizygotic twins. J Child Neurol, 2005. 20 (11): p. 876-84.

292. Raznahan, A., et al., Serotonin transporter genotype and neuroanatomy in autism spectrum disorders. Psychiatr Genet, 2009. 19 (3): p. 147-50.

293. Hoyer, D., et al., Serotonin 5-HT1D receptors. Ann N Y Acad Sci, 1990. 600 : p. 168- 81; discussion 181-2.

294. Green, A.R., P. Johnson, and V.L. Nimgaonkar, Interactions of beta-adrenoceptor agonists and antagonists with the 5-hydroxytryptamine2 (5-HT2) receptor. Neuropharmacology, 1983. 22 (5): p. 657-60.

295. Cheng, S.B., et al., Presynaptic targeting of alpha4beta2 nicotinic acetylcholine receptors is regulated by neurexin-1beta. J Biol Chem, 2009. 121

296. Comoletti, D., et al., Characterization of the solution structure of a neuroligin/beta-neurexin complex. Chem Biol Interact, 2008. 175 (1-3): p. 150-5.

297. Sudhof, T.C., Neuroligins and neurexins link synaptic function to cognitive disease. Nature, 2008. 455 (7215): p. 903-11.

298. Tabuchi, K., et al., A neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice. Science, 2007. 318 (5847): p. 71-6.

299. Yan, J., et al., Neurexin 1alpha structural variants associated with autism. Neurosci Lett, 2008. 438 (3): p. 368-70.

300. Cheslack-Postava, K., et al., beta2-Adrenergic receptor gene variants and risk for autism in the AGRE cohort. Mol Psychiatry, 2007. 12 (3): p. 283-91.

301. Campbell, H., et al., Increased Task Difficulty Results in Greater Impact of Noradrenergic Modulation of Cognitive Flexibility. Pharmacol., Biochem., and Behav., 2008. 88 : p. 222-229.

302. Aston-Jones, G., J. Rajkowski, and J. Cohen, Locus coeruleus and regulation of behavioral flexibility and attention. Prog Brain Res, 2000. 126 : p. 165-82.

303. Zwaigenbaum, L., et al., Clinical assessment and management of toddlers with suspected autism spectrum disorder: insights from studies of high-risk infants. Pediatrics, 2009. 123 (5): p. 1383-91.

304. Stahmer, A.C., The basic structure of community early intervention programs for children with autism: provider descriptions. J Autism Dev Disord, 2007. 37 (7): p. 1344-54.

305. Chavez, B., M. Chavez-Brown, and J.A. Rey, Role of risperidone in children with autism spectrum disorder. Ann Pharmacother, 2006. 40 (5): p. 909-16.

306. Norbury, C.F., J. Borck, L. Cragg, S. Einav, H. Griffiths, and K. Nation, Eye- movement patterns are associated with communicative competence in autism spectrum disorders. J Child Psychol Psychiatry, 2009. 50 : p. 834-42.

307. Hamilton, M., Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol, 1967. 6: p. 278-96.

122

APPENDIX A: VERBAL FLUENCY DATA

123

Placebo-ASD Subject Phonemic (letters) Semantic (categories) 1 9.33 12.67 2 9.29 6.88 3 10.66 10.60 4 3.79 8.88 5 9.63 11.75 6 7.67 6.00 7 8.33 9.00 8 6.00 11.67 9 10.32 11.46 10 7.33 10.00 11 3.79 8.02 12 4.13 8.02 13 7.67 10.00 14 6.67 3.33 AVERAGE 7.47 9.16 STDEV 2.35 2.57 Placebo-Controls Subject Phonemic (letters) Semantic (categories) 1 6.77 12.20 2 14.33 10.67 3 9.33 11.33 4 8.67 10.67 5 8.91 8.72 6 8.20 8.72 7 10.33 13.33 8 11.00 11.67 9 8.56 9.59 10 8.00 13.67 11 11.05 10.75 12 10.33 13.33 13 8.56 11.62 14 10.70 10.46 AVERAGE 9.62 11.19 STDEV 1.86 1.59

