PARIETAL NEUROPHYSIOLOGY DURING SUSTAINED ATTENTIONAL PERFORMANCE: ASSESSMENT OF CHOLINERGIC CONTRIBUTION TO PARIETAL PROCESSING
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of the Ohio State University
By
John Isaac Broussard, B.A., M.A.
****
The Ohio State University 2007
Dissertation Committee: Approved by Associate Professor Ben Givens, Advisor
Professor Martin Sarter
Professor John Bruno ______Associate Professor John Buford Advisor Graduate Program in Psycholology
ABSTRACT
There were three major aims of this dissertation, all of which pertained to neurophysiological correlates of sustained attention task performance in rats. The first aim was to examine whether the evoked neurophysiological responses of local field potential activity of the parietal cortex during the detection of signals and the rejection of nonsignals produces behavioral correlates similar to that of single unit activity. The second aim was to test whether removal of local cholinergic input to the PPC via infusion of a specific cholinotoxin reduces the signal-related increases in firing rate of
PPC neurons. The third aim was to test whether unilateral cholinergic deafferentation of the medial prefrontal cortex of rats specifically reduced the responses of PPC in the presence of a visual distractor. In the first study, it was determined that visual signal recruited an event-related potential (ERP) similar to the P300, an ERP component found in human PPC during the detection of infrequent and unpredictable stimuli.
Amplitude of the ERP varied as a function of signal duration. Analysis of the spectral content of the evoked response indicated increases in alpha power as a function of correct detection. In the second study it was determined that restricted loss of cholinergic input near the recording site significantly reduced the relative number of signal-related neurons in the PPC. This manipulation also increased the relative number of neurons responsive to the visual distractor, and increased the baseline firing rate of neurons initially activated by the signal. In the third study cholinergic
ii
deafferentation of mPFC did not reduce the number of neurons responsive to the visual signal, but analysis of the trial blocks indicated a specific impairment of these neurons to encode the signal in the presence of the distractor. Further, bilateral cholinergic deafferentation produced a side bias to the hit lever during distractor sessions. The results of these experiments suggest that cholinergic input to the PPC is necessary for the filtering of distractors and the optimization of signal-related activity, whereas cholinergic input to the mPFC of rats is specifically required to produce the behavioral flexibility to maintain task performance under attentional challenges.
iii
Dedicated to Yun-ju
iv
ACKNOWLEDGEMENTS
I would like to express my appreciation to my advisor, Ben Givens, for patience and encouragement throughout my graduate career. I thank Martin Sarter, who provided guidance and intellectual stimulation at every encounter. John Bruno has shared much knowledge and wisdom, and has helped me develop my skills as a writer and a presenter with useful classes and advice over the years. I am grateful to John Buford for his participation in this committee.
Special thanks are due to past and current students and post-doctoral associates.
Mike Gill was always quick to respond to my many, many technical inquiries early in my tenure here. Josh Burke and Chris Herzog were also very supportive for my first year.
Sharmila Venugopal’s contribution in aiding the programming and calibration of the hardware and software was invaluable. I thank Kate Karelina for her assistance with technical aspects of these experiments.
The many undergraduates who have contributed their time and effort have been invaluable in my studies. A short list includes: Michael Rosner, David Osher, Andrea
Kornbau, Karen Lindsay, Celia Less, Tim, Sameh Elguizaoui, Sara Shelton, Ryan Dagg,
Timothy Simmons, Amy Gosnell, and Amir Adeli.
v
VITA
December 05, 1978………………………Born – Abbeville, Louisiana
2001……………………………………...B.S. Psychology, Louisiana State University.
2004……………………………………...M.A. Psychology The Ohio State University.
2001-present…………………………….Graduate Teaching and Research Associate, The Ohio State University
PUBLICATIONS 1. Broussard, J., Sarter, M., & Givens, B. (2006). Neuronal correlates of signal detection in the posterior parietal cortex of rats performing a sustained attention task. Neuroscience. 143: 2, 407-417.
2. Hawkins, M.F., Uzelac, S.M., Baumeister, A.A., Hearn, J.K., Broussard, J.I., & Guillot, T.S. (2002). Behavioral Responses to Stress Following Central and Peripheral Injection of the 5-HT2 Agonist DOI. Pharmacology, Biochemistry, & Behavior. 73: 3, 537-544.
FIELDS OF STUDY
Major Field: Psychology
Minor Field: Behavioral Neuroscience
vi
TABLE OF CONTENTS
Abstract……………………………………………………………………………...... ii Dedication……………………………………………………………………….………..iv Acknowledgements………………………………………………………………………..v Vita……………………………………………………………………...... vi List of Tables………………………………………………………………………...... x List of Figures…………………………………………………………………………….xi
1. General Introduction………………………………………………………………..…..1 1.1. Posterior parietal cortex contributes to attentional processing……………….1 1.2 Identification of the rodent homologue of the PPC…………………………...4 1.3 The role of ACh in the evoked response……………………………………...6 1.4 ACh input in cortical synchrony………………………………………..……12 1.5 Frontal cortex and effortful cognitive control…………………………..……16 1.6 Specific Aims ………………………….………………………………...... 19
2. General Methods……………………………………………………………………...24 2.1 Subjects and apparatus……………………………………………………….24 2.2. Behavioral training…………………………………………………………..25 2.3. Electrode and infusion cannula implantation………………………………..27 2.4. Neurophysiological recording sessions……………………………………...28 2.5. Histology………………………………………………………………….....29 2.6. Behavioral measures………………………………………………………...30 2.7. Neurophysiological measures……………………………………………….31
3. Experiment One………………………………………………………………….……34 3.1. Introduction………………………………………………………………....34 3.2. Specific Methods…………………………………………………………....39 3.2.1. Recording and analysis of EEG…………………………………...39 3.3. Results………………………………………………………………….……41 3.3.1. Histology…………………………………………………….…….41 3.3.2. Behavioral effects of signal duration and distractor………………41 3.3.3. Event Related Potentials…………………………………………..42 3.3.4. Increases in delta and theta power reflect changes in signal duration…………………………………………………43 3.3.5. Phasic increases in alpha power reflect correct detection of signals ………………………………………………………...44
vii 3.3.6. Distractor evoked changes in phasic and tonic alpha power ………………………………………………………..…..45
3.4. Discussion………………………………………………………………...…46 3.4.1. Evoked P300 response and lower frequency oscillations………....46 3.4.2. Phasic Increases in the alpha power as a function of correct detection………………………………………………………….47 3.4.3. Tonic levels of alpha power: a marker of attentional effort?...... 48 3.4.4. Distinctions between LFP and Single Unit Activity…….…….....49 3.4.5. Duration- and detection- related changes in the evoked potentials.51 3.4.6. Cholinergic input and the production of theta and alpha rhythms..52 3.4.7. Parietal functions in visual attention performance of rats………...53
4. Experiment two…………………………………………………………………...... 66 4.1. Introduction…………………………………………………………………66
4.2. Specific Methods………………………………………………...... ……...68 4.2.1 Electrode and infusion cannula implantation……………………....68 4.2.2 Neurophysiological recording sessions…………………………….68
4.3. Results………………………………………………………….…………...70 4.3.1. Histology………………………………...... 70 4.3.2. Behavioral performance under standard and distractor conditions…………………………...….…………...…71 4.3.3. Significant activation of PPC neuronal activity during the detection of signals……………………....……...... 72 4.3.4. Effects of signal duration on signal-evoked PPC activity.……….73 4.3.5. Distractor-induced modulation of signal-evoked activation of PPC neurons………………………………….……74 4.3.6. Cholinergic deafferentation increased distractor-related PPC unit activity……………………...... 75 4.3.7. Cholinergic deafferentation reduced signal-evoked PPC unit activity ………………………………………………….….75 4.3.8. Cholinergic modulation of the SNR under attentionally challenging conditions………………………………………………………..76 4.3.9 Correct versus incorrect trials…………………………………….78
4.4. Discussion…………………………………………………………………..78 4.4.1. Effects of signal duration and distractor……………………...…..79 4.4.2. Cholinergic modulation of PPC function in rats………………....80 4.4.3. Cholinergic modulation of signal-evoked activity…………….…80 4.4.4. Cholinergic suppression of distractor-related activity…………...81
5. Experiment three………………………………………………………………………97
viii 5.1. Introduction………………………………………………………………….97
5.2. Specific Methods…………………………………………………………..102 5.2.1. Subjects…………………………………………………………..102 5.2.2. Surgery………………………………………………………..…103 5.2.3. Histology………………………………………………………..104 5.2.4. Behavioral Measures……………………………………………104 5.2.5. Neurophysiological Measures………………………………….105
5.3. Results……………………………………………………………………..105 5.3.1. Effects of response cost and time outs on premature responding..105 5.3.2. Histology…………………………………………………………106 5.3.3. Initial lesions do not impair behavioral performance ...…………107 5.3.4. Bilateral lesions increase the occurrence of omissions on nonsignal trials…………………………………………………………….107 5.3.5. Bilateral mPFC cholinergic loss produces a persistent distractor- induced side bias……………………………………………….108 5.3.6. Cholinergic deafferentation of mPFC reduced signal-evoked PPC unit activity in distractor trial blocks…………………..…109
5.4. Discussion………………………………………………………...... 110 5.4.1 Parameter manipulations inducing conditioned inhibition increase errors of omission…………………………………….111 5.4.2. Cholinergic deafferentation of mPFC and attentional performance…………………………………………………….112 5.4.3. Rat PPC may be required to resolve competition between two stimulus-response contingencies in attention tasks…………….114
6. General Discussion…………………………………………………………………..121 6.1. Summary of main findings……………………….………………………..121 6.2. Signal duration dependence of the p300 amplitude……………………….122 6.3. Effects of cholinergic deafferentation on signal-related cortical Processing……………………………………………………………..123 6.4. The role of the medial prefrontal cortex in attentional processing……….127
List of References……………………………………………………………………...131
ix
LIST OF TABLES
Table
1. Behavioral correlates of PPC unit activity during sustained visual attention in Experiment 2……………..………………………………………………………85
2. Behavioral correlates of PPC unit activity during distractor sessions…………...85
x
LIST OF FIGURES
Figure Page 2.1. Schematic illustration of the response rules and contingencies of the sustained attention task…………………………………………………………………….33
3.1. Illustration and example of the final neurophysiological recording sites……….56
3.2. Behavioral effects of signal duration and visual distractor……………………..57
3.3. Correlates of duration and detection in the PPC evoked response……………...58
3.4. Low frequency oscillations are enhanced as a function of signal duration……..59
3.5. Correct detection of signals is preceded by increases in alpha power……….....61
3.6. Upper gamma frequency is elevated when the distractor light is off…………...63
3.7. Task performance and changes in tonic and phasic alpha band power during the distractor trial block…………………………………………………………64
4.1. Experimental design for Experiment 2…………………………………………..86
4.2. Electrode cannula probe placement and AChE staining of PPC……………...…87
4.3. Behavioral effects of signal duration, distractor, and lesion on task performance…………………………………………………………………89
4.4. Example of a raster plot and histogram of a single PPC neuron during four response types…………………………………………………………………...90
4.5. Stimulus locked population PETHs showing PPC responses to signals for trials yielding correct and incorrect behavioral responses……………………………91
4.6. Effects of SAP on the distribution of signal and distractor correlates…………...92
4.7. Effects of SAP and distractor on baseline firing rate of signal-evoked and distractor-related neurons……………………………………………………….94
xi 4.8. Effects of SAP on PPC signal-related activity on hit and miss trials during distractor sessions……………………………………………………………….96
5.1. Changes in task parameters and experimental design for Experiment 3……….115
5.2. AChe staining in SAP lesioned mPFC slices relative to controls and PPC slices………………………………………………………………………117
5.3 Behavioral performance of animals following unilateral and bilateral lesions of mPFC…………………………………………………………………….…118
5.4. Interaction of distractor and mPFC SAP infusion on signal-related PPC activity…………………………………………………………………...119
xii
xiii
CHAPTER 1
INTRODUCTION
1.1. Posterior Parietal Cortex Contributes to Attentional Processing
The posterior parietal cortex (PPC) plays a role in many facets of cognition, including visuospatial attention, the transformation of sensory cues into motor output, and general representations of space (Critchley, 1953; Posner and Raichle, 1994;
Ungerleider, 1995; Driver and Mattingly, 1998; Colby and Goldberg, 1999; Andersen and Buneo, 2002). Clinical studies of patients with parietal lesions indicate that such patients have disorders of symbolic thought, acalculia, and visuospatial neglect
(Critchley, 1953). In the neural network model of attention described by Posner and
Peterson (1990), the PPC, along with the pulvinar and superior colliculus, comprise the posterior attention network required to orient a subject to a location in space. The PPC is specifically involved in disengaging from a currently attended target, the superior colliculus moves the “spotlight of attention” to a new location, and the pulvinar is involved in engaging attention to the new target. Thus, the blockade of PPC function via brain damage or temporary inactivation (using techniques such as transmagnetic stimulation, as reviewed in (Rushworth and Taylor, 2006) impairs a search for a target
1 defined by multiple, complex features, but not targets defined by a single feature that
“pop out” (Ellison et al., 2003; Ellison et al., 2004; Rushworth and Taylor, 2006), because complex search tasks require serial shifts of attention. Patients with parietal lesions are impaired at filtering irrelevant distractors, and have detection thresholds similar to primates with lesions of extrastriate systems (De Weerd et al., 2003a;
Friedman-Hill et al., 2003; Buffalo et al., 2005).
The PPC is hypothesized to be a source of top-down control of extrastriate areas of visual cortex (Kastner et al., 1999; Hopfinger et al., 2000; Hopfinger and West,
2006). Anatomically, there are rich reciprocal anatomical pathways between PPC and visual processing areas (Cavada and Goldman-Rakic, 1989, 1991), suggesting a putative circuitry for this top-down control. Several imaging studies in humans have shown activation of PPC in tasks that require top-down attentional control (Corbetta et al., 1993; Coull et al., 1998). The allocation of attention to cues in different modalities increases cerebral blood flow in the PPC, whether these shifts allocated attentional resources spatially or non-spatially (Shomstein and Yantis, 2004, 2006).
Neurophysiological studies of attention show that multiple stimuli in the receptive field of extrastriate visual neurons must compete for the response of the cell.
When attention is directed to a target in the receptive field, the response is biased in favor of the target and distractor stimuli are filtered out (McAdams and Maunsell,
1999; Reynolds et al., 2000; Chelazzi et al., 2001; De Weerd et al., 2003b; Bichot et al.,
2005; Buffalo et al., 2005). Zhang and Barash (2000; 2004) demonstrated that stimulus presentation produces an initial wave of activation of the PPC, and a second longer latency activation produced by top-down mechanisms signaling a cognitive evaluation
2 of a cue. In this experiment, the authors used different color cues to direct saccades of primates toward or away from the cue, in order to control for spatial location of the visual stimulus. Monkeys were trained first to saccade to the cue after a delay, and then trained to perform an antisaccade to a different cue. This top-down activation of PPC may be required for a subject to saccade away from a target, and is thought to represent a remapped visual response to an oppositely directed stimulus. This evidence suggests that PPC may be a site at which top-down behavioral context is reconciled with incoming bottom-up stimuli.
The role of the PPC may then be to bias the detection and selection of sensory inputs from multiple modalities and to project target information to motor areas. A series of neurophysiological studies provide evidence that indicates that PPC neural activity represents the intentions of a subject to move in space, and that the PPC acts to guide effectors such as hands and eyes throughout space (Mountcastle et al., 1974;
Kalaska, 1996; Snyder, 2000; Andersen and Buneo, 2002; Scherberger et al., 2005).
Contrasting evidence from a different lab indicates that PPC activity correlates with covert shifts in attention in the absence of effector movement (Colby and Goldberg,
1999; Bisley et al., 2004; Bisley and Goldberg, 2006; Ipata et al., 2006). Additional evidence has shown that PPC encodes “salient” visual stimuli, which can be defined as having immediate behavioral relevance as a result of abrupt onset of a stimulus in a stable setting or as dictated by behavioral context (Gottlieb et al., 1998). An integration of the two neurophysiological models of parietal function complements the observations found in clinical research from both brain damaged patients and transient parietal disruption through transcranial magnetic stimulation, namely that loss of
3 parietal function results in impairments in attending to and moving through contralateral space (Posner and Petersen, 1990; Rushworth and Taylor, 2006).
1.2. Identification of the Rodent Homologue of the PPC
Anatomically, the PPC of primates has been localized to subsections of
Brodmann area 7, and is primarily interconnected to aspects of the visual system, including the frontal eye fields, pulvinar, ventrolateral thalamic nuclei (Leichnetz,
2001), and superior colliculus (Pare and Wurtz, 1997). In rats, the region considered to be a homologue to the primate PPC is generally defined as a region 3.5-5.0 mm caudal to bregma and extending 1.5-5.0 mm lateral from the midline (Reep et al., 1994). This area is connected to auditory, somatosensory, and visual cortical areas, as well as ventral orbital cortex. Also, this region has reciprocal connections with the lateral dorsal and lateral posterior thalamic nuclei, similar to that of primates (Chandler et al.,
1992). Further, basal forebrain cholinergic neurons project to the PPC of rats (Bucci et al., 1999). Thus, these findings support the general hypothesis that the PPC of rat is homologous to that of primates, and that the PPC is important for integrating multiple modes of sensory input for attentional processing.
Although the imaging and neurophysiological studies mentioned earlier have been useful for determining the cognitive functions of the PPC and its subregions, relatively few studies have investigated the neurotransmitter systems underlying these processes (Davidson et al., 1999). The rat provides a good animal model for investigating neurotransmitter systems in cognitive tasks, and as such it is important to note functional homology between rat and primate PPC. Behavioral studies of rat
4 parietal function have demonstrated that unilateral parietal lesions impair the ability of rats to attend to visual, auditory, and tactile cues in contralateral sensory fields (Kolb and Walkey, 1987; King and Corwin, 1993). Interestingly, these results were replicated when the fibers between the PPC and prefrontal cortex were severed, suggesting that the parietal and prefrontal cortex worked together as part of the attentional network
(Burcham et al., 1997). Additionally, mechanical lesions of the PPC in rodents produced impairments in incorporating allocentric, but not egocentric cues in maze performance (Kolb and Walkey, 1987; McDaniel et al., 1998). Although these findings support the hypothesis that parietal function in rats is similar to that of primates, it is important to note that parietal lesions produce minimal deficits in certain tasks. Most notably, PPC lesions fail to disrupt performance on tasks that require discrimination between valid and invalid cues (Rosner and Mittleman, 1996).
Specific loss of cholinergic input to the parietal cortex resulted in a failure to process conditioned stimuli that predict changes in the value of unconditioned stimuli, an effect interpreted as attentional in nature (Bucci et al., 1998). This data has since been interpreted as demonstrating that the ACh reports a mismatch between bottom-up stimulus processing and top-down biasing and provides the plasticity through which the contextual framework can be updated (Yu and Dayan, 2002, 2005).
The sustained attention paradigm is one useful model of studying attention in rats. Initially developed by Bushnell and colleagues (Bushnell et al., 1994), and further modified and validated for visual attention (McGaughy and Sarter, 1995), the sustained attention tasks utilizes a variable intertrial interval (10±3 sec), variable signal duration, and pseudorandom presentation of signal and nonsignal trials. Bilateral cholinergic
5 lesions of the basal forebrain (McGaughy et al., 1996), and blockade of NMDA receptors in the basal forebrain impaired performance on signal trials (Turchi and
Sarter, 2001a; Kozak et al., 2006), whereas performance on nonsignal trials was spared.
This task is hypothesized to require an animal to suppress associational information related to nonsignal performance and access incoming sensory information (signals).
This switching from associational processing to input processing is hypothesized to be cholinergically mediated, in contrast to the effects on signal detection in tasks with a nonsignal reference (Sarter et al., 2005). Results from our lab have shown that single unit activity in parietal cortex is increased on correctly detected signal trials, but not on missed signal trials or nonsignal trials (Broussard et al 2006). A visual distractor has been successfully used in this task to reduce the detection of signals while elevating prefrontal unit activity (Gill et al., 2000) and frontoparietal ACh efflux (Himmelheber et al., 2000), findings which have been interpreted as meeting the rise in attentional demand in this task. Given the necessity of ACh neurotransmission for proper attentional performance (McGaughy et al., 1996; Burk et al., 2002; Martinez and Sarter,
2004), we hypothesize that ACh input to the PPC is necessary for an animal to detect visual signals, particularly in a task successively presenting signal and nonsignal trials.
1.3. The Role of ACh in the Evoked Response: Lessons from Primary Sensory
Cortex.
Although ACh has been implicated in several aspects of cognition, including sensory processing (Roberts et al., 2005; Zinke et al., 2006), attention, learning and memory (Gold, 2003; Sarter et al., 2003) and consciousness (Whitehouse, 2004) the
6 mechanisms underlying the role of ACh in cognition are not well known. In vivo electrophysiological recordings from visual cortex of cats and non-human primates have provided some insight into the mechanisms by which ACh modifies the plasticity of cortical receptive fields. A series of landmark studies from Wiesel and Hubel
(1963a; Wiesel and Hubel, 1963b, 1965) showed that early deprivation of vision in one eye results in a functional disconnection of the V1 from the deprived eye; following re- opening, the deprived eye had little to no visual capacity, and a majority of V1 cells were activated only by stimulation of the non-deprived eye. Visual cortex plasticity has since become an important model for experience dependent plasticity.
Several studies provide evidence for the role of ACh in this type of plasticity.
The blockade of muscarinic, but not nicotinic, receptors during monocular visual deprivation blocked the functional disconnection of the V1 (Gu and Singer, 1989).
More specifically, the blockade of M1 receptor with pirenzepine replicated this effect, whereas blockade of M2 receptors did not affect plasticity (Gu and Singer, 1993). Also, the iontophoretic application of ACh generally facilitated the response of single neurons to visual stimuli (Sato et al., 1987). Cholinergic agonists iontophoretically applied have also been shown to modify orientation tuning in the visual cortex (Muller et al., 1993). Importantly, the iontophoretically applied ACh paired with the presentation of a non-preferred orientation bar of light gradually shifts the cortical receptive field. The original optimal orientation generates a weaker response and the orientation bar paired with ACh application becomes the preferred orientation (Greuel et al., 1988). A more recent study indicates that cholinergically-mediated cortical plasticity also occurs in the motor cortex in learning a reaching task (Conner et al.,
7 2003; Conner et al., 2005), and is necessary for functional recovery of this task following brain injury.
An experiment by Kimura and colleagues illustrates the complexity of ACh modulation of cortical activity and its possible mechanisms (1999). In in vitro optical recordings, ACh suppressed the spread of excitation throughout the cortical layers when layers I/II were stimulated, but the suppression is much lower following stimulation of afferent input, an effect that is glutamate dependent. Based on this evidence, ACh is interpreted to suppress activation from local connections and facilitates activation from distal afferents (Kimura, 2000).
Ultrastructural localization of muscarinic receptors indicate that of the five muscarinic receptor subtypes, M1, M2, and M4 receptors are the most abundant in the cortex and hippocampus (Flynn et al., 1995). Cortical M1 receptors are enriched on pyramidal cells, particularly in layers II/III and VI. In the prefrontal and primary visual cortex, M1 receptors were localized at postsynaptic dendrites and spines receiving both asymmetric and symmetric synapses, suggesting that the M1 receptor facilitates glutamatergic, cholinergic, and possibly GABAergic effects (Mrzljak et al., 1993).
M1 receptors enhance intrinsic activity through three putative mechanisms.
Activation of the M1 receptor reduces K+ conductance in cortical neurons, which may enhance depolarization in response to excitatory input (McCormick and Prince, 1985).
Also, activation of the M1 receptor results in phosphatidyl inositol (PI) hydrolysis, which produces increasing concentrations of inositol 1, 4, 5 triphosphate, releasing Ca2+ from intracellular stores. A second product of PI hydrolysis is diacylglycerol, which stimulates protein kinase C, which produces a variety of downstream responses (Gu,
8 2003). However, a recent study indicates that the phasic activation of M1 receptors provides a hyperpolarization, and that tonic activation produces a depolarizing potential
(Gulledge and Stuart, 2005). The authors of this study argue that the hyperpolarizing and depolarizing responses to M1 activation influence information processing. The hyperpolarizing response may be effectively reducing the influence of intracortical connections, and the depolarizing response facilitates excitation of stimulus processing.
Thus, the proper timing of M1 receptor activation may be crucial for cholinergic enhancement of relevant stimuli.
The M2 receptor in the cortex is primarily coupled to the inhibitory Gi protein which inhibits adenylate cyclase, resulting in lowered K+ conductance and Ca2+ levels, inhibiting cellular activity (Volpicelli and Levey, 2004). Initially thought to be presynaptic autoreceptors inhibiting the release of acetylcholine, immunoelectron microscopic studies of primary visual cortex demonstrated M2 immunoreactivity is primarily in presynaptic axons making asymmetric synaptic contacts, suggesting that these are heteroreceptors that serve to modulate glutamatergic as well as cholinergic activity (Mrzljak et al., 1996). In frontal cortex, M2 immunoreactivity was observed on pyramidal cells in layer III and V and at non-pyramidal neurons at all layers of cortex.
Interestingly, a majority of M2 immunoreactivity persisted in the cortex following the injection of the specific cholinergic immunotoxin anti-p75-saporin into the nucleus basalis, suggesting that this receptor subtype resides mainly on non-cholinergic terminals (Mrzljak et al., 1998). An immunoreactivity study labeling both the M2 receptor and GABA indicate a large proportion of GABAergic dendrites also have M2 receptors (Erisir et al., 2001). The colocalization of GABA and M2 receptors in the 9 cortex indicate that M2 may act to disinhibit cortical activity via inhibitory action on the
GABA interneurons. In the parietal cortex, M2 receptors are localized on the supragranular layers of more caudal PPC near the occipital cortex. In the anterior portion of the PPC, a region involved with the production of hand movements, the M2 receptor is localized in the deeper infragranular layers (Scheperjans et al., 2005). This anatomical heterogeneity suggests a functional heterogeneity, but speculation about this differentiation is beyond the scope of this thesis.
Although the importance of muscarinic receptors in plasticity and evoked potentials was highlighted in many of the studies referenced above, there have been several studies indicating a role of nicotinic receptors in attention, learning. and memory (McGaughy et al., 1999; Dani and Bertrand, 2007). There are two broad classes of nicotinic receptors, which are comprised of a diverse set of subtypes. One class of receptors have high affinity for nicotine and other agonists but not α- bungarotoxin (αBgtx), while others bind to αBgtx, and bind to nicotine with a lower affinity (Gotti et al., 2006). In the cortex, the principle nicotinic receptor subtypes are
α7 and α4β2, most of which presynaptically control release of neurotransmitters. The
α7 receptor modulates release of glutamate in the cortex (Sher et al., 2004), whereas the
α4β2 receptor modulates dopaminergic and cholinergic release in the cortex (Cao et al.,
2005). Young β2 knockout mice do not show any significant changes in performance on a number of cognitive tasks (e.g. Morris water maze, fear conditioning, memory, etc.) but aged β2 knockout mice show consistent behavioral deficits in these tasks (Zoli et al., 1999).
10 The long term potentiation (LTP) of evoked responses in the hippocampus is a putative marker of learning. In hippocampal slices, nicotinic agonists have shown to facilitate hippocampal long-term potentiation (LTP) by acting on these two types of nicotinic receptors (McGehee et al., 1995; Matsuyama and Matsumoto, 2003). A recent in vitro study indicated that the activation of nAChRs facilitates the transition from short-term to long-term potentiation in a time dependent manner. In the presence of atropine, iontophoretically applied ACh induces a series of action potentials. If electrical stimulation coincided with ACh induced action potentials LTP is produced.
Stimulation within 1 s of the final ACh induced action potential shifts the hippocampal output to LTD (Ge and Dani, 2005). Taken together, these findings suggest an important role for nicotinic receptors in cognition.
Current theoretical models regarding the modulation of sensory processing by
ACh are integrating the contributions of muscarinic and nicotinic subtypes into a broader conceptual model. According to a recent model proposed by Zinke and colleagues (2006), ACh reduces lateral cortical integration by acting on M2 receptors, which are typically bound presynaptically to local interneurons (Mrzljak et al., 1996;
Kimura, 2000). Nicotinic α4β2 receptors facilitate presynaptic glutamate release from the thalamocortical afferents (Hasselmo and Bower, 1992; Vidal and Changeux, 1993;
Gioanni et al., 1999). Also, activation of M1 receptors on pyramidal cells cause cell depolarization, increased excitability, and reduced spike frequency adaptation of these cells. Although this model applies to primary visual cortex, it could represent some mechanisms by which ACh produces a shift from local intracortical processing to heightened thalamocortical processing, a shift which may play a role in signal detection
11 (Hasselmo and McGaughy, 2004; Sarter et al., 2005). A recent study characterized the inhibition of unitary (i.e. one synapse) cortico-cortical connections in the presence of carbachol and nicotinic agonists (Levy et al., 2006). Recordings from somatosensory cortex in vitro indicated that when a layer 5 pyramidal cell was stimulated, neighboring cortical cells <100 µm away generated excitatory post-synaptic potentials (EPSPs). In the presence of carbachol, the EPSP of cells neighboring the stimulated neuron was reduced relative to baseline, an effect that was blocked by atropine. Further investigation showed that blocking M2 receptors was more effective than blocking M1 receptors in reversing the carbachol-induced suppression. Nicotinic agonists reduced unitary EPSPs only in the absence of Mg+2, which suggest that only in the presence of stimulation significant enough to unblock NMDARs would induce unitary intracortical suppression.
1.4. Defining the Role of Cholinergic Input in Cortical Synchrony
Mammalian arousal is associated with low amplitude desynchronized electroencephalographic (EEG) activity, whereas high-amplitude, low-frequency synchronous activity is associated with epilepsy, anesthesia, and slow-wave sleep. Slow wave sleep is specifically characterized by prominent delta waves (0.5-4 Hz) and sleep spindles. Transitions between sleep to a waking state are driven by an inhibitory pedunculopontine (PPT) cholinergic input to spindle-generating reticular neurons and simultaneous depolarization of delta-generating thalamocortical neurons (Steriade and
Descarries, 2006). Nucleus basalis neurons are most active during waking states and decrease their firing as the power in cortical delta waves increase (Buzsaki et al., 1988).
12 More recently it has been shown that immunohistochemically-labeled NB cholinergic neurons rhythmically burst during higher states of arousal (Manns et al., 2000; Lee et al.,
2005). These and other studies (Detari et al., 1999) provide ample evidence that PPT and NB cholinergic input modulate cortical arousal.
Although global, low frequency synchrony is associated with states of unconsciousness, the synchronization of local cell assemblies within specific cortical areas is hypothesized to bias input selection, facilitate neuronal plasticity, and correlate with attention (Buzsaki and Draguhn, 2004). Recordings from the motor cortex of monkeys (Murthy and Fetz, 1996; Baker et al., 1999) and humans (Aoki et al., 2001) indicate an increase in gamma (30-60 Hz) activity during precise grasping or threading movement. In Macaque extrastriate cortex (V4), visuospatial attention enhances synchrony between single unit and local field potential (LFP) data (Fries et al., 2001), and parietal neurons are similarly synchronized to the gamma frequency during a memory-saccade task (Pesaran et al., 2002). In the hippocampus, the theta frequency (4-
8 Hz) can modulate the firing rate and timing of a single neurons and can gate task- relevant place cells in rats (Huxter et al., 2003). A new report from subdural electrode recordings in humans reports a coupling of theta and high frequency gamma during a challenging working memory task (Canolty et al., 2006). Interestingly, the trough of theta gates both the higher frequency spike trains in rats and the high gamma oscillations in humans (Canolty et al., 2006; Hasselmo, 2006). This evidence supports the hypothesis that transient coordination of neuronal assemblies facilitates adaptive behavior. The current proposal seeks to test hypotheses about transient synchronization,
13 as measured by LFP recorded from the PPC during the detection of visual signals and correct rejection of nonsignals in a sustained attention task.
Hypotheses regarding the role of cholinergic input to the cortex in sensory processing have previously been tested by measuring extracellular post-synaptic potentials to exogenous stimuli. In typical studies, an animal is anesthetized and evoked responses are recorded from primary sensory cortex (auditory, visual, or somatosensory) while an animal is presented with different stimuli. The cholinergic system has been shown to enhance somatosensory evoked potentials, and this enhancement is thought to be mediated by NMDA receptors in cortex (Verdier and
Dykes, 2001). These authors hypothesized that basal forebrain activation results in
GABAergic disinhibition of cortical interneurons, followed by the cholinergic attenuation of potassium permeability, producing an environment favorable for opening
NMDA channels. Loss of cholinergic input to the cortex results in minor changes in basal EEG (Wenk et al., 1994; Berntson et al., 2002). Infusion of AMPA and NMDA into the basal forebrain result in increases in the power of the high frequency gamma band, but this change is small relative to the change in ACh efflux (25% to 200-300%;
(Fournier et al., 2004). In contrast, the considerable effects of cholinergic deafferentation on auditory evoked potentials in rats (Berntson et al., 2003b), or blockade of muscarinic receptors on eye blink potentials in rabbits (Wang et al., 1999) suggests that cholinergic modulation of EEG may be more robust in performing animals.
Further evidence of the necessity of cholinergic neurotransmission in learning and memory comes from studies investigating exogenous (i.e. electrical) stimulation of
14 the basal forebrain, when paired with a neutral CS tone, develops receptive field plasticity in auditory cortex (Edeline, 1993). This conditioning paradigm also induces a
“behavioral memory” (i.e. predictable changes in heart rate) and increases in the gamma frequency in the auditory cortex, which is an indicator of neural activation (McLin III et al., 2002). In anesthetized rats, stimulation of specific sites in the basal forebrain would enhance evoked-responses in somatosensory or visual cortex. Since the authors could not identify the type of NB neurons they stimulated, there are two putative mechanisms for these changes. One is that exogenous stimulation could be directly driving cholinergic neurons to fire, thus increasing ACh efflux, subsequently facilitating stimulus-specific changes in evoked activity. There is support for this explanation, as increases in ACh efflux and unit activity was in fact found in somatosensory and visual cortex following exogenous stimulation of NB (Jimenez-Capdeville et al., 1997). Also, atropine sulfate blocked conditioning-enhanced somatosensory evoked-potentials
(Maalouf et al., 1998; Dykes et al., 2001), indicating that NB ACh neurons are involved in the enhancement of evoked potentials and neuronal plasticity in the cortex.
Application of carbachol induces gamma oscillation in the primary visual cortex of cats only when it is paired with the presentation of visual stimuli, indicating a direct cholinergic contribution to the gamma frequency, which enhance the neuronal response to stimuli (Rodriguez et al., 2004). These results indicate a prominent role of ACh in the transient synchronization of neuronal assemblies.
A second mechanism by which NB stimulation can produce plasticity is through GABAergic NB neurons with axons projecting to the cortex directly inhibiting cortical neurons. Since many of these projections contact GABAergic interneurons in
15 the cortex (Freund and Gulyas, 1991), activation of NB could produce cortical disinhibition. Simultaneous recordings of PFC and NB in freely-moving rats indicate a population of NB neurons that was not responsive to changes in the sleep-wake cycle
(thus putatively non-cholinergic). These NB neurons would synchronize and fire spike bursts, producing transient increases in PFC gamma frequency in the awake state (Lin et al., 2006). The authors propose that these neurons are GABAergic projection neurons, and it has been proposed that these neurons may control coupling of theta and high- frequency oscillations necessary for task-performance in humans (Canolty et al., 2006).
The prevalence of muscarinic receptor immunoreactivity on GABAergic neurons in the primary visual cortex of macaques is evidence that the GABAergic and cholinergic contributions to states of cortical arousal may not be readily dissociable (Disney et al.,
2006).
1.5. Frontal Cortex and Effortful Cognitive Control
Frontal cortex is critical for a broad range of goal-directed behavior, which generally requires the ability to maintain goal-relevant information in mind and bias associational and sensory resources towards processing relevant stimuli while ignoring irrelevant stimuli (Coull et al., 1998; Swick and Knight, 1998). There is a close functional relationship of the PPC and frontal areas, particularly the lateral prefrontal and anterior cingulate cortices, which are critical for mediating the allocation of resources between competing stimulus-response associations (Mesulam, 1981; Jonides et al., 1998; Cabeza et al., 2000; Crone et al., 2006; Sarter et al., 2006). The co- activation of PPC and frontal areas is commonly found during tasks requiring the
16 filtering of irrelevant stimuli (Hazeltine et al., 2000; Marois et al., 2000; Lee et al.,
2006), yet with progressive increases in attentional demand the role of these two regions can be dissociated. Bunge and colleagues (2002) found that PPC areas were activated on trials in which both congruent and incongruent flanking distractors (i.e. congruent distractors flanked the target but cued the responder to make the same response as the target, incongruent distractors flanked the target and directed the responder to make a response opposite of the target). Frontal cortex was only significantly activated on trials with incongruent distractors. These results suggest that
PPC activation was sufficient for maintaining stimulus-response associations of a well- learned attention task, and that PFC was required only on trials which provide conflicting response alternatives.
There is ample evidence that PFC is involved in the suppression of irrelevant stimuli in posterior cortical areas. Patients with PFC lesions have exaggerated evoked- responses to irrelevant somatosensory and auditory stimuli (Yamaguchi and Knight,
1990). This effect was not replicated in patients with parietal lesions, and controls from this study indicate a direct suppression at the thalamic or sensory level. A recent study by Nelson and colleagues (2005) investigated some of the mechanisms by which PFC can modulate PPC processing. In this study perfusion of AMPA and carbachol (a nonspecific ACh agonist) into the PFC increased ACh efflux distally in the PPC.
Perfusion of nicotine and NMDA into the PFC did not increase PPC ACh efflux, suggesting that muscarinic receptors in the PFC are necessary for PFC to elicit ACh efflux in the PPC. Perfusion of carbachol or nicotine throughout the PPC, while eliciting increases in local ACh efflux, failed to modulate PFC ACh levels. Moreover,
17 local administration of AMPA into the PPC failed to elicit ACh efflux. These findings suggest PFC can broadly regulate the activity of the cholinergic system at distal sites within the cortex.
Anatomically, the NB and substantia innominata (SI) receive substantial glutamatergic input from the prefrontal cortex which has led many to suggest that the
PFC can modulate cortical ACh by regulating basal forebrain cholinergic nuclei
(Zaborszky et al., 1997; Golmayo et al., 2003). Orthodromic stimulation of primary sensory cortices produces activation of modality specific frontal sites, and orthodromic stimulation of these frontal sites result in activation of the basal forebrain (Golmayo et al., 2003), suggesting a feed-forward loop between primary sensory, frontal, and basal forebrain sites. In an elegant study using antisense oligonucleotides to disrupt NMDA receptors in the SI, Turchi and Sarter (2001b) were able to demonstrate a marked attenuation in the detection of relevant signals. Blockade of NMDA receptors via infusion of APV in the basal forebrain causes a selective impairment in the detection of signals, while correct rejection on nonsignal trials is maintained. More recently, this finding has been replicated in an operant dialysis study, which demonstrated that NB infusion of APV increased cortical ACh efflux in prefrontal cortex (Kozak et al., 2006).
The increased levels of prefrontal ACh are hypothesized to be necessary to maintain attentional performance during challenging conditions.
Based on these data, it is hypothesized that cholinergic input to the medial prefrontal cortex (mPFC) is necessary to optimize behavioral performance under increasing attentional demand. By contrast ACh input to the PPC is hypothesized to mediate attentional processing in standard task performance, as long as there is
18 sufficient stimulus-response competition between signal and non-signal trials. Previous work in our lab has shown that parietal activity is activated following the presentation of visual signals only on those trials resulting in a hit response. Trials in which the animal failed to detect the signal also lacked this excitatory response. The primary aim of this proposal is to validate the findings in the previous study by measuring the local field potential activity (LFP) of task performing animals, and to investigate the contributions of prefrontal cortex and basal forebrain cholinergic system to the attentional processing in the PPC.
1.6 Specific Aims
Aim #1: Test of the effects of signal duration and distractor stimuli on event-related potentials (ERPs) and oscillations in the local field potentials (LFPs) during performance of a sustained attention task.
Hypothesis 1: If signals that are more easily detected increase the amplitude of the p300 component of evoked potential in rats, then the signal with the longest duration in the sustained attention task will produce the p300 of the greatest amplitude.