124

Propranolol-ASD

Semantic Phonemic (letters) Subject (categories) 1 11.01 15.19 2 8.00 12.00 3 10.00 13.00 4 5.00 6.67 5 8.67 16.00 6 6.88 6.88 7 7.92 10.32 8 5.16 11.46 9 11.00 14.33 10 6.53 9.18 11 5.33 6.67 12 5.67 7.67 13 8.26 12.61 14 9.29 7.74 AVERAGE 7.77 10.69 STDEV 2.08 3.28 Subject Propranolol-Controls Semantic Phonemic (letters) 1 (categories) 2 7.67 14.67 3 16.40 12.79 4 8.56 9.88 5 8.91 10.75 6 9.67 10.00 7 8.33 9.33 8 9.98 15.98 9 8.20 11.62 10 8.00 7.33 11 7.84 14.23 12 8.67 11.00 13 10.70 11.04 14 9.00 9.33 10.00 12.00

AVERAGE 9.42 11.42 STDEV 2.20 2.36

125

APPENDIX B: EYE CONTACT DATA

126

PLACEBO-ASD

MEAN DATA PER VIDEO

VIDEO EYES(s) NOSE (s) MOUTH (s) PROPORTION FIX ON EYE F01 3.27 1.32 0.46 0.65 F02 3.45 1.15 0.31 0.70 F03 2.54 0.93 0.75 0.60 F04 3.07 0.38 0.56 0.77 F05 2.46 0.37 0.65 0.71 F06 3.72 0.29 0.17 0.89 F07 3.61 0.33 0.18 0.88 F08 2.42 1.61 0.46 0.54 F09 3.30 0.61 0.50 0.75 F10 3.12 0.59 0.44 0.75 F11 2.17 0.50 0.38 0.71 F12 1.94 2.19 0.73 0.40 F13 4.37 0.34 0.36 0.86 F14 3.63 0.24 0.52 0.83 F15 2.33 0.87 0.56 0.62 F16 3.32 0.92 0.51 0.70 M01 3.63 1.22 0.33 0.70 M02 2.77 1.49 0.29 0.61 M03 2.66 0.80 0.59 0.66 M04 2.85 1.30 0.56 0.61 M05 2.44 1.32 1.02 0.51 M06 2.90 0.28 0.89 0.71 M07 4.26 0.90 0.22 0.79 M08 2.21 1.18 1.02 0.50 M09 2.45 1.11 0.68 0.58 M10 3.63 0.98 0.50 0.71 M11 2.50 1.28 0.16 0.63 M12 3.13 0.77 0.64 0.69 M13 3.13 0.93 0.15 0.74 M14 2.90 0.86 1.31 0.57 M15 2.20 1.69 0.69 0.48 M16 4.14 0.96 0.33 0.76

TOTAL 35.89 12.24 7.62

AVERAGE 3.02 0.93 0.53 0.68 127

STDEV 0.64 0.47 0.27 0.12 PROPORTION OF FIX 0.67 0.21 0.12

PROPRANOLOL-ASD

MEAN DATA PER VIDEO

VIDEO EYES(s) NOSE (s) MOUTH (s) PROPORTION FIX ON EYE F01 4.00 0.73 0.15 0.82 F02 3.52 0.94 0.31 0.74 F03 2.79 1.28 0.01 0.68 F04 3.41 1.47 0.02 0.70 F05 2.20 0.42 0.23 0.77 F06 3.82 0.49 0.12 0.86 F07 3.44 0.47 0.65 0.75 F08 3.53 0.75 0.19 0.79 F09 2.70 0.77 0.37 0.70 F10 3.16 1.32 0.12 0.69 F11 3.15 0.72 0.23 0.77 F12 4.46 0.91 0.20 0.80 F13 3.72 0.55 0.25 0.82 F14 4.09 0.69 0.08 0.84 F15 2.12 1.10 0.37 0.59 F16 2.88 1.20 0.57 0.62 M01 3.50 0.80 0.21 0.78 M02 3.54 0.81 0.39 0.75 M03 3.38 0.99 0.35 0.72 M04 3.19 1.31 0.85 0.60 M05 3.30 1.40 0.14 0.68 M06 3.77 1.27 0.58 0.67 M07 3.49 0.39 0.32 0.83 M08 2.97 0.98 0.35 0.69 M09 2.89 1.03 0.19 0.70 M10 3.68 0.82 0.19 0.78 M11 2.36 0.92 0.29 0.66 M12 3.21 1.33 0.37 0.65 M13 3.95 0.25 0.26 0.89 M14 4.56 0.66 0.13 0.85 M15 3.77 1.23 0.84 0.65 128