Aim #1 is designed to test the general hypothesis is that the rat PPC is a functional homologue of primate and human PPC. If this were the case, then rat PPC would produce similar neurophysiological correlates of attention to human and primate
PPC. The ERP is hypothesized to represent synaptic potentials, afterhyperpolarization of somatodendritic spikes, and voltage-gated oscillations of the membrane potential
19 (Logothetis, 2003b). These periodic oscillations are thought to represent synaptic input and periodically elevate the potential close to threshold, providing discrete windows for a neuron to respond to external stimulation (Buzsaki and Draguhn, 2004). The synchronization of neural ensembles to periodic oscillations increases the efficacy of neural representation of stimuli and may be a correlate of attentional selection
(Steinmetz, 1995). In humans, signals that were easiest to discriminate from the background produced a p300 component of the greatest amplitude (Yordanova et al.,
2001). Thus, we will test the specific hypothesis that signals of the longest duration will produce a p300 component of greater amplitude than shorter signals.
Hypothesis #2: If power in the alpha band represents the allocation of attentional resources, then increased alpha power in response to the visual distractor will correlate with higher false alarm rates.
Changes in the alpha band are hypothesized to represent the top down expectation of a signal. In tasks employing a retention interval, such as GO/NO-GO or working memory tasks, power in the alpha band increases late in the retention interval.
Further, when signals are predictable, power in the alpha band increases prior to the onset of the signal. Thus, it is predicted that power in the alpha band will increase prior to the tone on signal trials. By contrast, because the onset or trial type (signal, nonsignal) is not predictable, we predict that alpha power will not increase in the intertrial interval during standard trial blocks.
The presence of the visual distractor produces significant increases in the false alarm rate in this task, indicating that rats are attending to the visual distractor as if it
20 were a behaviorally relevant signal. Over the course of a distractor trial block, the false rate declines as rats habituate to the distractor. Due to the predictable occurrence of the visual distractor, it is hypothesized that alpha power will increase prior to the onset of the visual distractor. Further, it is predicted that as rats decrease their false alarm rate during the distractor block, distractor-related increases in alpha power will be attenuated.
Aim #2: Test of the effects of 192 IgG saporin infusions on signal-evoked PPC unit activity using in vivo, freely-moving neurophysiological recordings.
Hypothesis #1: If cholinergic transmission contributes to signal-evoked responses within the PPC, then local cholinergic deafferentation will attenuate the signal- evoked response of PPC neurons.
Parietal neural activation is hypothesized to represent a salience map, in which constituent neurons encode locations in visual space, and the salience or intensity of representation of visual objects is reflected in the firing rate of these neurons (Colby and Goldberg, 1999; Itti and Koch, 2001). In rats, PPC neuronal activity increases as visual signals are detected but not when signals are missed or during performance on nonsignal trials (Broussard, 2006). Given the integral role of the basal forebrain cholinergic system for the detection of visual signals, but not on performance during nonsignal trials, it is of particular interest as to whether cholinergic input mediates attention-related activation in the PPC. The hypothesis that the attention-related activation is dependent upon intact cholinergic input to the PPC will be tested by
21 removing cholinergic input to the PPC via local infusion of 192 IgG saporin. It is predicted that this manipulation will attenuate signal-evoked response of PPC neurons to the visual signal, but not to other task-related correlates. Further, we will test the interaction of distractor and lesion on signal-evoked activation. If the
Aim #3: To test the effects of ipsilateral, contralateral, and bilateral cholinergic deafferentation to the medial prefrontal cortex (mPFC) on detection-related PPC activity and behavioral performance in the presence of a visual distractor.
Hypothesis #1: If cholinergic transmission in the mPFC contributes to the production of signal-evoked PPC responses in the presence of the distractor, and the this top-down modulation of parietal neurophysiology is within hemisphere, then cholinergic deafferentation of mPFC ipsilateral to the PPC recording probe will attenuate the signal-evoked neuronal response only in the presence of the distractor.
Cholinergic input to the medial prefrontal cortex is hypothesized to mediate increases in attentional effort. Attentional effort can be define as the degree to which a motivated subject attempts to detect signals and reject nonsignals in the face of detrimental conditions. Typical methods employed to create detrimental conditions to challenge attention are increased time on task, pharmacological challenges, or environmental distractors (Sarter et al., 2006). Here we test the specific hypothesis that cholinergic input to the mPFC is necessary for optimizing detection-related activation of PPC neurons in the presence of a distractor, but not during standard conditions. As
22 in previous experiments, probes will be unilaterally implanted in the PPC of trained rats.
Then, the effects of infusions of 192 IgG saporin (SAP) into the ipsilateral or contralateral mPFC on PPC unit activity in attention-task performing animals will be assessed. Specifically, the effects of ipsilateral infusions of SAP into the mPFC on PPC unit activity of attentional task-performing rats will be compared to the effects of contralateral infusions of SAP on PPC unit activity. In vivo electrophysiological recordings from mPFC of attention-task performing rats showed increase activation in the presence of a visual distractor, an effect that was cholinergically mediated (Gill et al., 2000). Because cholinergic activation of the mPFC was sufficient to stimulate PPC cholinergic efflux (Nelson et al., 2005), we propose that cholinergically mediated increases in mPFC firing may act to suppress or filter out distractor stimuli in this task, an effect that will be reflected in PPC unit activity.
Hypothesis #2: If cholinergic transmission within the mPFC is necessary for the mediation of attentional effort, then bilateral deafferentation of the mPFC will specifically impair attentional performance in the presence of the visual distractor.
We will test the effects of bilateral SAP infusions on behavioral performance, with the specific prediction that cholinergic transmission in the mPFC will impair performance in distractor trial blocks, but not standard trial blocks. Pilot data indicated that cholinergic lesions of the mPFC did not affect performance of the sustained attention task during standard conditions, but impaired the rats ability to detect signals and reject nonsignals in the presence of a distractor (Gill et al., 1999).
23
CHAPTER 2
GENERAL METHODS
2.1. Subjects and Apparatus
Six week old male Long-Evans rats (250-300g; Harlan, Indianapolis, IN) were housed singly in climate controlled cages on a 12:12-h light: dark cycle (lights on at 6 a.m.). Rats were handled extensively upon arrival and provided food and water ad libitum until three days before training, after which water was gradually restricted to one hour per day. Rats received a water reinforcer during daily training sessions and following training received supplemental water. All animals were trained 5-7 days a week. Animal care and experimentation were performed in accordance with protocols approved by the Ohio State University Institutional Animal Care and Use committee.
The operant chambers have been described in detail elsewhere (Gill et al., 2000).
Briefly, three panel lights, three fixed levers, and a house light near the ceiling of the chamber were present on one panel of the operant chamber. The opposite panel contained a recessed water port with a water dispenser and a tone generator on the outside of the operant chamber. All chambers were placed within a sound attenuated shell. An operant chamber with a similar configuration fitted for neurophysiological recording and video recording was used for testing sessions.
24 2.2. Behavioral Training
The sustained attention task was modified from the original task characterized by
McGaughy and Sarter (1995). The current task uses fixed response levers, as opposed to retractable levers, in order to minimize electrical interference with the neurophysiological signal.
There were 4 stages of training in the sustained visual attention task. A house light was illuminated during all phases of training. Rats were initially trained to press each of two levers on an FR-1 schedule of reinforcement. Animals were not rewarded for more than five consecutive presses on a single lever in order to prevent side bias. Once rats made at least 50 responses on each lever during a 1 hr session for 3 consecutive days, they were trained in the sustained visual attention task.
The rules of the sustained attention task were presented in the second stage of training (Fig. 1). The rules required rats to detect the presence of signal events
(illumination of the central panel light for 500 ms) and correctly reject non-signal events
(central panel light remained off). Both types of events were followed 1 s later by a 200 ms tone that initiated a 4 s response window. Following the presentation of the signal and subsequent tone, a left lever press was positively reinforced with a drop of water and scored as a Hit. A right lever press on a signal trial was not reinforced and scored as a
Miss. On non-signal trials, a right lever press was positively reinforced and scored as a
Correct Rejection, whereas a left lever press on non-signal trials was not reinforced and was scored as a False Alarm. If a response was not made within 4 s, the trial was scored as an omission. Either a response or an omission initiated a variable intertrial interval
(ITI, 12 ± 3 s). At this stage of training, an incorrect response led to a correction trial in
25 which the same type of trial repeated for up to 5 consecutive times or until the rat made a correct response. Each behavioral training session included a 36 min task period preceded and followed by 5 min, task-free periods.
In the third stage, correction trials were removed and signal and non-signal trials were presented with equal probability throughout each 36 min session. For the animals to advance to the final stage of training, accurate performance on 70% on both signal and non-signal trials was required, with less than 30% omissions for three consecutive days.
During the final stage of training, the task was modified to further tax attention.
These modifications are based on findings in humans, in which variations in signal duration and increases in the event rate have been shown to challenge attentional performance (Parasuraman et al. 1987). Signals of three durations (25, 50, and 500 ms) were presented with non-signal trials and the intertrial interval was reduced to 10 ± 3 s.
Criterion was raised to 75% correct responses on both types of trials and to less than 25% omissions. Testing that occurred using this task was called a Standard Session.
After reaching criterion performance for three consecutive days, rats performed in a Distractor Session, which was identical to the final stage of training except that a distractor (house light flashing at 0.5 Hz) was presented during the second 12 min block of the task period. Previous studies using this task have demonstrated that this distractor impairs performance on signal and non-signal trials (Gill et al., 2000). Each rat received at least 3 Distractor sessions that were separated by at least 2 Standard sessions.
Following the third distractor testing session, rats were transferred to an operant chamber equipped for electrophysiological recording and required to reach criterion levels of performance on the final training stage. All rats reached criterion within 2-5 months of
26 training. The rats underwent implantation surgery following three additional distractor sessions in this new environment.
2.3. Electrode and Infusion Cannula Implantation
Two tetrodes were inserted into a 26-ga cannula (15 mm) and extended ~1 mm beyond the distal end of the cannula. The cannula and tetrodes were affixed to a moveable headstage (as described in Broussard et al., 2006). The eight lead wires were soldered into separate channels of an eight channel headstage (Plexon Inc, Dallas, TX).
Two separate Teflon-coated, stainless steel, 250 µm electrodes (A-M Systems, Everitt,
WA) were soldered into to the headstage as well and served as a reference (placed in the brain) and ground (wrapped around a screw on the skull) for recordings. A 26 ga, 17 mm, infusion cannula was placed within ~1 mm of the electrode guide cannula for the purpose of infusing saline or SAP near the recording site.
Rats trained to criterion performance were anesthetized with isoflurane gas mixed in oxygen. Body heat was maintained at approximately 37° C with a thermal pad
(Deltaphase IV, Braintree, MA). The tetrodes and a single infusion cannula were implanted unilaterally using the stereotaxic coordinates A/P -4.5 mm, M/L ± 2.5 mm, and
D/V -1.0 mm from the dura surface, in accordance with previous anatomical evidence of the rat homologue of PPC (Reep et al., 1994). Four additional burr holes were drilled into the skull into which machine screws were threaded. The reference electrode was placed in the contralateral somatosensory cortex, and the ground electrode was wrapped around a machine screw. The carrier, headstage, and stainless steel electrodes were affixed to the skull with dental cement. Lidocaine and antibiotics were applied to the
27 wound immediately after surgery. Rats were allowed one week to recover from surgery in their home cages with free access to food and water, after which access to water was reduced before resumption of behavioral testing and neurophysiological recording.
2.4. Neurophysiological Recording Sessions
Electrical potentials were collected with a head-mounted operational amplifier and sent via a cable to a commutator that relayed the signal to two differential amplifiers
(A-M Systems, Carlsburg, WA). The analog signals were amplified (10,000X), bandpass-filtered between 300 Hz and 5 kHz, and digitized by an analog-to-digital board at 250 kHz (CED Power 1401, Cambridge Electronics Design, Cambridge, England).
Signals that exhibited peak amplitude exceeding a user-defined threshold on either electrode were sampled at 11 kHz using Spike 5 software (Cambridge Electronics
Design). Multiple unit activity on each tetrode was separated into single units based on principle component analysis and k means clustering of the waveforms. The interspike interval of each neuron was examined and all units showing an ISI < 2.5 ms were discarded from further analysis. The isolated single units were recorded during two consecutive standard sessions and one distractor session. The microdrive was advanced after distractor sessions and a new population of neurons recorded. The Med PC IV software (Med Associates, St Albans, Vermont) sent a pulse to a clock board to signal each behaviorally relevant event (signal light, tone, house light, right and left lever presses during the response window and the ITI, water delivery, water port entry) with
0.1 ms precision.
28 2.5. Histology
Following the final electrophysiological recording session, animals were anesthetized with sodium pentobarbital (100 mg/kg) and the final recording site was marked with a small electrolytic lesion (15 µA for 30 s on each channel of the 2 tetrodes) using a stimulator and stimulus isolation unit (Grass Instruments, Quincy, MA). Rats were transcardially perfused with 0.9% cold saline followed by 4% paraformaldehyde.
The brains were post-fixed for 24 hr in 4% paraformaldehyde and transferred to a 30% sucrose solution in 0.1 M phosphate buffer (pH 7.4). Each brain was sliced into 50 µm sections; sections within 100 µm of the electrolytic lesion were stained for Nissl substance with cresyl violet, and remaining PPC slices were stained for AChE-positive fibers (Tago et al., 1986). These procedures determined the extent of cholinergic loss due to SAP infusion. Sections were placed for 7-10 min in 0.1% H2O2 , and then rinsed with
0.1 M maleate buffer (pH=6.2). Subsequent to rinsing, sections were incubated in a solution of 5 mg acetylthiocholine, 0.147 g sodium citrate, 0.075 copper sulfate, and
0.0164 potassium ferrocyanide in 1.0 L of 0.1 M maleate buffer for 45 min. Following completion of this incubation, sections were rinsed using 50 mM Tris buffer (pH=7.6) and placed in a secondary incubation solution. This solution contained 0.05 g diaminobenzidine (DAB) and 0.375 nickel ammonium sulfate in 125 ml of 50 mM Tris buffer. After 10 min, 20 ml of 0.1% H2O2 was added for an additional 2 min to produce sufficient cortical layering of the stain. Sections were then rinsed in 5 mM Tris buffer and mounted on gelatin-coated slides. After drying overnight, sections were dehydrated in ethanol and de-fatted in xylene prior to cover slipping. Microphotographs were taken of the Nissl-stained and AChE-stained sections of PPC and analyzed using standard
29 thresholding procedures in ImageJ (Rasband, 1997-2006). An ANOVA was then conducted comparing the percentage of black pixels in a photographed of lesioned slices to contralateral PPC (in experiment 2) or mPFC slices from control brains (experiment 3).
2.6. Behavioral Measures
The behavioral measures generated for statistical analysis include response accuracy on signal trails [hits/ (hits + misses)] and on non-signal trials [correct rejection/
(correct rejections + false alarms)], response latency, and errors of omission. The behavioral measures were generated from distractor and baseline sessions. Baseline testing sessions with >40% overall omissions were excluded from the statistical analyses.
Percentage data were angularly transformed (Zar, 2007) before analysis to correct for the skewed distribution of percentage scores. In addition, a vigilance index was used to assess overall performance. In order to determine an overall measure of performance, a
“vigilance index” (VI) was determined by collapsing the relative number of hits (h) and false alarms (fa) into one index using a modified formula for a nonparametric index of signal sensitivity (VI= (h-f)/([2(h-f)-(h+f)2]). This index varies from -1 to 1, with 1 being perfect discrimination of signals and non-signals and 0 being the loss of ability to discriminate signals and nonsignals. Finally, the proportion of responses to a particular side, or side bias (SB) was estimated for each block of trials (SB= (h+f)/ (total responses).
The SB measure varies from 0 to 1, with 0.5 indicating equal responses on each lever, 0 specifying all responses were on the correct rejection/miss lever, and 1 denoting all responses were on the hit/false alarms lever. Because shorter (25, 50 ms) signals are harder to detect than longer signals, a “neutral” side bias in this task is between 0.3-0.4.
30 Repeated-measures ANOVAs were conducted on the behavioral data using
Session (Baseline and Distractor) and Signal Duration (25ms, 50ms, 500ms) as within subjects factors for analyzing the dependent measures of signal response accuracy, non- signal response accuracy, errors of omission, and reaction time on signal and non-signal trials, as well as response lever side bias and VI. A Huyn-Feldt correction was applied to all repeated measures to control for possible violations of the sphericity assumption of homogeneity of variances.
2.7. Neurophysiological Measures
Single units exhibiting more than 800 total spikes during the recording sessions were used for the electrophysiological analysis (0.30 spikes/s). Nearly all cells detected exceeded this threshold, and the average firing rate was 2.46 ± 0.18 spikes/s (Range:
0.092 spikes/s-22.29 spikes/s). Posterior parietal unit activity was divided into 20 ms bins and changes in activity were analyzed in 2 s epochs. Analysis centered on stimulus presentation (e.g. signal light, tone, and distractor light) and behavioral responses (e.g. correct rejections, hits, misses, false alarms) which were assessed using peri-event time histograms (PETH). Because unit activity often violates the assumption of normality,
Wilcoxon signed-rank tests were used to analyze pre- and post-event epochs. A χ2 analysis was applied to assess the proportion of neuron pairs exhibiting increases or decreases in task-related activity as a function of distractor or cholinergic deafferentation.
Once the population of neurons exhibiting stimulus-driven activity was determined, a calculation of the signal-to-noise ratio (SNR) was used to quantify how much the neuronal activity exceeded the background activity when a visual stimulus was presented.
31 SNR was calculated for pre-lesion and post-lesion animals as the stimulus driven response (Rstim) divided by the sum of the stimulus driven response and background activity (Rspont) for each individual PETH (SNR = Rstim/(Rstim+Rspont). Values range from -
1 to 1, with those approaching 1 indicating that all spiking activity in the epoch is evoked by the signal.
32
FIGURES
Figure 2.1.) Illustration of the response rules and recording epochs analyzed. Signals are illuminations of a panel light (25, 50, or 500 ms), which are followed 1 s later by a 250 ms tone, which opens a 4 s operant window. Left lever responses (Hits) are rewarded (R=reward) on signal trials, and right lever presses (Misses) are not. Nonsignal trials are initiated with a tone identical to that in signal trials, right lever presses are rewarded (Correct Rejections) and left lever presses are not (False Alarms). A response or an omission (no response in 4 s) initiates a variable ITI (10±3 s) for the next trial. The 1 s interval prior to the tone on all trials is analyzed to test attention related activation of the PPC prior to a response.
33
CHAPTER 3
EXPERIMENT ONE
MODULATION OF THE PARIETAL EVENT RELATED RESPONSE
BY SIGNAL DURATION AND VISUAL DISTRACTOR.
The deployment of attentional resources between competing sets of stimulus-response contingencies is thought to be resolved in associational areas such as the parietal cortex.
Recently, neuronal activity in the parietal associational area in rodents has indicated an important behavioral correlate in a sustained attention task. Single neurons increased their firing rate as animals prior to correct detection of a visual signal. Although correct detection has been demonstrated to be a function of signal duration, the firing rate of parietal neurons were not modulated by signal duration (Broussard et al., 2006). If the parietal cortex of rats encodes the detection and selection of stimuli for further processing, one might predict that signals of varying intensity would produce correlates of varying intensity. The lack of an observed effect of signal duration on unit activity from the previous and current studies (see also Experiment 2) may be a function of technical limitations; signals of longer duration may recruit more neuronal activity from the PPC, but the small number of recorded neurons per session (~8) preclude detection of different correlates. Recording the local field potential (LFP) is a complementary dependent measure to spiking data which can be used to further validate this model. The LFP is
34 hypothesized to encode different task parameters than single unit activity in monkeys
(Scherberger et al., 2005), and so we chose to analyze the local field potentials within the parietal cortex of sustained attention task-performing rats.
There are several measures of the field potential which correlate with either stimulus properties or behavioral performance. One such measure is the p300 response, a positive potential that peaks around 300 ms post signal in humans. The p300 response is maximal in humans and primates at parietal sites (Linden, 2005). The standard paradigm used to generate the p300 response is the “oddball task”, in which infrequent targets are successively presented with frequent irrelevant targets. The amplitude of the p300 varies as a function of stimulus discriminability, and thus is an attractive measure for our study.
Longer signals in the sustained attention task (500 ms) are more easily detected, and are predicted to produce a higher amplitude p300 response. Another component of the field potential related to task performance is the long-latency (500-1000 ms post signal) contingent negative variation (CNV). The CNV was first measured from the scalp of humans and has two components. The first component is generated in anterior areas over the frontal eye field, and is developed after a stimulus calls for a decision. The second component is found over more central areas and is related to the execution of a response plan (Singh and Knight, 1990). In parietal cortex, the CNV is more likely to correspond to central generation of the CNV and reflect correct responses to targets.
Oscillatory activity among neurons has been hypothesized to play a major role in attentional processing (Singer and Gray, 1995; Klimesch, 1999; Klimesch et al., 2007). It has been well documented that oscillations in the delta (0.5-4 Hz) and theta (4-8 Hz) frequencies synchronize activity between distant cortical regions (Varela et al., 2001;
35 Buzsaki and Draguhn, 2004; Canolty et al., 2006). Persistent slow wave synchrony is a characteristic of a loss of consciousness and non-REM sleep. However, transient changes in slow wave synchrony correlate with the presentation of oddball stimuli (Spencer and
Polich, 1999; Demiralp et al., 2001), suggesting that the mammalian brain responds to highly salient, unpredictable stimuli by powerfully synchronizing several cortical regions.
Changes in the LFP can be categorized as either event-related (phasic) or tonic.
In humans, visual attention tasks produce phasic changes in the alpha band (typically 8-
12 Hz) which are characterized by increased synchronization of the alpha band in frontal sites and corresponding decreases in occipital sites (Sauseng et al., 2005a). As tasks become more demanding than simple detection, coherence of the alpha band between sites throughout the scalp increases, resulting in a state of “alpha equilibrium”, which is hypothesized to enhance cortico-cortical communication (Klimesch et al., 2007). Tonic levels of alpha band are increased with age, sleep restriction, and time on task (Boksem et al., 2005). Similarly, manipulations of various parameters increasing attentional demand increase basal levels of alpha band. These findings suggest that tonic modulation in alpha power reflect an attempt by the subject to elevate cognitive performance under challenging conditions.
Until recently, there were few animal models assessing the role of the alpha band in task performing animals. In rodents, detection of simple, predictable somatosensory stimuli produce event-related desynchronization in the 7-12 Hz frequency. However, tasks requiring compound stimuli, which require response inhibition during a cue-target interval, elicit a different phasic response in the alpha band (Wilke et al., 2006). One such study investigated the synchrony between the primary visual and parietal cortex
36 (area 7) of cats performing a GO/NO-GO task (von Stein et al., 2000). Interestingly, cues which triggered a GO response produced robust synchronization of alpha and theta frequencies between parietal and visual cortices, relative to novel, irrelevant cues. In addition, cross-correlation at these frequencies between parietal and visual cortex was greater on correct trials relative to incorrect trials. The authors hypothesized that modulation of the alpha frequency may represent a form of top-down processing, in contrast to the stimulus dependent processing found in higher frequency bands.
Previous work in our lab demonstrated an increase in the firing rate of single and multiple unit activity in the posterior parietal cortex (PPC) during the detection of visual signals in task performing animals (Broussard et al., 2006). Interestingly, variations in signal duration did not affect detection related changes in the firing rate.
Because LFPs capture the summed potentials from a large number of neurons and have been shown to correlate with different parameters of task performance than single and multiple unit activity of primate PPC (Scherberger et al., 2005), we will assess parietal
LFP activity of attention-task performing rats to determine whether signal duration modulates the population response. LFP activity is hypothesized to represent synaptic potentials, afterhyperpolarization of somatodendritic spikes, and voltage-gated oscillations of the membrane potential (Logothetis, 2003a). These periodic oscillations are thought to represent synaptic input and periodically elevate the potential close to threshold, providing discrete windows for a neuron to respond to external stimulation
(Buzsaki and Draguhn, 2004). The synchronization of neural ensembles to periodic oscillations increases the efficacy of neural representation of stimuli and may be a correlate of attentional selection (Steinmetz and Constantinidis, 1995). We will test the
37 efficacy of neural synchronization under standard conditions and in the presence of a visual distractor, a manipulation that produces impairments in the ability of rats to detect signals, that modulates signal detection-related spiking activity, and that can be used to manipulate attentional effort (Sarter et al., 2006).
In the current study, we attempt to quantify the contributions of delta and theta frequency activity to the evoked response, and their respective correlates of performance.
In our task, animals were trained for several sessions to discriminate between successive presentation of visual signals and nonsignals. Two main manipulations were made to increase demands of signal detection. Signals of varying duration were used, which produced reliable effects on accurate detections (Gill et al., 2000). The second manipulation was the presentation of a visual distractor, which produced significant decrements in both the detection of signals and the correct rejection of nonsignals
(McGaughy and Sarter, 1995). In a previous study, a similar visual distractor initially impaired performance on nonsignal trials, followed by the subjects regaining correct detection performance later in the distractor block. This re-evaluation of response strategy correlated with an increase in cortical acetylcholinergic efflux (Himmelheber et al., 2001). This measure has since been hypothesized to be a correlate of attentional effort (Sarter et al., 2006). In addition to testing hypotheses about the phasic response of the LFP to signals of varying duration, we will assess whether the distractor, or subsequent strategy shift, will produce changes in the tonic LFP activity, in particular the alpha frequency band.
38 3.2. Specific Methods
Five male hooded Long-Evans rats (Harlan, IN), age 2-3 months were trained on the sustained attention task outlined in Chapter 2. This task requires the correct detection of visual signals and the correct rejection of nonsignals. Following 3-6 months of training, animals were surgically implanted with a cannula containing tetrodes into the parietal cortex (relative to Bregma, in mm): AP -4.5, DV 1.0, L 2.5, (Paxinos and Watson,
1998). Following a 7 day convalescence period, animals were re-trained to criterion performance, after which they were tested 6-7 d/week.
3.2.1. Recording and analysis of EEG
Electrical potentials were collected with a head-mounted operational amplifier and sent via a cable to a commutator that relayed the signal to four-channel differential amplifiers (A-M Systems, Everitt, WA). The analog signals were amplified (1,000X), bandpass-filtered (low pass, 0.1 Hz, high pass; 5 kHz), and then digitized by a CED 1401
Micro data acquisition unit (Cambridge Electronics Design, Cambridge, England). To analyze the visually evoked response, the EEG signal from 1s before stimulus onset to 2s after stimulus onset was averaged across trials. The maximum amplitudes and peak latencies of the visual ERP were assessed. The amplitude of the positive potential in the
0-400 ms window following the signal onset was measured. A second window measured the negative amplitude and latency of the CNV in the 400-900 ms window post signal
Latencies were measured relative to the stimulus onset and amplitudes measured relative to the mean value in the 200 ms preceding the stimulus onset.
39 In order to investigate the evoked changes in oscillatory activity, the spectrogram of single-trial EEG sweeps was obtained using a log-power spectrogram calculated for the [-2, 2]-s interval around each stimulus event (signal, nonsignal). Because the distractor used a 0.5 Hz flicker rate, spectrograms were calculated for the [-1, 1] s interval around each distractor onset. Each spectrogram displayed the frequency from 0.5 to 100
Hz and used a 0.05-s bin width. These log-power spectrograms were averaged within each session for all relevant behavioral events, and the log power of the oscillation bands corresponding to delta (0.5-4 Hz), theta (4-8 Hz), alpha1 (9-12 Hz), alpha2 (12-16Hz), β
(16-30 Hz), low gamma (30-60 Hz) and high gamma (60-100Hz). The mean magnitude of the respective frequency bands was calculated for six time frames: three 250 ms time windows preceding target presentation (-750 to -500 ms, -500 to -250 ms, and -250 to 0 ms) and three windows of 250 ms duration following target presentation.
Four-way ANOVAs with factors SESSION (4 levels), DURATION (25, 50, 500 ms), DISTRACTOR (distractor block of trials, standard block of trials), and TIME (the six time windows), were used to analyze event related changes in spectral power at each frequency band. Due to the paucity of 500 ms signal trials with a miss response, misses could not be adequately analyzed for these signals. Instead, separate four-way ANOVAs with ACCURACY (hit, miss), SESSION, DISTRACTOR, and TIME were employed to compare event-related differences between hit and miss trials at the shorter 50 ms signal duration. In order to test the hypotheses that the initial block of distractor trials produces greater increases in tonic alpha power relative to the second block, in line with task performance in humans (Klimesch et al., 2007), two-way ANOVAs were used to
40 investigate discrete effects of the distractor, with TIME before and after distractor light
onset as the first factor and TIME ON TASK as the second factor.
3.3. Results
3.3.1. Histology
Verification of the placement of the recording electrodes in the cresyl violet-stained brain sections was determined by locating the electrode tract and electrolytic lesion. The coronal section presented in the Figure 3.1 illustrates that the placement of recording electrodes in all rats was within the deep layers of the PPC. All placements were 2.5-3.0 mm lateral from midline and 4.3-4.5 mm posterior to bregma.
3.3.2. Behavioral effects of signal duration and distractor
The detection of brief visual signals during attentional testing was signal duration- dependent (F (2, 8) = 157.89, p<0.05, see Figure 3.2.), with an increased detection rate occurring at longer signals. The presentation of a visual distractor (flashing houselight at
0.5 Hz during trial block 2) also significantly decreased the detection rate (F (2, 8) = 6.32, p<0.05; Quadratic contrast: F (1, 4) = 9.52, p<0.05). Analysis of the VI, an index assessing both signal detection and correct rejection rate, indicated a significant effect of distractor on overall performance F(5, 20) = 3.85, p<0.05). Trend analysis indicated only a significant quadratic contrast F (1, 4) = 9.52, p<0.05, all other contrasts, p>0.05), demonstrating that animals regain performance when the distractor ceases in the final 12 min block. Previous studies from our colleagues have shown that over time, performance in the presence in the
41 distractor improves, and the mean performance from these animals seems to improve (see
Fig. 8), but planned comparisons of the two 6 min trial blocks in the presence of the distractor demonstrate that the effect is insignificant (F (1, 4) = 1.30, p>0.05).
The correct detection of visual signals produced responses of the shortest latency.
Correctly detected signals did not vary in response latency as a function of duration (F (2, 8)
= 1.27, p>0.05) and thus were collapsed for further analysis. A SIGNAL X ACCURACY analysis indicated a significant difference between signal and nonsignal trials (F (1, 4) =
9.82; p<0.05), a significant difference between correct and incorrect performance (F (1, 4) =
18.38; p<0.05), and a significant interaction between the factors SIGNAL and
ACCURACY (F (1, 4) = 28.96; p<0.05). Reaction time for hits (0.69± 0.06 s) was significantly shorter than for misses (1.21 ± 0.10 s; F (1, 4) = 24.07; p<0.05), correct rejections (1.20 ± 0.10 s; F (1, 4) = 17.48; p<0.05), and false alarms (1.14 ± .13 s; F (1, 4) =
17.48; p<0.05).
3.3.3. Event-Related Potentials
Signals of the longest duration had a significantly greater P300 response relative
to 50 ms signals (F (2, 8) = 14.09; p<0.05), the area under the curve for this positive
potential was also significantly greater for longer signals (F (2, 8) = 15.11; p< 0.05, see
Figure 3.3). The negative peak was similar for both durations on hit trials (F (1, 4) = 0.32;
p>0.05), as well as the negative area under the curve for the latter half second of the
delay period (F (1, 4) = 0.13; p>0.05). Due to the lack of misses at longer signals,
comparisons between detected and missed signals could only be made using the 50 ms
signals. Somewhat surprisingly, there were no significant effects or interactions of the
42 visual distractor on the amplitude or area of the P300 response. There was no main effect on the peak of the parietal P300 potential (F (1, 4) = 4.16; p>0.05) for correctly detected 50 ms signals, relative to missed ones. The same held true for the area under the curve of the positive potential (F (1, 4) = 2.30; p>0.05).
The CNV response varied as a function of both detection and attentional load, but not duration. The distractor significantly increased the amplitude of the CNV (F (1, 4) =
8.70; P<0.05). Interestingly, the distractor interacted with detection in that the distractor significantly increased the area under the curve of the hit trials, but not miss trials
(DISTRACTOR X ACCURACY: F (1, 4) = 11.75; P<0.05; Hit trials: F (1, 4) = 7.77;
P<0.05; Miss trials: F (1, 4) = 4.89; P>0.05). This finding suggests that increasing amplitude of the CNV during detection corresponds to increases attentional demand imposed by the distractor.
3.3.4. Increases in the delta and theta bands reflect changes in signal duration.
As can be inferred from the Figure 3.4, there was an enhancement of delta after presentation of the long signal (DURATION X TIME: F (10, 20) = 8.66, p<0.05). During baseline trials, 500 ms signals produced a significant activation (relative to baseline power) in the first 250 ms epoch following signal onset (F (1, 4) = 14.08, p<0.05), an effect not found on 50 ms signals (F(1, 4) = 1.4, p>0.05, 50ms). In the final time bin (500-
750ms post signal), 50 ms signals (F (1, 4) = 21.27, p<0.05) produce significant enhancement of the delta band relative to baseline power. Neither the correct detection of signals nor the distractor modulated power in the delta band. Thus, changes in the delta frequency correlated with signal duration, similar to the P300 response.
43 Signal duration and detection produce effects in theta similar to that of delta.
Duration significantly interacted with TIME (F (5, 20) = 5.99, p<0.05). Significant increases in theta power occurred in the first 250 ms bin following the onset of the 500 ms signal (F (1, 4) = 78.87; p<0.05), and in the following bin for the 50 ms signal (F (1, 4) =
15.90; p<0.05). Analysis of the 50 ms signals indicated no significant effects of
ACCURACY or DISTRACTOR on theta power (p>0.05).
3.3.5. Phasic changes in the alpha band reflect correct detection of signals
Perturbations in the alpha band correlated well with both stimulus saliency and performance accuracy. The longer signals significantly increased power in the alpha band relative to shorter signals (TIME x DURATION: F (1, 4) = 4.15; p<0.05). As seen in the lower frequencies, 500 ms signals induced significant increases in the first time epoch following signal onset (F (1, 4) = 12.75; p<0.05), whereas 50 ms signals induced significant increases in the alpha band 500 ms following signal onset (F (1, 4) = 43.73; p<0.05). Further analysis of the 50 ms signal by correct detection indicated a main effect of ACCURACY (F (1, 4) = 9.54; p <0.05) and ACCURACY x TIME interaction (F (1, 4) =
4.41; p<0.05, see Figure 3.5).
The upper alpha (13-16 Hz) and β (16-30 Hz) bands showed similar patterns of activation. Longer signals produced greater increases in spectral power relative to shorter signals (upper alpha: F (5, 20) = 5.13, p<0.05; β: F (5, 20) = 3.73, p<0.05), and at shorter signals detection enhanced the spectral power in alpha (F (5, 20) = 5.31, p<0.05), but not β
(p>0.05). There were no effects of distractor, duration, or detection on gamma bands
(p>0.05).
44
3.3.6. The distractor evokes changes in both the phasic and tonic alpha band
Spectral analysis of the [-1, 1] time window centered on the onset of the distractor houselight indicated that the higher frequency bands exhibited significant increases in power when the houselight is off (Fig. 7). As indicated in figure 8, the onset of the distractor houselight produced a transient increase in alpha bands followed the onset of the distractor houselight, followed by a desynchronization (alpha: F (5, 20) = 11.94, p<0.05; alpha: F (5, 20) = 6.38, p<0.05). Significant quadratic contrasts (alpha: F (1, 4) =
29.23, p<0.05; alpha2: F (1, 4) = 15.17, p<0.05), and visual inspection of the spectrogram averages indicate the peak of activation occurred in the time bin following the onset of the houselight (-250 to 0). Both lower gamma (30-60Hz) and upper gamma (60-100 Hz) exhibited significantly higher spectral power when the distractor was off (gamma: F (5, 20)
= 4.10, p<0.05; Upper gamma: F (5, 20) = 3.07, p<0.05, see Figure 3.6).
In order to test whether the physiological response to the distractor houselight habituates as a function of time on task, a third factor TIME (early, late) was added in this analysis, in which the spectral content of the first and last 150 occurrences of the distractor onset were compared. The theta and alpha frequencies had significantly higher basal power earlier in the distractor block (alpha1: F (1, 4) = 10.02, p<0.05; alpha2: F (1, 4) =
10.50, p<0.05, see Figure 3.7). This time on task effect did not modulate the phasic response to distractor onset exhibited in the alpha and gamma frequencies mentioned earlier; indicating that the earlier increases in theta and alpha power was independent of the stimulus properties of the distractor lights.
45
3.4. Discussion
This study produced several findings of interest. First is that both accurate performance and amplitude of the parietal evoked P300 potential increased as a function of signal duration. The longest signals also generated earlier and greater enhancement of the delta and theta bands. The long latency CNV potential was significantly greater on hit trials relative to miss trials and was modulated by the distractor, but did not vary as a function of signal duration. Analysis of the spectral content of the local field potential during this epoch also indicated a long latency (250-1 s post stimulus) enhancement of the alpha bands as a function of accurate performance. Power in the higher frequency bands (30-100 Hz) was particularly higher when the distractor houselight was off.
Finally, decreases in vigilance performance in the distractor block corresponded with increases in tonic alpha activity, whereas improvements in the second distractor block corresponded with decreases in tonic alpha activity.
3.4.1. Evoked P300 response and lower frequency oscillations
The P300 response seen in the current study replicates in rats what has been shown in human on multiple occasions (Spencer and Polich, 1999; Demiralp et al., 2001;
Schurmann et al., 2001). Stimuli that are presented with low probability evoke increases in P300 amplitude and area under the curve, which correlate to greater power in the delta and theta frequency bands. This may in part explain why the unpredictable signal light elicited robust, early increases in delta and theta power, whereas a distractor light with a
0.5 Hz flicker rate failed to modulate either frequency. Neocortical low frequency
46 oscillations, specifically in the delta range originate from ascending thalamocortical neurons (McCormick and Pape, 1990). Thus, the lack of an effect of ACCURACY on these frequencies is not surprising as perturbations in this range correlate with stimulus parameters. Unfortunately, ceiling performance at the 500 ms signal precluded analysis of undetected signals in this condition.
3.4.2. Phasic increases in alpha power and correct detection of signals
Since the initial investigations of Hans Berger (1929; Hughes and Crunelli, 2005), the alpha rhythm has been associated with a “resting” or “idling” state (Hughes and
Crunelli, 2005). Elevated baseline alpha band activity in the parietal cortex of humans is associated with anticipation of forthcoming visual events (Klimesch et al., 1998;
Bastiaansen et al., 2002; Thut et al., 2006). The deployment of visual attentional resources then produces decreases in alpha activity in sensory areas (Sauseng et al.,
2005b), which are hypothesized to reflect enhanced cortical excitability. Recently, an elegant study illustrated that the inverse mechanism may also apply in the occipital cortex.
Rihs et al. (2007) used a modified Posner task to demonstrate robust alpha band increases in the hemisphere contralateral to unattended visual field. These findings suggest that alpha synchronization actively suppresses visual attention to unattended space or stimuli in order to sharpen the current spotlight of attention. The event related desynchronizations in the 7-12 Hz range have also been found in rats (Wiest and
Nicolelis, 2003) and typically occur in the first 250 ms following a stimulus. The desynchronization of the alpha bands following the onset of the distractor light can be interpreted simply as an active suppression of a predictable and irrelevant stimulus
47 (though do note that visual stimuli can produce a brief synchronization prior to desynchronization, as seen in Rihs et al., 2007).
The long latency phasic increases in alpha power to detected signals in the current study are in accordance with the human literature. Following the presentation of signals, alpha oscillations become desynchronized (Bastiaansen et al., 2002), but show a long latency (i.e. 900 ms) resynchronization of alpha (Woertz et al., 2004). Further, recordings from the Pz site in humans, overlying the parietal cortex, indicated increases in the alpha band as task demands increased (Spencer and Polich, 1999). Parietal recordings from cats performing a GO/NO-GO task providing a cue and target stimulus have demonstrated robust synchrony in the 8-12 Hz band for expected cue stimuli, relative to novel stimuli (Roelfsema et al., 1997). In this context, the long-latency, detection-related increases in alpha synchronization found in our experiment can be interpreted as an anticipation of the expected tone stimulus.
3.4.3. Tonic levels of alpha power: a marker of attentional effort?
Manipulations like the distractor used in the current task have produced decreases in frontoparietal cholinergic efflux in the initial six min block, during which attentional performance was most impaired. In the second six min trial block, behavioral performance was improved, and a corresponding increase in frontoparietal ACh efflux was observed (Himmelheber et al., 2001). Evidence from these and other experiments have been hypothesized to be markers of “attentional effort”, in which motivated subjects attempt to overcome parametric or pharmacological challenges to performance (Sarter et al., 2006). In the current study, the initial distractor block of trials produces a decrease in
48 the vigilance index and a corresponding increase in tonic alpha activity. As rats encounter more error signals on both signal and nonsignal trials, VI performance approaches standard levels in the second distractor trial block, and tonic alpha rhythms decrease in intensity. In humans, tasks that require the detection of visual stimuli is related to small alpha power in the reference interval (Ergenoglu et al., 2004). Low reference alpha power is hypothesized to indicate that cortical activity is increased
(Klimesch et al., 2007). Thus, the decreased tonic alpha power late in the distractor block may represent increased attentional effort to better discriminate between the distractor light and the signal light.