M16 3.12 0.97 0.39 0.70

TOTAL 41.08 11.25 4.06 8.76 AVERAGE 3.36 0.91 0.30 0.74 STDEV 0.58 0.32 0.21 0.08 PROPORTION OF FIX 0.74 0.20 0.07

PLACEBO-CONTROL MEAN DATA PER VIDEO VIDEO EYES(s) NOSE (s) MOUTH (s) PROPORTION FIX ON EYE F01 3.38 1.27 0.59 0.34 F02 4.68 1.46 0.39 0.47 F03 2.08 2.14 0.34 0.21 F04 3.69 0.68 0.43 0.37 F05 2.94 0.55 0.54 0.29 F06 4.20 0.25 0.47 0.42 F07 2.95 0.89 0.41 0.29 F08 2.99 0.86 0.46 0.30 F09 1.43 1.63 0.78 0.14 F10 1.80 2.84 0.33 0.18 F11 2.05 2.51 0.30 0.20 F12 3.25 2.07 0.16 0.32 F13 1.47 1.87 0.54 0.15 F14 2.91 1.07 0.11 0.29 F15 2.10 0.89 0.47 0.21 F16 2.71 2.01 0.38 0.27 M01 3.86 1.23 0.08 0.39 M02 4.05 1.24 0.15 0.40 M03 4.01 0.74 0.63 0.40 M04 3.31 0.78 0.81 0.33 M05 3.42 0.82 0.42 0.34 M06 3.88 0.62 0.20 0.39 M07 3.48 0.73 1.15 0.35 M08 2.96 1.74 0.78 0.30 M09 1.80 2.53 0.02 0.18 M10 1.81 1.59 0.51 0.18

129

M11 1.41 3.49 0.19 0.14 M12 1.35 2.52 0.61 0.14 M13 1.69 1.20 0.82 0.17 M14 1.42 1.52 1.59 0.14 M15 1.24 2.59 0.28 0.12 M16 1.49 2.09 0.51 0.15

TOTAL 85.79 48.43 15.45 8.58 AVERAGE 2.68 1.51 0.48 0.27 STDEV 1.02 0.79 0.32 0.10 PROPORTION OF FIX 0.57 0.32 0.10

PROPRANOLOL-CONTROL MEAN DATA PER VIDEO VIDEO EYES(s) NOSE (s) MOUTH (s) PROPORTION FIX ON EYE F01 3.47 1.08 0.32 0.71 F02 3.86 0.99 0.09 0.78 F03 3.08 1.91 0.31 0.58 F04 2.77 1.37 0.23 0.63 F05 2.20 1.08 0.83 0.54 F06 4.30 0.63 0.16 0.84 F07 3.12 1.70 0.36 0.60 F08 2.25 2.46 0.43 0.44 F09 6.02 0.70 0.28 0.86 F10 4.54 1.56 0.26 0.71 F11 5.59 0.52 0.48 0.85 F12 5.54 0.34 0.65 0.85 F13 4.27 0.78 0.36 0.79 F14 5.43 0.55 0.30 0.86 F15 2.52 1.11 0.34 0.63 F16 3.79 0.88 0.87 0.68 M01 2.83 1.52 0.14 0.63 M02 2.12 1.93 0.14 0.51 M03 2.86 1.32 0.26 0.64 M04 1.75 1.21 0.90 0.45 M05 2.83 1.02 0.25 0.69 M06 3.42 1.26 0.32 0.68 M07 3.00 0.47 1.50 0.60

130

M08 2.69 1.04 1.01 0.57 M09 4.03 0.68 0.30 0.80 M10 5.76 0.38 0.18 0.91 M11 3.76 1.55 0.29 0.67 M12 4.77 0.89 0.49 0.78 M13 4.48 0.58 0.44 0.81 M14 4.81 0.51 0.91 0.77 M15 4.19 1.26 0.16 0.75 M16 4.47 0.79 0.52 0.77

TOTAL 120.50 34.07 14.08 22.42 AVERAGE 3.77 1.06 0.44 0.70 STDEV 1.17 0.51 0.32 0.12 PROPORTION OF FIX 0.71 0.20 0.08

131