In vitro studies of rat neocortex may yield a mechanistic link between increases in parietal ACh efflux and modulation of the alpha rhythm. Lukatch and MacIver (1997) demonstrated 3-12 Hz oscillations prevalent in layer 5 of the parietal associational cortex by bath applying carbachol (a cholinergic agonist) and bicuculline (a GABAA antagonist).
Several control experiments indicated that muscarinic and disinhibition of GABAergic neurons were necessary to produce these oscillations. Thus, increased ACh efflux in the presence of the distractor could modulate basal alpha frequencies seen in the current experiment.
3.4.4. Distinctions between LFP and Single Unit Activity
Extensive reviews focusing on the distinction between unit activity and LFP activity have been published (Logothetis, 2003a; Buzsaki, 2006). Single unit recordings measure the extracellular field potential when microelectrodes are placed close to the soma or axons of a neuron, and reports the action potentials produced by the nearest population of
49 neurons. The firing rate of neurons has been a critical measure for comparing the neural activity of sensory processing or behavior for decades (Mountcastle et al., 1974;
Boudreau et al., 2006). Measuring single unit activity provides no information about subthreshold inputs to dendritic arbors or integrative processing in the soma. Also, this technique is biased to larger neurons, which generate a greater flow of membrane current, resulting in a larger spike. Thus, recordings from the cerebral cortex may only represent small populations of pyramidal cells.
By contrast, LFPs represents the cooperative activity of neural populations.
Changes in the LFP have been hypothesized to reflect the averaged synchronized synaptic activity of a neural population within 3 mm of the electrode tip. Rhythmic LFPs of high amplitude and low frequency, classified originally in the EEG literature as delta waves, are generated by the interplay of thalamocortical and neocortical activity and are typically modulated by the ascending neurotransmitter systems, such as ACh, NE, and histamine (Steriade et al., 1993; Eggermann et al., 2001; Lee et al., 2005). The origin of medium range frequencies, such as the alpha and beta frequencies (8-30 Hz), is less understood, though in vitro studies indicate that cholinergic and GABAergic mechanisms are necessary for the generation of theta and alpha in rats (Lukatch and MacIver, 1997).
Finally, changes in higher frequency oscillations (gamma 30-100 Hz) in the neocortex seem to be a function of neural synchrony within smaller populations within the cortex.
Studies that simultaneous measure single unit and LFP activity demonstrated that the firing rate of single neurons can be gated in part by the oscillations in the local field
(Costa et al., 2006). Specifically, as the phase of the LFP approaches a nadir, it can lower the firing threshold within that field, thereby temporally constraining the firing of
50 neurons. Conversely, neurons are embedded within a field and can modulate the field potential, particularly at higher frequencies.
3.4.5. Duration- and Detection-Related Changes in the Evoked Potentials
As predicted, the local field potential varied as a function of signal duration, an effect not found in single unit activity. Further, the amplitude of the P300, and the corresponding changes in the delta frequency, were no different in distractor or standard blocks of trials in this task. One might predict that if the P300/delta response were generated predominantly by thalamocortical connections, as is the case with most low frequency, high amplitude oscillations, it should be modulated by the amount of background contrast during task performance. Due to the nature of the visual distractor, approximately half of all signal trials occur in the dark and half occur when the houselight is illuminated. It is plausible that there are too few trials in the dark to adequately assess this difference in contrast. Regardless, the P300 response to signals was specifically analyzed when the houselight was either on or off, and no effect was found (data not shown). The CNV response, which has been demonstrated in humans to correlate with top down execution of correct responding (Le Dantec et al., 2007), was also found in task performing rats during proper detection of visual signals. These findings complement and validate earlier experiments from our lab (Broussard et al.,
2006). The amplitude of the p300 response was sensitive to signal duration, and the
CNV correlated with correct detection of signals and rejection of nonsignals.
51 3.4.6. Cholinergic input and the production of theta and alpha rhythms
Predominant research into the mechanisms of oscillatory activity in the brain have
focused primarily on ascending thalamocortical input (Destexhe and Sejnowski, 2001)
and hippocampus (Hasselmo, 2006). In the hippocampus, septal cholinergic and
GABAergic afferents are necessary for the production of theta oscillations1, which are
hypothesized to facilitate learning and memory (Hasselmo, 2006). Also, cholinergic
agonists produce both delta and theta activity in the rat hippocampus (Fellous and
Sejnowski, 2000). As mentioned earlier, Lukatch and MacIver (1997) demonstrated the
necessity of muscarinic activation for generation of 3-12 Hz oscillations. Further, it has
been postulated that enhanced cholinergic neurotransmission can shift power from the
theta rhythm to the alpha rhythm, suggesting that cholinergic mechanisms modulate alpha
power as well (Hughes and Crunelli, 2005).
The advantage of studying the rodent model of attentional performance is that an
experimenter can assess the neuromodulatory influences on behavior, and neural
correlates of behavior. The findings in experiment 1 could have been bolstered had the
cholinergic input to the PPC been removed, similar to the design of subsequent
experiments. Due to technical considerations, however, specific hypotheses about
cholinergic function in the LFP and single unit activity could not be simultaneously
tested. Currently, we can only broadly infer that cholinergic input to PPC modulates the
1 Hippocampal theta in rats has been defined as encompassing a 3-12 Hz range, whereas cortical theta in humans is binned in the 4-8 Hz frequency range, and cortical alpha in mammals has been defined as anywhere from 7-13 Hz. Because we are investigating cortical activity, we opted to bin alpha in the 8-12 Hz range. Lukatch and MacIver found 3-12 Hz activity in vitro in rat Occ2, the same region from which we are recording in vivo. Cholinergic and GABAergic activity was required for oscillations in the 3-12 Hz range, and as temperature approached normal body temperature, neocortical oscillations were found in the 10-12 Hz range. 52 ERP based on the results cited above. In tasks using aversive conditioning, the amplitude of the evoked potential was decreased following infusions of SAP into the nucleus basalis magnocellularis (Berntson et al., 2003b). Recent findings indicate that footshock produces significant decreases in delta frequency power of the EEG (generally considered a marker of arousal), an effect that required intact cholinergic input (Knox,
2005). Although there is quite a different circuitry mediating aversive (Berntson et al.,
2003a) and appetitive responding (Robinson and Berridge, 2001; Berridge, 2004), the necessity of cholinergic inputs for measures of cortical arousal and attentional performance has been demonstrated, and is worth further parsing. Future experiments should investigate the cholinergic mechanisms mediating LFP activity in attention task performing animals.
3.4.7. Parietal function in visual attention performance of rats
Previous investigation of electrophysiological correlates of parietal processing in rats have produced evidence of the encoding of head direction cells (Chen et al., 1994a,
1994b), responses to tones in different spatial locations (Nakamura, 1999), and more recently the detection of visual signals (Broussard et al., 2006). Specific investigation of single and multiple unit activity from small recording sites produced increased activation as a function of detection, but not signal duration. That is to say, detected signals at any duration produced significant changes in the firing rate of single neurons. Thus, the current signal duration dependent changes in the evoked potential can be interpreted in two ways. One is that because the local field potentials represent cortical synchronization of thousands of neurons, unpredictable and salient signals can induce changes to a larger
53 population of neurons than less salient signals. Because the tetrodes in these recording studies are placed relatively close together (~100µm) they may be recording from essentially the same population and fail to differentiate between multiple populations as effectively as the field potential. An alternative interpretation is that the field potential, particularly increases in the low frequency components, represents long range inputs from the thalamus, striatum (Costa et al., 2006), or possibly the nucleus basalis (Lin et al.,
2006).
This study is one of a series investigating the electrophysiological responses of parietal cortex in task performing rats. One interesting and unexpected finding was the increase in gamma power during the off cycle of the distractor light. A population of neurons from a different set of rats (in experiment 2) exhibits increases in baseline firing rate when the houselight is off. Given this is an irrelevant stimulus, it is plausible that the absence of light itself is a stimulus that activates local neurons, and the increases in parietal gamma power is a function of this local parietal activation.
Previous evidence regarding the function of parietal cortex in rats stems from lesion studies, which produced a myriad of functional deficits. These deficits included impairment in contralateral paw use, an inability to properly use distal cues for proper navigational maze performance (Kolb and Walkey, 1987; Kolb et al., 1994; McDaniel et al., 1995; Long and Kesner, 1998; McDaniel et al., 1998). Cholinergic deafferentation of parietal areas did not affect attentional performance in the 5 choice serial reaction time task (Muir et al., 1996; Maddux et al., 2007). However, cues providing enhanced incentive value for rats (as measured by duration spent in the food cup) lose this enhancement with cholinergic deafferentation of the PPC (Bucci et al., 1998). Further,
54 partial reinforcement, which generally enhances attention to a conditioned stimulus, fails to do so following cholinergic deafferentation of PPC (Maddux et al., 2007). In this study, separate groups of rats received lesions of the ascending cholinergic projections to either the mPFC or PPC. Deafferentation of mPFC impaired 5CRSTT performance, particularly on signals of shorter duration, whereas PPC deafferentation did not. The authors suggest these findings validate their assertion that cholinergic afferents to mPFC are necessary for signal-driven modulation of detection processes, whereas PPC directs cognitive modulation of detection processes. In the current experiment, signal evoked changes in the lower frequency range (0.5-8Hz) were contingent upon signal duration and not correct detection, whereas changes in the alpha oscillation (8-16 Hz) increased as the animals expected a tone to initiate an operant window. Because alpha oscillations also varied as a function of attentional demand, and due to the necessity of muscarinic receptor activation for the production of 8-12 Hz rhythms, it is possible that these rhythms may be a marker of cholinergic function in the rat.
55
FIGURES
Figure 3.1.) Illustration and example of the final recording sites for moveable tetrodes into the PPC. (A) An example showing the final placement of electrode and cannula. The tract produced by the electrode cannula is readily visible. (B) A schematic illustration of the final placement of the moveable electrodes, which extended 1.0 mm beyond the guide cannula and were located at AP -4.5, within ML ± 2.5-3.0, and DV 1.0. 56
Figure 3.2.) Behavioral effects of signal duration and visual distractor. A.) Signal duration dependent performance during testing. The relative number of hits (mean ± SEM) increases as a function of signal duration. During the distractor block, significant decrements in detection performance were recorded on the 50 ms and 500 ms trials.
57
Figure 3.3.) Separate characteristics of the evoked response correspond to different aspects of task performance. Signals of longer duration evoked a significantly greater P300 response. A long latency CNV (~ 500 ms following signal onset) correlates with correct detection of the signal. Missed signals produced a CNV of significantly smaller amplitude. Due to the relatively few miss trials for 500 ms signals per session, these traces were not analyzed.
58
Figure 3.4.) Signal duration correlates with evoked increases in the delta oscillation. A- B). Log-transformed STFT spectrograms analyzed the [-1, 1] epoch around the signal onset. A.) Fifty ms signals evoke significant increases in the delta and theta frequency in the 250-500 ms time epoch. B.) Signals of longer duration evoked significantly earlier, and more powerful, responses in the delta and theta frequencies. C.) A 2D plot of the log power in the delta band of signals at both durations.
59
Figure 3.4.
60
Figure 3.5.) Correct detection of signals is indexed by significant perturbations in the alpha frequencies. The mean, baseline [-1, 0] power of each frequency bin was subtracted from each time bin; the resulting non-blue bins represent shifts from the mean baseline. A.) When signals are detected, the alpha and upper alpha frequencies significantly increased in power relative to baseline power. The onset of the tone, which opens the operant window, elicits broad spectrum perturbations. B.) Missed signals produce significantly fewer perturbations in the spectrum relative to baseline power.
61
62
Figure 3.6.) Upper gamma frequency is significantly greater when the houselight is off. Log transformed spectrograms were produced for each trial, analyzed in [-1, 1] epochs centered on the onset of the flashing houselight. Data was analyzed in the 60-100 Hz frequency bin, but perturbations were only observed in the 76-100 Hz range. A population of parietal neurons from a different group of animals exhibited increases in firing rate in response to darkness.
63
Figure 3.7.) Basal power in the alpha is diminished with incremental improvement in task performance. A.) The distractor produces deficits in vigilance index, a measure of signal and nonsignal accuracy. Vigilance index ranges from [1,-1], with 1 equating to 100% accuracy, -1 equating to 100% inaccuracy, and 0 equating to chance performance. The distractor produced significant decrements in attentional performance in trial blocks three and four. B.) Changes in the alpha frequency vary as a function of time on task and light onset. Power in the alpha band is represented on the ordinate and time relative to distractor onset is represented on the abscissa. Onset of the distractor light produces a brief synchronization, followed by desynchronization. Overall power in the 8-16 Hz range was greater in the first 6 min trial block relative to the second 6 min trial block.
64
Figure 3.7
65
CHAPTER 4
EXPERIMENT TWO
CHOLINERGIC OPTIMIZATION OF SIGNAL-EVOKED
PARIETAL SINGLE UNIT ACTIVITY
4.1 Introduction
The posterior parietal cortex (PPC) is implicated in a wide variety of cognitive functions, including the shifting of attention between competing sets of stimulus-response contingencies (Bunge et al., 2002; Yantis et al., 2002). This shifting of resource processing is particularly evident in visuospatial attention (Hung et al., 2005; Sereno and
Amador, 2006). Further, the discrimination of relevant stimuli from irrelevant distractors may originate from PPC, as suggested by effects of brain damage in patients (Driver and
Vuilleumier, 2001; Friedman-Hill et al., 2003) Results from our recent experiments indicated attentional performance-associated neuronal activity patterns in the PPC that provide correlational evidence of a role of PPC neurons in the detection of visual signals
(Broussard et al., 2006). For example, we found that, during a sustained attention task, a population of rat PPC cells are phasically and selectively activated by unpredictable signal cues on trials resulting in correct detection, but not when rats failed to detect the
66 signal. Performance on non-signal blank trials also failed to activate this population.
Variation in stimulus duration did not modulate the PPC response, indicating that PPC single neuron activation is not a response to the sensory features of the stimulus but instead reflects the detection of stimuli and the execution of a response plan.
A parallel set of studies have provided ample evidence of the necessity of basal forebrain cortical cholinergic activity for attentional performance in rats (McGaughy and
Sarter, 1998; Burk et al., 2002; McGaughy et al., 2002). Selective deafferentation of cortical cholinergic input, produced by bilateral infusions of 192 IgG saporin (SAP) into the nucleus basalis, produced a reduction in the number of correct responses to signal trials, without affecting the animals’ ability to respond correctly to non-signal trials
(McGaughy et al., 1999). In comparison to control tasks, tasks that tax attentional resources also produced robust increases in ACh efflux (Dalley et al., 2001; Arnold et al.,
2002; McGaughy et al., 2002) and increased the proportion and capacity of choline transporter in the medial prefrontal cortex (mPFC) of rats (Apparsundaram et al., 2005).
Further, behavioral or pharmacological manipulations that produced attentional challenges, produced additional increases in frontal-parietal ACh release (Himmelheber et al., 2000; Kozak et al., 2006). In a separate study distractors produced robust increases in mPFC neurophysiological activity, an effect that was contingent upon intact cholinergic input (Gill et al., 2000). In contrast, little is known about the role of cholinergic input on PPC processing during the detection of signals.
The present experiment was designed to test the hypothesis that the local deafferentation of cholinergic input to the PPC will reduce signal-evoked increases in neural activity. The unilateral infusions of SAP in the vicinity of the recording site
67 produced cholinergic deafferentation but did not impair attentional performance, allowing for an unbiased assessment of the role of cholinergic input on signal processing in the
PPC. A visual distractor tested whether cholinergic deafferentation would reduce signal- evoked activity of PPC neurons under conditions challenging attentional capacity (Sarter et al., 2006). The results support the hypothesis that cholinergic transmission in the PPC optimizes the processing of signal stimuli, particularly under challenging conditions.
4.2 Specific Methods
4.2.1. Electrode and infusion cannula implantation
Two tetrodes were inserted into a 26-ga cannula (15 mm) and extended ~1 mm beyond the distal end of the cannula. The cannula and tetrodes were affixed to a moveable headstage (as described in Broussard et al., 2006). The eight lead wires were soldered into separate channels of an eight channel headstage (Plexon Inc, Dallas, TX).
Two separate Teflon-coated, stainless steel, 250 µm electrodes (A-M Systems, Everitt,
WA) were soldered into to the headstage as well and served as a reference and ground for recordings. A 26 ga, 17 mm, infusion cannula was placed within ~1 mm of the electrode guide cannula for the purpose of infusing saline or SAP near the recording site.
Rats trained to criterion performance were anesthetized with isoflurane gas mixed in oxygen. Body heat was maintained at approximately 37° C with a thermal pad
(Deltaphase IV, Braintree, MA). The tetrodes and a single infusion cannula were implanted unilaterally using the stereotaxic coordinates A/P -4.5 mm, M/L ± 2.5 mm, and
68 D/V -1.0 mm from the dura surface. Four additional burr holes were drilled into the skull into which machine screws were threaded. The reference electrode was placed in the contralateral somatosensory cortex, and the ground electrode was wrapped around a machine screw. The carrier, headstage, and stainless steel electrodes were affixed to the skull with dental cement. Lidocaine and antibiotics were applied to the wound immediately after surgery. Rats were allowed one week to recover from surgery in their home cages with free access to food and water, after which access to water was reduced before resumption of behavioral testing and neurophysiological recording.
4.2.2. Neurophysiological recording sessions
Following five distractor sessions of recording from intact PPC (15 sessions total), rats were again anesthetized using isoflurane, and received a 0.5 µl infusions of SAP
(n=5) in Dulbecco’s saline (Advanced Targeting Systems, San Diego, CA; 0.15 µg/µl.
Rats in the sham group (n=2) received 0.5 µl infusion of Dulbecco’s saline. Behavioral testing resumed 10 days post-infusion. After 1-2 days of re-training, animals were back to criterion levels of behavioral performance, and fifteen more post-infusion testing sessions were recorded.
Single units exhibiting more than 800 total spikes during the recording sessions were used for the electrophysiological analysis (0.30 spikes/s). Nearly all cells detected exceeded this threshold, and the average firing rate was 2.46 ± 0.18 spikes/s (Range:
0.092 spikes/s-22.29 spikes/s). Posterior parietal unit activity was divided into 20 ms bins and changes in activity were analyzed in 2 s epochs. Analysis centered on stimulus presentation (e.g. signal light, tone, and distractor light) and behavioral responses (e.g.
69 correct rejections, hits, misses, false alarms) which were assessed using peri-event time histograms (PETH). Because unit activity often violates the assumption of normality,
Wilcoxon signed-rank tests were used to analyze pre- and post-event epochs. A χ2 analysis was applied to assess the proportion of neuron pairs exhibiting increases or decreases in task-related activity as a function of distractor or cholinergic deafferentation.
Once the population of neurons exhibiting stimulus-driven activity was determined, a calculation of the signal-to-noise ratio (SNR) was used to quantify how much the neuronal activity exceeded the background activity if a visual stimulus was present. SNR was calculated for pre-lesion and post-lesion animals as the stimulus driven response
(Rstim) divided by the sum of the stimulus driven response and background activity (Rspont) for each individual PETH (SNR = Rstim/(Rstim+Rspont). Values range from -1 to 1, with those approaching 1 indicating that the signal evoked response is much higher than the background activity.
4.3 Results
4.3.1. Histology
Verification of the placement of the recording electrodes was conducted on cresyl violet brain sections. The final placement of the electrodes was determined by localizing the electrolytic lesion and reconstructing the dorsoventral path of the electrodes through the PPC. The coronal section represented in Figure 2 illustrates that the final placement of recording electrodes in all rats was within the deep layers of the PPC (III-VI).
Histological evaluation of the removal of cholinergic afferents within the PPC was
70 conducted on AChE-stained brain sections. Images taken at 20X were analyzed using a within-subjects ANOVA (LESION F1, 35 = 119.605, p<0.0001). Each of the 5 SAP infused rats exhibited at least a 75% decrease in the density of AChE-positive fiber staining. The lesion extended rostrally toward primary somatosensory cortex (~-2.8 mm posterior to bregma), caudally toward secondary visual cortex (~-6.3 mm posterior to bregma) and laterally toward auditory cortex (~6.5 mm lateral, 4.5 mm dorsal).
4.3.2. Behavioral performance under standard and distractor task conditions
After several months of training, animals had developed a stable pattern of behavior. The detection of visual signals during standard attention task performance was signal duration-dependent (F2, 8=254.52, p<0.01), with a significant linear trend (F1,
8=447.30, p<0.01). Analysis of latency on hit trials indicates a signal duration effect with rats responding at shorter latencies to 500 ms signals (25 ms: 0.93s ± 0.05, 50 ms: 0.79s ±
0.04, 500 ms: 0.69s ± 0.04; F2, 8=20.49, p<0.01). In Standard sessions, neither signal nor non-signal performance was affected by time on task (Signal: F2, 8=0.55, p>0.05; Non- signal: F2, 8= 1.064, p>0.05).
As illustrated in Figure 3, the presentation of the visual distractor decreased the accurate detection of signals (F2, 8=6.5, p<0.01) and correct rejection of non-signals (F2,
8=18.82, p<0.01). Post-hoc analyses indicate that the distractor had no effect on performance on 25 ms signals (F2, 8 = 1.84, p>0.05), but impaired hit performance on the two longer signals (50 ms: F2, 8=15.78, p<0.01; 500 ms: F2, 8=4.72, p<0.05). Analysis of the rats latency to respond indicates that the distractor significantly delayed responding to the signal (Baseline latency 0.74 ± 0.05s, Distractor latency 0.88 ± 0.06s; F2, 8=3.63, 71 p<0.05), and the signal duration effect on latency was maintained in distractor sessions
(F2, 8=18.58, p<0.05).
As predicted, cholinergic deafferentation did not affect the rats ability to detect signals in either Standard sessions (Signal: F1, 4=0.16, p>0.05; Non-signal Trials: F1,
4=0.32, p>0.05) or in Distractor sessions (Signal: F1, 4=0.002, p>0.05, Non-signal F1,
4=0.47, p>0.05). Further, distractor-induced decreases in the detection of signals and correct rejection of non-signals was unaffected by cholinergic deafferentation. The latencies of all responses under standard conditions were not significantly different following the lesion (Hit: F1, 4=1.08, p>0.05, all other responses p>0.05). Interestingly, cholinergic deafferentation produced longer latencies on hit trials during the distractor sessions (Pre-lesion 0.74s, Post-lesion 0.85s; F1, 22=4.94, p<0.05), an effect not found on other response types (all p>0.05). Saline infusions had no effect on latency or performance (all p>0.05).
4.3.3. Significant activation of PPC neuronal activity during the detection of signals
Multiple unit activity from the PPC was recorded for each 46 min session.
During each session, two to twenty-four units were isolated from multiple unit activity using principle component analysis of aspects of the waveform and template matching using Spike 5 software (Cambridge Electronics Design, Cambridge, England). Prior to cholinergic deafferentation, a total of 350 units were isolated during standard testing sessions and 195 units were isolated during distractor sessions.
Neuronal responses in Standard sessions were characterized by increases in unit activity following the presentation of the signal light on hit trials. Signal-evoked activity
72 was the most prevalent behavioral correlate as 26% of neurons (92/350) exhibited significant activation relative to ITI activity (p<0.01, Wilcoxon signed rank test). The neuron shown in Figure 4 exemplify the increases in the firing rate of a single PPC neuron in response to the signal on hit trials, but not non-signal trials or signal trials resulting in a miss. This population exhibited a signal driven increase in firing rate 100 ms after the onset of the signal; the latency of the peak of the activation occurred at an average of 220 ms post-signal (Fig. 5). This population exhibited a 53% increase in firing rate during the 1 s epoch following the signal (Baseline: 2.74 ± 0.15 spikes/s; Post- signal: 4.24 ± 0.69 spikes/s). These results are similar to those found previously from recordings from PPC neurons of attention task-performing rats (Broussard et al., 2006).
4.3.4. Effects of signal duration on signal -evoked PPC activity
In order to assess the effects of signal duration on PPC neuronal activity, the subset of neurons which exhibited significant signal-evoked activity (92/350) were further analyzed by signal duration. Longer signals activated a greater proportion of neurons, with 25 ms signals eliciting a significant activation from 22/92 cells (23%), 50 ms signals from 43/92 cells (47%), and 500 ms signals from 64/92 cells (69%). In order to further investigate whether PPC activity was a function of signal duration, the signal- to-noise ratio was calculated from the firing rate of each cell at each duration and analysis of the SNR of these neurons indicated no significant effect of duration (p>0.05, Friedman test).
73 4.3.5. Distractor-induced modulation of signal-evoked activation of PPC neurons
The presentation of a visual distractor decreased the proportion of neurons exhibiting signal-evoked activation. Specifically, the relative number of neurons exhibiting signal-evoked activity (36/208, 17%) declined significantly relative to standard sessions (92/350 [26%], χ2 = 9.9, p<0.001). Further, a second population of neurons was recruited by the visual distractor. As illustrated in Figure 6 the visual distractor is a houselight flashing at 0.5 Hz. This population of neurons (40/208) is active during the 1s period that the houselight is off. Signal-evoked (36/208) and distractor driven (40/208) neurons comprised two distinct populations; only three neurons were activated by both the signal and the distractor light.
4.3.6. Cholinergic deafferentation increased distractor-related PPC unit activity
Cholinergic deafferentation significantly increased the proportion of neurons exhibiting distractor-related increases in unit activity. As illustrated in Figure 4.6, a population of 52/167 neurons (31%) were significantly activated when the distractor light was off, a significantly larger proportion of neurons relative to the pre-lesion population
(χ2=20.32, p<0.001). Unlike the pre-lesion data, in which two distinct populations of
PPC neurons correlated with either the distractor light or signal light onset, 38% (7/18) of deafferented PPC neurons exhibiting signal-evoked activation were also activated by the distractor. This was not a function of electrode depth or infusion, as saline infusions did not affect the proportion of distractor related neurons.
74 4.3.7. Cholinergic deafferentation reduced signal-evoked PPC unit activity
Cholinergic deafferentation of the PPC significantly reduced the proportion of neurons exhibiting signal-evoked activation in Standard sessions. During standard task performance, 65/331 (19%) neurons exhibited signal-evoked activation on hit trials, as opposed to 92/350 (26%) neurons from the prelesion population (χ2 = 8.174, p<0.05).
Cholinergic deafferentation did not significantly reduce the overall firing rate of all PPC neurons (pre-lesion 2.34 ± 0.22 spikes/s, post-lesion 2.38 ± 0.21 spike/s, t=0.70 P>0.05).
This effect was not a function of electrode depth or number of sessions, as data from saline-infused animals showed similar proportion of signal-evoked activity from the pre- lesion PPC population (see Tables 1 & 2).
Cholinergic deafferentation also significantly reduced the proportion of neurons exhibiting signal-related activation in the presence of the visual distractor. Only 10.9 % of total neurons isolated (18 of 167 total) exhibited signal-evoked activation. This is the product of two main effects. As presented earlier, the distractor significantly reduced the proportion of signal-evoked neurons, and cholinergic deafferentation further reduced signal-evoked responses of PPC neurons. This proportion of neurons is significantly smaller than the proportion of prelesion neurons exhibiting signal-evoked activation in the presence of the distractor (Prelesion Distractor: 36/208, 17%; Postlesion Distractor
18/167, 10.9%; χ2=9.9, p<0.005). Further, this subset of neurons is significantly smaller than the proportion of neurons exhibiting signal-evoked activation on standard sessions following cholinergic deafferentation (Postlesion Standard: 65/331, 19%; Postlesion
Distractor: 18/167, 10.9%; χ2=126.88, p<0.001). These results suggest that ACh
75 contributes to the optimization of the signal-evoked response under increased attentional demand of the distractor.
4.3.8. Cholinergic modulation of the SNR under attentionally challenging conditions
In order to further investigate the effect of the distractor, baseline firing rate (task- related + ITI unit activity) from signal-evoked PPC neurons was analyzed at each 12 min trial block of Distractor sessions. Prelesion signal-evoked PPC neurons elicit a lower baseline firing rate in the presence of the distractor: F2, 70= 5.04, p<0.05), an effect that is maintained into the third, non-distractor block of trials (F1, 35= 5.35, p<0.05, Linear contrast). This effect was specific to this subset of neurons, as the average baseline firing rate of all 208 neurons collected was not significantly different in Distractor sessions
(t=0.99, p>0.05). This effect was not a function of time on task, as the population of 92 neurons from Standard sessions was not modulated by trial blocks (F2, 184=1.41; p>0.05).
Comparison of the SNR of these neurons in the distractor block and standard blocks of trials indicate no effect of distractor. Together, these findings suggest that in the presence of a distractor, the baseline firing rate of signal-evoked PPC neurons is lowered, thus maximizing the signal to noise ratio of unpredictable and relevant stimuli.
In order to test whether cholinergic input to the PPC was necessary to filter out distractor stimuli, signal-evoked activity was analyzed by block of trials. The distractor reduced the number of cells activated by the signal to a greater degree following cholinergic deafferentation. In the first block of trials 11/18 (61%) neurons were significantly activated by the signal on hit trials. Further, in contrast to the pre-lesion population, the baseline firing rate (task-related + ITI unit activity) of signal driven
76 neurons was not significantly reduced post-lesion (F2, 34=0.66; p>0.05), which may be evidence that the maintenance of an elevated firing rate during the distractor block reduces the SNR of signal-evoked PPC neurons (see Fig. 7).
Following cholinergic deafferentation, baseline firing rate of signal-evoked neurons remained activated throughout the distractor and subsequent block of trials (F2,
34= 0.66, p>0.05). This maintenance of a relatively high baseline firing rate suggests that the SNR might be decreased in the absence of endogenous ACh. Further, the firing rate of distractor-related neurons remains unaltered throughout the task for intact animals (F2,
78=0.44, p>0.05), but is significantly higher during the distractor block of trials post lesion (F2, 102=8.34, p<0.05; Quadratic contrast: F1, 51= 13.97, p<0.05). To test the effects of these elevated basal firing rates on signal-evoked activity, we calculated the median
SNR for each cell during a standard task trial block and in the presence of the visual distractor, and compared the two conditions. As presented earlier, the SNR was not affected by the distractor prior to lesion (Z=-0.833; p>0.05, Wilcoxon), but there was a trend toward significance following cholinergic deafferentation (Z=-1.851, p=0.06), suggesting that ACh neurotransmission may be involved in maintaining a higher SNR in the presence of a distractor. Given that cholinergic deafferentation resulted in an overall loss of significantly activated cells suggests that signal-evoked activity was reduced overall throughout the population, which may be a function of a higher background firing rate.
77 4.3.9. Correct versus incorrect trials
A group of PPC neurons from each condition (Standard Unlesioned, Distractor
Unlesioned, Standard Postlesion, and Distractor Postlesion) exhibited significant activation following the presentation of the visual signal. As seen in Figure 8A, prelesion
PPC activity indicated a stronger signal-evoked activity on hit trials relative to miss trials.
This pattern of activation was consistent in unlesioned PPC in standard sessions (Z=5.638, p<0.01; Wilcoxon) and in the presence of the distractor (Z=3.959, p<0.01).
Cholinergically deafferented PPC maintained this pattern of activation in standard sessions (Z=5.527, p<0.01). However, examination of cholinergically deafferented PPC activity in the presence of the visual distractor indicated that the magnitude of the signal- evoked activity was similar for both hit and miss trials (Z=.283, p>0.01, Fig. 8b). The failure of PPC neurons to produce a reliable discriminatory signal upon which the rat can act suggests a cholinergic enhancement of signal properties in the PPC, particularly as task demands increase.
4.4. Discussion
Our results suggest that ACh transmission in the PPC optimizes the production of attentional correlates. Removal of ACh input to rat PPC reduced signal-evoked increases in neuronal activation, particularly in the presence of a visual distractor. Further, the removal of cholinergic input to the PPC produced proportionally fewer neurons exhibiting signal-evoked increases in firing rate in the absence or presence of a distractor relative to both pre-lesion and unlesioned populations. Cholinergic deafferentation also
78 significantly increased the proportion of distractor-activated neurons. Signal-evoked neurons from intact PPC displayed an overall reduction in baseline firing rate during the distractor block. In contrast, neurons in the cholinergically-deafferented PPC maintained their baseline firing rate in the presence of the distractor, and trend toward reducing their signal-to-noise ratio.
4.4.1. Effects of signal duration and distractor
Signals of longer duration produced more consistent increases in firing rate of signal-evoked PPC neurons. This may be a function of fewer hit trials at shorter signals, as PETH data with less than 15 trials can produces considerable variation (Clayton et al.,
2004). A secondary measure, the SNR of the signal-evoked PPC population, showed no differentiation between signals of different duration. In contrast to the previous study
(Broussard et al., 2006); the distractor produced a significant response in a large population of PPC neurons, equivalent to that of the signal-evoked population of neurons receiving endogenous cholinergic input. One explanation for this discrepancy is the previous study had a smaller population of neurons derived from fewer distractor sessions.
In light of recent findings from our lab indicating little modulation of signal-related activity by the visual distractor, the modulation of prelesion and unlesioned PPC neurons by the visual distractor warrants discussion (Broussard et al., 2006). The earlier study was the first to investigate parietal responses to visual stimuli in task performing rats, which relied on the anatomical definitions of the proposed rat analogue of primate PPC
(Reep et al., 1994). Recordings from these nine animals encompassed a range from 2.5-
4.5 mm lateral from the midline and in which four probes were more than 3.5 mm lateral
79 from the midline. Detection-related activity from these neurons had a rather long latency
(~500 ms) and although further subdivision of the neuronal populations by animal was precluded by a relatively small population (~120 neurons); we observed that recordings from the more medial sites produced shorter latency visual responses. Thus, in this study we recorded exclusively from a more medial range (2.5-3 mm). As a result, signal- related activation from this population was found to have a shorter latency of peak activation (~220 ms) with greater than baseline activity beginning around ~120 ms post stimulus. Given that this population was more responsive to visual stimuli, it follows that imposing visual distractors would recruit activity from and modulate the signal-driven activity of this population.
4.4.2. Cholinergic modulation of PPC function in rats
A growing body of evidence supports the role of rat PPC in visual attention, and recent studies point to a role of ACh in the evaluation of visual signals. Mechanical lesions of the PPC impaired the ability of rats to respond to distal visual cues or
“allocentric space” (McDaniel et al., 1998) and produce auditory and visual neglect (King and Corwin, 1993). Bilateral cholinergic deafferentation of the PPC produced impairments in incremental conditioning (Bucci et al., 1998), and attention to learning in a novel context (Maddux et al., 2007), indicating that cholinergic input to PPC is necessary for the proper valuation of visual signals. The restricted, unilateral lesions in the current study, as predicted, did not impair signal or non-signal performance in either standard or distractor versions of this task. The longer response latency to signal trials during distractor sessions indicates a mild behavioral deficit.
80 The successive presentation of signal and non-signal trials in the sustained attention task provides additional uncertainty and requires a rat to hold two stimulus- response associations in working memory (Parasuraman et al., 1987). Proper performance on non-signal trials, which is maintained in the absence of ascending cholinergic modulation (McGaughy et al., 1996; Burk et al., 2002), requires a rat to respond to a tone stimulus. The presentation of the visual signal required an animal to switch stimulus-response associations to the tone so that the rat must respond to an opposing lever upon tone presentation. The attentional challenge imposed by a visual distractor reduced the proportion of signal-evoked neurons. Cholinergic deafferentation significantly reduced the proportion of neurons exhibiting signal-evoked activation in both the standard and distractor conditions. Further, the population of neurons exhibiting signal-evoked activation had a higher spontaneous firing rate, a lower SNR (i.e. the signal-evoked activity relative to the spontaneous firing rate was lower in the presence of the distractor following cholinergic deafferentation), and a greater signal-evoked response in miss trials. Taken together, these findings support the hypothesis that cholinergic neurotransmission is necessary for the redirection in stimulus-response associations required in signal detection.
4.4.3. Cholinergic suppression of distractor-related activity
Although cholinergic modulation of distractor stimuli has been found in somatosensory cortex (Alenda and Nunez, 2007), this is the first report from PPC recordings. In the current study, a separate population of neurons was found to increase their firing rate when the visual distractor light was off. It is important to note that during
81 training the operant chamber was constantly illuminated with a houselight. The off state of the distractor represents a novel condition in which the rat must perform. Following cholinergic deafferentation, the proportion of neurons activated by the distractor increased dramatically. ACh transmission may be necessary to mediate the suppression of a constant, irrelevant stimulus, thereby facilitating the processing of the visual signal.
Under normal task conditions, cholinergic deafferentation produces a significantly smaller but sizable population of signal-evoked neurons. The constant distractor dominates the response of cholinergically-deafferented PPC neurons and produces an elevated baseline firing rate of signal-related neurons. The conflicting effects of cholinergic deafferentation on distractor and signal stimuli may be explained by the nature of these two stimuli. PPC neurons have been shown to encode both sensory and cognitive aspects of visual stimuli (Zhang and Barash, 2000, 2004). The distractor flashes approximately 365 times during the 12 min block of trials; in contrast, the signal is randomly presented approximately 27 times during the same period. The reduction of signal-evoked activation of cholinergically-deafferented PPC neurons suggests that top- down processes act via cholinergic mechanisms to optimize signal processing. In the absence of this input, PPC neurons maintain or accentuate a sensory response, as reflected by the increase in evoked response in miss trials and to the distractor.
The present data provides evidence for a cholinergic modulation of signal-evoked activity in the PPC of attention task-performing animals. Cholinergic input is hypothesized to modulate neocortical processing by suppressing the glutamatergic synaptic input of local cortical circuits and enhancing afferent, putatively sensory, input
(Hasselmo and McGaughy, 2004; Sarter et al., 2005). Recordings from sensory cortex of
82 anesthetized rats indicated that release of ACh confined the spread of intracortical activation, while at the same facilitating feed-forward or afferent synapses (Hasselmo et al., 1992; Kimura, 2000; Linster and Hasselmo, 2001). In sensory cortex of anesthetized animals, application of ACh reduced the variability of the neuronal response of neurons to visual stimuli (Zinke et al., 2006) and modulated visual receptive fields (Roberts et al.,
2005). Phasic application of ACh hyperpolarized neurons, whereas tonic elevation of
ACh levels resulted in depolarization of the same neurons (Gulledge and Stuart, 2005).
These findings have defined the mechanisms by which ACh neurotransmission modulates cortical plasticity. The elevated baseline firing rate and weakened response of SAP- infused PPC to the signal indicates an impairment of cholinergically deafferented PPC neurons to switch from local PPC neuronal activity to the processing of infrequent, salient stimuli.
Computational models of in vitro data suggest that the intrinsic connections within local cortical assemblies provide the capacity to store distributed association necessary for recall. By suppressing the intrinsic inputs within cortical assemblies, cholinergic input increases the relative value of incoming afferent input (i.e. enhances the signal to noise ratio, (Hasselmo and Wyble, 1997; Linster and Hasselmo, 2001). In the absence of cholinergic input, these local cortical circuits have an elevated spontaneous firing rate, resulting in a relatively muted response to sensory stimuli. The distractor suppressed the spontaneous firing rate of signal-evoked PPC neurons. This effect was cholinergically dependent, indicating that attentional challenges specifically induce cholinergic modulation of parietal processing (Sarter et al., 2006). This is consistent with the finding that similar attentional challenges increase ACh efflux in frontoparietal areas.
83 The suppression of the spontaneous firing rate via increases in cholinergic efflux may represent a mechanism by which basal forebrain cholinergic input can optimize the detection of incoming stimuli (Sarter et al., 2005).
In summary, separate populations of PPC neurons increased firing rate in response to visual signals and distractors. Signal-evoked PPC neurons inhibited their firing rate in the presence of a visual distractor and maintain a high signal-to-noise ratio throughout the task. Following cholinergic deafferentation, significantly fewer neurons exhibited signal-evoked activity, the firing rate of signal-evoked neurons remained elevated in the presence of the visual distractor, and the signal-to-noise ratio of these neurons was moderately lower than that of the pre-lesion neurons during the distractor block of trials. Further, signal-evoked neurons elicited a higher firing rate on miss trials following cholinergic deafferentation, suggesting that these neurons are not selectively firing during the redirection of attention to the visual signal, but rather in anticipation of a general response. These findings suggest a role of ACh optimizing the processing of salient task-relevant stimuli in the PPC.
84 Prelesion Unlesioned Postlesion Hit excitatory 26% (92) 27% (37) 18% (62) * Miss excitatory 8% (28) 4% (5) 8% (27) CR excitatory 8% (28) 7% (9) 8% (28) FA excitatory 5% (19) 8% (11) 5% (22)
Table 1. Behavioral correlates of PPC unit activity during sustained visual attention Data are expressed as the percentage of units (absolute number noted in parentheses) displaying significant behavioral correlates from the total number of units recorded (prelesion 350; unlesioned 136; postlesion 331).
*Significantly different from both prelesion and unlesioned ( χ2; P < 0.05)
Prelesion Unlesioned Postlesion Hit excitatory 17% (36)† 17% (13) 10% (18) ** Miss excitatory 2% (5) 3% (2) 7% (13) CR excitatory 6% (13) 2% (1) 6% (10) FA excitatory 8% (16) 9% (5) 7% (13) HL activated 19% (40) 5% (4) 31% (52)*
Table 2. Behavioral correlates of PPC unit activity during distractor sessions Data are expressed as the percentage of units (absolute number noted in parentheses) displaying significant behavioral correlates from the total number of units recorded (prelesion 208; unlesioned 73; postlesion 167).
*Significantly different from both prelesion and unlesioned ( χ2; P < 0.05) **Significantly different from prelesion distractor performance, and postlesion standard performance. †Significantly different from standard sessions
85
FIGURES
Figure 4.1.) As explained in the Materials and Methods, animals were implanted with a drivable mount including recording electrodes and infusor cannula and were allowed to convalesce and then habituate to task performance. Single unit activity was then recorded from task-performing animals in vivo for two standard sessions (houselight illuminated throughout 36 min session; "S") and a distractor session (houselight flashes at 0.5 Hz for the middle 12 min block of trials; "D"). Following each distractor session, the tetrodes were lowered to record from a new neuronal population. After fifteen sessions from intact PPC neurons were collected, a 0.5 6l bolus of the selective cholinotoxin SAP (n=5) or Dulbecco's saline (n=2) was infused near the recording site of animals. Electrophysiological recordings resumed 10 d later, allowing the destruction of ascending cholinergic input to PPC. Fifteen post-infusion sessions (10 standard, 5 distractor) were then recorded from each animal, with electrodes lowered following each distractor session.
86
Figure 4.2.) Electrode path and SAP induced AChE-positive fiber loss within the PPC. Top left: Schematic of a coronal section through the level of the PPC (4.3-4.5mm posterior to bregma) illustrating the final recording sites of the recording electrodes in the left or right hemispheres. Symbols represent ●: Rats were infused with SAP in right hemisphere; ○: Rats were infused SAP in left hemisphere; ■: Rat was infused with saline in the right hemisphere, □: Rat was infused with saline in the left hemisphere. Top right: Photomicrograph illustrating the electrolytic lesion at the final recording site of Nissl stained PPC, magnified at 4x. Electrolytic lesion from each animal was within PPC layers III–V of the left hemisphere. Middle left: Restricted loss of AChE-positive fibers to PPC near the final recording site in left parietal cortex. Note the density of AChE staining in the contralateral cortex. Middle right: AChE-positive staining from an animal receiving saline infusions. Bottom left: Higher-magnification photomicrograph (20X) of the loss of AChE-positive fibers throughout the deep layers of the PPC. Bottom right: High magnification photomicrograph of the contralateral AChE fibers from the PPC of the same animal as on the left side.
87
Figure 4.2.
88
Figure 4.3.) Behavioral performance as a function of signal duration, trial type and trial block. Plots show the average percentage of correct trials during baseline block of trials (trial blocks 1 & 3, black plots) and in the presence of a constant distractor (trial block 2, gray plots). Responses to signal trials were assessed at three durations. The bar graphs show the average percentage of correct rejections during baseline block of trials and a distractor block of trials, in black and gray bars, respectively. Data are taken from the same animals prior to (left), and following (right), infusion of SAP (left) into the PPC.
89
Figure 4.4.) Example of a raster plot and histogram of a single PPC neuron during the four response types: Clockwise from top left: correct detection of a signal (Hit), Correct Rejection of nonsignal trials (CR), incorrect rejection of a signal (Miss), and incorrect detection of nonsignal trials (FA). The yellow circles represent the onset of the visual signal; the black line represents the onset of the tone, opening the operant window on all trials. Behavioral responses of the rats on each trial are represented by symbols following the tone. Rasters are organized by trial number with the first trial on the top. The number of trials in each session is indicated in parentheses at the top of each graph.
90
Figure 4.5.) Stimulus locked population PETHs showing PPC responses to signals for trials yielding correct and incorrect behavioral responses. Nonsignal trials are locked to an event marker starting the trial. The response to the visual signal initially peaks 220 ms following signal presentation and activation remains sustained until the hit response. The signal-evoked response was greater on hit trials relative to miss trials. Curves represent the RT distributions for correct responses on signal and nonsignal trials. False alarm and miss trials have a similar distribution and latency as correct rejections and have been omitted for clarity.
91
Figure 4.6.) Effects of SAP lesion on the distribution of signal and distractor correlates. The two prevalent correlates were the distractor and the signal light. Top: Population of neurons significantly activated during the “off” phase of the distractor houselight. Bottom: Distribution of significant correlates for pre-lesion and post-lesion sessions. Bottom Left: Following a 12 min block of trials in which the houselight remains illuminated, the off phase activates 40/162 neurons prior to infusion of SAP. Pre- deafferentation, this population is separate from the population correlated with the detection of signals (only 3 neurons are activated by both the signal and the off phase of the distractor. The peak activation of this population was 20 ms following the offset of the houselight. Bottom right: Following deafferentation, a greater proportion of PPC neurons (χ2 = 20.32; P<0.0001) are activated by the off phase of the distractor (52/115), and fewer neurons are activated by the signal light. For both graphs, neurons that have mixed correlates are activated by both the signal and the distractor.
92
Figure 4.6.
93
Figure 4.7.) The effects of SAP lesions and distractor on the baseline firing rate of signal-evoked distractor-related neurons. A. The baseline firing rate of signal-evoked and distractor-related prelesion PPC neurons. The distractor significantly reduced the baseline firing rate of signal-evoked neurons, an effect that was maintained in the third, non-distractor block. Distractor-related neurons showed no changes in baseline firing rate throughout the task. B. Following SAP lesions, the baseline firing rate of signal- evoked neurons remains elevated throughout the distractor and subsequent trial block. Also, the baseline firing rate distractor-related neurons are significantly increased in the presence of the distractor. C & D. The effects of distractor and SAP lesions on the SNR, which measures the signal-evoked response in relation to the ITI firing rate. C. The population of prelesion signal-evoked neurons maintains a high SNR during standard task performance and during the distractor block of trials. D. The population of postlesion signal-evoked neurons has a moderately reduced SNR in the distractor relative to the standard blocks of trials. Considering that SAP lesions already significantly reduces the proportion of signal-evoked neurons, the elevated baseline firing rate of stimulus driven neurons may be one mechanism by which cholinergic deafferentation impairs adequate processing of relevant signals.
94
Figure 4.7.
95
Figure 4.8.) The effects of SAP lesions on PPC signal-related activity on hit and miss trials during distractor sessions. A, left: A signal-locked PETH for hit and miss trials (20 ms bins, Gaussian filtered over three bins) of 36 signal-evoked neurons from prelesion recordings. The signal elicits a peak activation of these neurons on hit trials at an average of 220 ms following the signal, and the firing rate remains elevated through the 1 s delay and response. Right: The average firing rate of each neuron during the 1 s epoch following the signal, with hit trials plotted against miss trials; 28/36 neurons (green dots) have a significantly higher firing rate on hit trials. Only 2/36 neurons (red dots) have a higher firing rate on miss trials. B, left: A signal-locked PETH, same format as A, from 18 signal driven neurons from postlesion recordings. PPC neurons are activated on both hit and miss trials. Right: The average firing rate of each neuron during the 1 s epoch following the signal; 6/18 neurons have a higher firing rate on hit trials relative to miss trials and 3/18 neurons have a higher firing rate on miss trials.
96
CHAPTER 5
EXPERIMENT THREE
EFFECTS OF CHOLINERGIC DEAFFERENTATION OF MEDIAL
PREFRONTAL CORTEX ON TASK PERFORMANCE AND PARIETAL
NEUROPHYSIOLOGY
5.1. Introduction
The prefrontal cortex is critical for executive functioning and is crucial for the maintenance of several memory representations necessary for mediating action over time.
Executive functioning is defined as the processes that are necessary for organizing cross- temporal contingencies of complex behavior. A cross-temporal contingency can be operationally defined as either retrospective (i.e. working memory) or prospective
(preparatory set). Lesions of the prefrontal cortex in humans and primates specifically impair performance on tasks requiring the maintenance of cross temporal contingency
(Fuster, 1997, p. 222). Briefly, the prefrontal cortex serves to represent individual acts in a complex sequence, and executes those acts in the proper order. Tasks which require some degree of conditioned inhibition, such as the GO/NO-GO tasks or delay tasks used often in primate studies, require the prefrontal cortex (DiMattia et al., 1990). Frontal areas are also essential for the formulation and execution of structures of behavior,
97 particularly when subjects must make a decision between two or more behaviors, and the cues that signal the proper behavior are unpredictable and variable. In primates, the dorsolateral prefrontal cortex is necessary for biasing and selection of stimuli for attentional processing (Posner and Dehaene, 1994; Gehring and Knight, 2002), whereas medial areas, such as the orbitofrontal cortex, are necessary for the exclusion of irrelevant stimuli. In this experiment, we will test the broad hypothesis that the medial cortex in rat is necessary for the exclusion of irrelevant visual distractor in a sustained attention task.
The prefrontal cortex displays considerable anatomical variation across species, especially with regards to the presence or absence of a granular zone and reciprocal connections from the mediodorsal nucleus of the thalamus (Preuss, 1995). Recent advancement in the field of retrograde and anterograde tracers have demonstrated that prefrontal cortex in primates produce reciprocal connections with thalamic nuclei outside of the mediodorsal thalamus, and that the MD projects to other parts of the brain (Ongur and Price, 2000). The ambiguous nature of the reciprocity between MD and apparent prefrontal areas requires more specific criteria for inclusion. Uylings and van Eden
(1990) revised this criterion, suggesting that reciprocal connections from apparent PFC neurons to MD must be of greater number and have greater synaptic density relative to other thalamic nuclei.
A second measure by which rat and primate prefrontal cortical areas are homologous is the reciprocal connections between cortex and ascending monoaminergic and cholinergic nuclei. The nucleus basalis magnocellularis sends cholinergic and non- cholinergic afferents throughout the cortical mantle relatively diffusely. In primates and rats, only the medial PFC projects back onto the NB. The locus coerulus, the ascending
98 system providing noradrenergic tone throughout the neocortex, also receives direct projections solely from the prefrontal cortex of rats and primates. Further, there is specific anatomical distribution of distinct portions of the mPFC to distinct parts of the striatum; the infralimbic area of rat mPFC project to the shell of the nucleus accumbens, whereas prelimbic area projects to the core (Ongur and Price, 2000). These and other studies have shown parallels between rats and primate frontal cortical areas in the cortical gating of neuromodulatory tone (Uylings et al., 2003). By these criteria, the prelimbic, infralimbic, and orbitofrontal cortex of rats constitutes a homologue of primate PFC.
Based on anatomical and functional criteria, there are several distinct subregions within rat prefrontal cortex, namely the medial, lateral, and ventral divisions (Dalley et al., 2004a). The medial frontal division is comprised of both a dorsal and ventral subdivision. The dorsal-medial subdivision includes precentral and anterior cingulated cortex, whereas the ventral subdivision includes the prelimbic, infralimbic, and medial orbital cortices. Due to the results of the functional studies outlined below, we will confine our discussion to the medial prefrontal contribution to behavior in rats.
As in humans and primates, the prefrontal cortex of the rat contributes to working memory processes. Specifically, tasks involving a delayed response contingency, such as delayed spatial alternation (Kesner and Holbrook, 1987), and delayed non-match to sample tasks (Rich and Shapiro, 2007) require an intact mPFC. An additional impairment caused by mPFC lesions is the inability to shift toward new strategies or rules.
In rats trained to discriminate between bowls containing food on the basis of digging medium or odor, damage to the medial prefrontal cortex produces a selective deficit in extradimensional set shifting, that is, shifting from one modality-reward association to
99 another, whereas intradimensional set shifting remained unimpaired (Birrell and Brown,
2000). Similarly, mPFC lesioned rats are impaired when working memory tasks require a shift in the association of rules to performance. For example, shifting from spatial to visually cued rules in the Morris water maze requires an intact mPFC.
The integrity of the mPFC is also necessary for correct performance in visual attention tasks (Dalley et al., 2004a). Disruption of glutamatergic transmission in the prefrontal cortex of rats impairs response inhibition in an attention task (Murphy et al.,
2005). Infusions of ibotenic acid, a toxin which preferentially destroys non-cholinergic neurons, into the mPFC produces randomized task performance in the sustained attention task (Miner et al., 1997). Additionally, lesions of the prelimbic area increase perseverative responding under conditions minimally taxing attention. Given the necessity of cholinergic neurotransmission for attentional performance, the specific role of cholinergic input to rat mPFC is also worth mentioning. Blockade of muscarinic receptors via localized infusions of scopolamine into the mPFC produce deficits in allocentric spatial navigation (Nieto-Escamez et al., 2002). Cholinergic deafferentation of mPFC via infusion of cholinergic immunotoxin SAP produces increases in perseverative and impulsive responding in visual attention tasks (Dalley et al., 2004b).
As mentioned in chapter 1, rats that are well-trained in a visual attention task produce significant increases in cortical ACh efflux, an effect not found on control tasks (Arnold et al., 2002).
Cholinergic input to the medial prefrontal cortex is hypothesized to mediate attentional processing requiring attentional effort, in which a motivated subject attempts to detect signals and reject nonsignals in the face of detrimental conditions, such as
100 increased time on task, pharmacological manipulations, or distractors (Sarter et al.,
2006). The visual distractor in this task produces initial impairments in VI performance or a side bias, followed by a recovery to nearly stable levels of performance. This improvement in performance in the distractor block over time requires the rat to encode the increase in error trials and change behavioral strategies to optimize task performance. Here we test the specific hypothesis that cholinergic input to the mPFC is necessary for mediating the shift in behavioral strategies necessary for improvements in performance.
In addition to behavioral effects, we will be recording the in vivo single unit activity of PPC neurons during distractor task performance. The effects of ipsilateral, contralateral, and bilateral infusions of SAP into the mPFC on PPC unit activity in attention-task performing animals will be assessed. Specifically, the effects of ipsilateral infusions of SAP into the mPFC of PPC unit activity of attentional task- performing rats will be compared to the effects of contralateral infusions of SAP on
PPC unit activity. In vivo electrophysiological recordings from mPFC of attention-task performing rats showed increase activation in the presence of a visual distractor, an effect that was cholinergically mediated (Gill et al., 2000). Because cholinergic activation of the mPFC was sufficient to stimulate PPC cholinergic efflux (Nelson et al.,
2005), we propose that cholinergically mediated increases in mPFC firing may act to suppress or filter out distractor stimuli in this task, an effect that will be reflected in
PPC unit activity. Thus, aim of this experiment was to test the hypothesis that loss of ipsilateral, but not contralateral, cholinergic input to the medial prefrontal cortex
(mPFC) will reduce detection-related PPC activity in the presence of a visual distractor.
101 In contrast to cholinergic input to the PPC, which reduced signal related activity in normal and distractor blocks of task performance, we predict that signal-related activity in the PPC will only be affected in the distractor block.
5.2. Specific Methods
5.2.1. Subjects
Male Hooded Long-Evans weighing between 200-250 g, were obtained from
Harlan Industries (Indianapolis, IN) and individually housed on a 12:12 light-dark cycle with lights on at 6:00 a.m. Following two weeks of handling and ad lib access to food and water, a restricted water schedule was imposed, limiting access to 30 min a day during operant training. Experiments were approved by the Institutional Laboratory
Animal Care and Use Committee of The Ohio State University in facilities approved by the Association for the Assessment and Accreditation of Laboratory Animal Care.
Rats were trained on a variant of the sustained attention task introduced in the
General Methods. The general task schema is the same, in which visual signals of varying duration are interspersed with nonsignal blank trials. Because cholinergic deafferentation of the mPFC produced premature and perseverative responding in rats performing a visual attention task (Dalley et al., 2004b), the task was modified to prevent reinforcement of premature responses (See Figure 5.1). First, in order to minimize perseverative responding, each time a rat pressed during the ITI, the ITI was reset, delaying the trial, and subsequent reward. Because the ITI was variable, the animal did not have added control over the task by delayed stimulus onset. Second, a premature
102 response to signals following the onset of the signal, but before the onset of the tone, canceled the trial, which was recorded as a premature response. These procedures were implemented after rats reached criterion performance on the second stage of training, in which the basic task rules are learned. Rats that had attained a 75% accuracy rate for three days in performance in the operant box equipped for neurophysiological recording were subjected to surgical procedures.
5.2.2. Surgery
Rats were administered isoflurane and placed in a stereotaxic apparatus. In addition to implanting a recording a drivable apparatus including two tetrodes, ground e and a Plexon headpiece (recording electrodes implanted at AP: -4.5mm, ML: ± 2.5mm,
DV 1mm), two infusion cannula were implanted in the prelimbic area of the frontal cortex of both hemispheres (AP: +3.2mm, ML: ± 1.6mm, DV -3.8mm). Following the completion of surgery, rats received the antibiotic amoxicillin (30 mg/kg) and topical administration of local anesthetic over all wounds.
The following day, rats were briefly anesthetized and 0.5µl of the selective cholinotoxin SAP (ATS, San Diego CA) was infused at 0.1 µl/min into one hemisphere.
One group of rats (n=8) received the initial prL infusion ipsilateral to the recording probe, and the second group (n=6) received the initial prL infusion contralateral to the recorded hemisphere. Ten days were allowed to pass before recording ensued in order for the animal to convalesce and to allow for the degeneration of cholinergic fibers. Rats were then re-trained on the attention task while connected to a unity gain headset plugged into a commutator. Following two days of criterion performance, physiological and
103 behavioral data was collected. Four of the animals failed to retrain to criterion performance following surgery and initial SAP infusion. Of the 10 remaining animals, three had developed illness halfway through data collection, leaving 7 animals from which a complete data set was obtained. One group (n=5) had received initial SAP infusions ipsilateral to the recording probe, and the second (n=2) group received initial
SAP infusions contralateral to the recording probe.
Behavioral and neurophysiological data was collected for nine days, with two days of standard session recordings followed by a single distractor. On the day following each distractor session, the movable drive was lowered by 50 µm.
5.2.3. Histology
At the conclusion of the study, rats were deeply anesthetized and transcardially perfused in a manner identical to that described in chapter 3. AChE positive fibers were quantified as in chapter 3, with additional frontal slices also analyzed.
5.2.4. Behavioral Measures
Behavioral measures similar to those recorded in previous chapters were also used here. Additionally, premature responses were recorded and analyzed. Repeated measures ANOVAs were conducted on the behavioral data using Signal Duration (25ms,
50ms, 500ms) and Lesion (unilateral, bilateral) as within subjects factors for analyzing the dependent measures of signal response accuracy, non-signal response accuracy, errors of omission, and reaction time on signal and non-signal trials, as well as response lever side bias. An increase in the number of omissions in this variant of the task precluded
104 simultaneous analysis of duration and distractor. Thus, a separate repeated measures
ANOVA analyzed the same dependent measures using Trial Block (Block 1, Distractor,
Block 3), and Lesion as within subjects factors.
5.2.5. Neurophysiological Measures
We recorded and analyzed single unit activity from parietal neurons using techniques described in the General Methods. Briefly, unit activity was binned (20 ms) and analyzed using PETH analysis in 4 s epochs [-2, 2] centered on task stimuli and behavior. Neurons which significantly modulated their firing rate (Wilcoxon, α = 0.01) were subjected to further analysis. Specifically, a χ2 analysis was applied to assess the proportion of neuron pairs exhibiting increases or decreases in task-related activity as a function of distractor, signal duration, or unilateral or bilateral cholinergic deafferentation of mPFC. The signal-to-noise ratio of each neuron was also calculated and subjected to further analysis based on these factors.
5.3. Results
5.3.1. Effects of response cost and time outs on premature responding.
As predicted, imposing a response cost on impulsive or premature responding reduced the occurrence of such responding. In this variant of the task, rats responded prematurely 3.42± 1.27 times per session on the cued lever, and 0.24 times on the non- cued lever. By comparison, rats performing on a variant of this task that does not impose a response cost on premature responding (from experiment 2), responded
105 prematurely 6.68 ± 4.45 times/session. In both variants of the task there is great degree of variability between subjects. Three of the seven subjects in the current experiment did not prematurely respond during any of the distractor sessions. Rats did prematurely respond significantly more on cued levers relative to non-cued (F (1, 6) = 6.58, p<0.05).
Comparison of unilaterally lesioned to bilaterally lesioned animals produced no effect on premature responses, (p>0.05).
The response cost also increased the number of omissions in the task. In experiment 1, rats omitted on 13.67% ± 2.21 of signal trials and 19.35% ± 2.86 of nonsignal trials. By contrast, animals in the current experiment omitted 29.6% of signal trials and 35% of nonsignal trials during distractor sessions. Standard procedures for increasing motivation for water reward, (e.g. reducing access to water, varying time of access to water), had marginal impact on these omission rates.
The third consequence of this modification of the task was a higher attrition rate.
Of the 18 animals trained on this task, 4 had failed to acquire basic task rules, and another
4 failed to reacquire adequate task performance following surgery. The post-surgical attrition rate may be due to other factors, however, such as a larger headset or mPFC deafferentation.
5.3.2. Histology
The final placement of recording electrodes in all rats was within the deep layers of the PPC (III-VI), as in experiment 2. Histological evaluation of the removal of cholinergic afferents within the PPC was conducted on AChE-stained brain sections (See
Figure 5.2.). Images of mPFC and PPC slices taken at 20X were compared to a control
106 animal which received no infusions. The tissue from the control animal was stained using the same protocol as SAP infused animals. Comparisons were made using independent t-tests with the number of black pixels as the dependent measure Each of the
7 SAP infused rats exhibited a 70% decrease in the density of black pixels of mPFC stained slides in both hemispheres (all p<0.05). No animals showed a decrease in AChE staining in PPC cortex.
5.3.3. Initial lesions do not impair behavioral performance
Cholinergic input to the right hemisphere mPFC is specifically implicated in attentional performance (Apparsundaram et al., 2005). In order to test the effects of right hemispheric cholinergic deafferentation relative to left hemispheric lesion, a between subjects factor was added, using behavioral measures from the initial sessions following unilateral lesion. Right (n=3) and left (n=4) lesioned animals did not differ in signal detection accuracy, VI, side bias, or omissions (all p>0.05). However, the lack of a hemisphere effect may be a function of the small sample used in this study.
5.3.4. Bilateral lesions increase the occurrence of omissions on nonsignal trials
Bilaterally lesioned rats exhibited accuracy levels similar their earlier unilaterally lesioned performance. Subjects retained signal duration dependent performance (F (2, 10)
= 55.33, p<0.05). The visual distractor produced significant decrements in attentional performance (as measured by the vigilance index; F (2, 12) = 18.74, p<0.05). Bilateral lesions had no effect on signal accuracy (F (1, 6) = 1.66, p>0.05), measures of the VI (F (1,
6) = 0.58, p>0.05), nor did bilateral lesions interact with duration or distractor blocks to 107 affect performance. Measures of response latency showed no effect of lesion on signal or nonsignal response latency (all p>0.05). Analysis of the relative number of omissions indicated that bilateral lesions significantly increased the number of omissions on nonsignal trials in the distractor block (LESION x SIGNAL x BLOCK; F (5, 30) = 3.62, p<0.05).
5.3.5. Bilateral mPFC cholinergic loss produces a persistent distractor-induced side bias
As indicated in Figure 5.3., the visual distractor produces an initial side bias to the hit lever. This represents an increase in the detection of signals and decrease in the correct rejection of nonsignals. In intact animals (Himmelheber et al., 2001), the repeated occurrence of errors on nonsignal trials produces a shift toward the “neutral” side bias scores (0.3-0.4) in the second 6 min distractor trial block. Unilaterally lesioned animals produce this same pattern of performance. Following lesion of the second hemisphere, rats gradually increase their side bias throughout the distractor block, and remained biased to the hit lever in the absence of the distractor (F (5, 25) = 2.86; p<0.05).
Post hoc within subjects ANOVAs indicated a bias to the hit lever in trial blocks 5 (F (1, 6)
= 5.49; p<0.05) and 6 (F (1, 6) = 4.74; p<0.05).
108 5.3.6. Cholinergic deafferentation of mPFC reduced signal-evoked PPC unit activity in distractor trial blocks.
Cholinergic deafferentation of the mPFC did not significantly reduce the proportion of neurons exhibiting signal-evoked activation in Distractor sessions. During distractor task performance, following ipsilateral lesions, 9/85 (10%) neurons exhibited signal-evoked activation in distractor sessions. From the two animals initially lesioned contralaterally, 6/38 (16%) neurons exhibited signal related activation. Although these proportions of signal-related activity are relatively small, subjecting these two groups to a second lesion increased the proportion of neurons exhibiting signal related activity similar to baseline levels, as 26/103 (25%) neurons exhibited signal-evoked activation on hit trials, a population comparable to previous findings in intact animals (Broussard et al.,
2006). Analysis of individual trial blocks indicated that signal-related activity was relatively absent in the distractor block following ipsilateral mPFC SAP infusion. Only 4 neurons from collected under ipsilateral and bilateral conditions (188; 2%) exhibited signal-related neuronal activation during the distractor block. Unlike Experiment 2, in which PPC cholinergic deafferentation produced a relatively noisy firing rate in signal- activated neurons, this population remained rather quiescent during task performance (see
Figure 5.4). We calculated the median SNR for each cell during a standard task trial block and during the distractor block, and compared the two conditions. Following ipsilateral lesions, the SNR was significantly reduced for signal-activated neurons in distractor blocks relative to standard blocks (Z=-2.29; p<0.05, Wilcoxon). Following contralateral lesions, there was not a significant effect of distractor on SNR, though this may stem from the fact that there were only 3 neurons to analyze. When both groups
109 received a second lesion, resulting in bilateral deafferentation, the distractor again significantly reduced the SNR (Z=-4.05; p<0.05, Wilcoxon). Further analysis of this population in non-distractor trial blocks indicated that neuronal activation did not vary as a function of signal duration (χ2 = 1.53, p>0.05, Friedman test).
5.4. Discussion
This experiment provides further information regarding the changes in PPC
cortical neurophysiology that takes place during attentional processing and operant task
performance. The results of experiment 2 were replicated, with the demonstration of
increases in neuronal firing rate during the detection of cues. The imposition of a time
out for perseverate and premature responses, primarily to minimize the impulsivity
found in cholinergically lesioned animals in other visual attention tasks (Dalley et al.,
2004b), produced a greater number of omissions to both signal and nonsignal trials.
Cholinergic deafferentation of mPFC did not reduce the number of neurons exhibiting
detection related activity during distractor sessions, but reduced the SNR of parietal
neurons in the distractor block, which seemed to be a function of a lower baseline firing
rate in that block of trials. Thus, a population of parietal neurons reliably produced
signal-related increases in firing rate under baseline conditions, but became inhibited in
the distractor block. However, bilateral cholinergic deafferentation of mPFC did not
specifically impair the detection of signals in the distractor trial blocks, relative to
unilaterally lesioned sessions. Under both unilaterally and bilaterally lesioned
conditions, animals reliably detected signal and correctly rejected nonsignal at a
proportion similar to that of animals in Experiment 2. However, bilateral cholinergic
110 lesions increased the number of omissions on nonsignal trials, and the side bias toward
the hit lever remained intact, indicating an inability to integrate information about error
rates to shift toward a more successful attentional strategy.
5.4.1. Parameter manipulations inducing conditioned inhibition increase errors of
omission
In the variant of the sustained attention task developed by Bushnell (1994) and modified for visual attention (McGaughy and Sarter, 1995), the presentation of the signal is followed after a delay by the extension of two response levers. At the end of a trial, levers are retracted. Due to technical considerations, namely in order to minimize electrical noise, this task was further adapted to assess neurophysiological recordings in vivo neurophysiological recordings. Response levers were fixed and the opening of the response window relied not upon the extension of levers but rather a tone, which is identical for both signal and nonsignal trials. This manipulation restricts the action- outcome association of lever pressing to the 4s interval following the tone. Unlike the task modified by McGaughy and Sarter, premature and perseverative responses can be recorded in this task.
The cost of these measures of impulsive behavior is that animals can respond as often as possible with little response cost, which can result in an animal minimally attending to task relevant stimuli and still solve the task by constantly pressing the correct rejection lever until a salient light stimulus directs responding to the hit lever.
Overtraining tends to remedy this behavior over time in most rats, but roughly 10% of
111 animals perseverate at this strategy. As a result, premature trials are not analyzed in non- impulsive animals, and persistently impulsive animals are culled from study.
Given that bilateral cholinergic deafferentation of mPFC increased impulsive behavior in the 5-choice-serial reaction time task (5-CSRTT), task parameters were further modified to minimize this potentially confounding effect in order to validly assess sustained attention. Like this variant of the task, the response apparatus in the 5-CSRTT are available throughout the task. The 5-CSRTT employs a time out, which is a type of negative punishment which delays future access to reinforcement (Klein, 1996).
Despite the imposition of a time out for premature responses, omissions rates are similar in the 5-CSRTT to those of subjects from experiment 1 (Risbrough et al., 2002).
Because this task requires a conditioned inhibition during the delay between the signal and the opening of the response window, the imposition of a time out in this task may produce additional conditioned inhibition, resulting in higher omission rates. In a classic paper studying the effects of extinction on tasks requiring varying levels of effort,
Mowrer and Jones (1943; as reprinted in Mowrer, 1960) summarized thus:
This study takes as its point of departure the hypothesis that the elimination of a response, or habit, through its own nonrewarded repetition involves a conflict in which the fatigue thus generated instigates a response (resting) which is incompatible with and therefore tends to inhibit the original response.
5.4.2. Cholinergic deafferentation of mPFC and attentional performance
Cholinergic deafferentation of mPFC did not impair the correct detection of signals and rejection of nonsignals under standard conditions. There was an increase in nonsignal omissions in the second block of distractor trials, compared to sessions in which animals were unilaterally lesioned. Cholinergic deafferentation did not increase
112 the number of premature responses, most likely due to the response cost producing sufficiently strong conditioned inhibition. Further, the lack of an effect on accuracy and latency indicates that cholinergic input to the mPFC is not necessary for the discrimination between signals and nonsignals on a trial by trial basis. As mentioned in the introduction, in tasks with short delays do not require the medial prefrontal cortex.
Rather, the specific deficit observed in bilaterally deafferented animals is an inability to incorporate the error signals in order to shift response strategy to optimize performance.
Himmelheber et al (2001) originally observed that the visual distractor produced an initial side bias followed by a shift toward a neutral side bias in the second 6 min trial block. A side bias to the hit lever is produced by a subject incorrectly responding to the houselight as if it were a signal light. However, because the probability of the occurrence of a nonsignal trial (0.5) is greater than that of 25 or 50 ms signals (0.32) the gains in reward collection on shorter signal trials are outweighed by a significant loss in reward on the nonsignal trials. Normal rats can utilize the negative feedback of increased FA to shift to a side-neutral response pattern, requiring them to suppress the processing of the visual distractor. This return to near baseline performance was accompanied by an increase in cortical ACh efflux. Behavioral results from this study suggest that cholinergic input to the mPFC is necessary for mediating this shift in response pattern as animals perseverate in responding to the hit lever.
113 5.4.3. Rat PPC may only be required to resolve competition between two stimulus-
response contingencies in attention tasks.
The perseverant side bias in bilaterally deafferented rats during and following the distractor block corresponds with a decrease in signal-related unit activity in the PPC.
One might predict that increased relative responding to the signal trial would result in greater parietal activation. However, tasks using only signal trials have been shown not to require an intact parietal cortex (Muir et al., 1996), indicating that, on a trial by trial basis, PPC neurons are activated to resolve competing stimulus-response contingencies.
114
FIGURES
Figure 5.1.) Changes in task parameters and experimental design for experiment 3. Basic task rules are the same as in the General Methods and Figure 2.1. A.) Illustration of Correct and Premature responding and consequences. The top panel illustrates the onset of the visual signal followed 1s later by a tone, which opens the operant window. Responding to the correct lever following the tone results in a water reward and initiates the selection of the ITI for the next trial. If an animal impulsively responds during the ITI, the ITI is reset until the animal inhibits responding throughout the ITI period. The bottom panel illustrates the cancellation of a trial due to the premature responding in response to the light, prior to the onset of the tone. B.) One group of rats (n=5) received an infusion of SAP in the mPFC ipsilateral to the PPC recording probe. Following a 10- day recovery period, animal were retrained while hooked up to a unity gain headset for recording. After 9 days of standard (S) and distractor (D) sessions, animals will receive a second infusion of SAP in the contralateral, unlesioned hemisphere. Ten more days between infusion and testing are again provided to allow for cholinergic terminals to be deafferented, followed by another nine days of recording. C.) A second group of rats (n=2) received the same pattern of treatment, except that the contralateral mPFC was lesioned, then the ipsilateral hemisphere.
115
Figure 5.1.
116
Figure 5.2.) Cholinesterase deafferentation of mPFC slices relative to controls and parietal slices. A.) Photomicrograph (4X) of a cholinergically deafferented subject. B.) A 20X photograph from the same subject in the prelimbic region near the midline. C.) A 4X photograph of an intact subject, again in the prelimbic region. D.) A 20X photograph of the prelimbic region of a control animal. E & F.) The AChE fiber staining in the parietal cortex showed similar density to control slices at 4X (E.) and 20X (F.)
117
Figure 5.3.) Behavioral performance of animals following unilateral and bilateral lesions of mPFC, compared to animals in experiment 2. A.) The Side Bias scores. In the fourth, distractor block, the side bias returns to neutral levels in animals with unilateral deafferentation of mPFC. Bilateral deafferentation produces a significant side bias toward the hit lever in blocks 5 and 6, both non-distractor blocks. B & C.) There are considerably more omission trials on this variant of the task, relative to the previous task. C.) In the second distractor block (4), bilaterally lesioned animals had significantly higher omissions. In previous studies, animals regain performance on nonsignal trials in the second block of the distractor trials. Following bilateral lesions, animals fail to produce this adaptive performance.
118
Figure 5.4.) The interaction of distractor and mPFC cholinergic deafferentation on signal-related PPC activity. Population based PETHs were constructed from neurons that were significantly activated in either distractor or standard trial blocks. A.) Contralateral mPFC deafferentation produced a population of neurons activated by the signal during both the standard and distractor blocks. B.) Ipsilateral deafferentation produced a population of neurons which were only activated by the signal during standard trial blocks. C.) Following the second SAP infusion, all subjects (n=7) were pooled. The neurophysiological responses were similar to that following ipsilateral infusion only in that the detection of visual signals activated PPC neurons in standard trial blocks.
119
Figure 5.4.
120
CHAPTER 6
GENERAL DISCUSSION
6.1. Summary of main findings
The LFP of the PPC of rats produced attentional correlates that were modulated by the duration of the signal and the presence of the distractor. In both standard task performance and in the presence of the distractor, the p300 field potential had greater amplitude in response to 500 ms visual signals than for shorter signals. A longer latency potential, the CNV, had significantly greater amplitude when rats detected a signal, irrespective of signal duration, and the distractor increased the amplitude of the CNV.
Thus, the CNV has similar correlates to the single unit responses seen previously, in which long latency increases in firing rate precede correct detection performance
(Broussard et al., 2006). The spectral content of the evoked field potential indicates an increase in alpha power preceding the tone on hit, but not miss trials, corresponding to the signal-evoked increases in CNV amplitude. Increased alpha band activation also preceded the onset of the houselight early in the distractor task, during which rats produced the highest false alarm rate. As attentional performance improved throughout the distractor trial block, decreases in alpha band power preceding distractor onset corresponded to decreases in the false alarm rate.
121 Infusions of the specific cholinotoxic agent SAP reduced the proportion of neurons reactive to the signal, an effect that was amplified by increased attentional demand from a visual distractor. Cholinergically deafferented PPC neurons that were activated by the signal were more likely to also be activated by the distractor. Further, these neurons had an elevated baseline firing rate in the presence of the visual distractor, resulting in a lower signal to noise ratio relative to populations of neurons recorded from animals pre-lesion.
Infusions of SAP to mPFC contralateral to PPC recording probes did not reduce the number of signal-evoked parietal neurons, nor did signal-evoked PPC neurons vary their pattern of activation over trial block. Ipsilateral infusions of SAP did not significantly reduce the number of signal-evoked neurons, but the neuronal response to the signal was attenuated while the distractor was present, and in trial blocks subsequent to the presentation of the distractor. Bilateral deafferentation produced similar effects as ipsilateral deafferentation on the physiological responses. Contrary to our predictions, bilaterally deafferented subjects failed to produce significantly greater premature responding. Premature responding may have been inadvertently precluded by an alteration in the task that was intended as an improvement. Specifically, in the third experiment, a time out was introduced as a means to reduce impulsive responses. This change in task resulted in an increase in omissions. Bilateral lesions produced a specific behavioral deficit: following the return to standard task conditions in the third trial block, performance on the nonsignal trials was impaired. Although this finding did not support predictions made for Experiment 3, the ability of rats to regain performance following the
122 return to standard task conditions may require increases in attentional effort, and thus recruit cholinergic transmission in the mPFC. This idea is discussed further in section 6.4.
6.2. Signal duration dependence of the p300 amplitude
The local field potential provides additional information to single unit activity.
The source of the LFP is largely synaptic activity, whereas spiking activity is largely generated from pyramidal cells. Thus, the LFP is hypothesized to represent the inputs to and subthreshold processing within a cortical region, and spiking activity represents the output. The present study provides new evidence for a role of the PPC in rat visual attention. First, the temporal profile of the LFPs in the PPC is very different when animals produce hit versus miss responses. Most notably, there is an increase in the amplitude of the contingent negative variation, which corresponds to increases in the alpha power. Importantly, the alpha power increases late in the interval between signal and tone onset, indicating that the detection of the light produces a correlate of expectation, or perhaps response inhibition, prior to the tone. These data also show that the LFP activity in PPC is modulated by sensory salience and attentional load. First, the amplitude of the p300 response was longer following signals of longer duration. This correlates well with performance in the sustained attention task, in which the longer signals generate greater accuracy and shorter latencies. In humans, amplitude of the p300 component reflects the allocation of attentional resources during information processing
(Yordanova et al., 2001).
The results from Experiment 1 support the hypothesis that PPC LFP activity produces correlates of attention that are modulated by the distractor and by signal
123 duration. This finding complements that of the single unit data in Experiment 2 and our previously published results, both of which indicates a long latency increase in the firing rate of PPC neurons on hit, but not miss trials.
6.3. Effects of cholinergic deafferentation on signal-related cortical processing
Visual stimuli are represented in the primary visual cortex and are processed through both the dorsal and ventral visual stream. The dorsal stream, of which the PPC is a constituent, encodes a visuospatial map which tracks location and direction of stimuli.
In the primate PPC specifically, retinotopic and somatotopic maps are represented in anatomically parcellated subdivisions (Colby and Goldberg, 1999). In addition to a general function of spatial map representation, the PPC is hypothesized to encode shifts of attention to different locations and/or modalities (Shomstein and Yantis, 2006). In tasks which impose a delay between the cue and response, visual cues generate activity in the PPC which is relayed to frontal areas for further processing. This phenomenon was demonstrated in primates by Chafee and Goldman-Rakic (2000), in which inactivation of the PPC by cooling reduces cue-related activity in the dorsolateral prefrontal cortex. By contrast, the DLPFC recruits PPC activity during the delay of working memory tasks.
This suggests that PPC initiates the representation of the visual cue, whereas the DLPFC directs the maintenance of the cue “on-line” in the PPC. Although the rat mPFC is thought to be a functional homologue of the primate ACC and not DLPFC, this evidence suggests that the direction that information flows is from the PPC to frontal areas which feed back onto posterior areas during a retention interval. Anatomically, rat PFC
124 (specifically the ventral orbital area) sends efferents to the PPC (Reep et al., 1994), suggesting a putative circuitry by which mPFC can modulate PPC activity.
Modulation of the parietal firing rate can be thought to link the visuospatial input with behavior for both covert and overt attention. That the most prevalent correlate of
PPC single unit activity was a signal-driven response on hit trials suggests that the role of the rat PPC is primarily to signal the behavioral relevance of the cue. The PPC of primates (in particular the lateral intraparietal area, or LIP) is hypothesized to provide an intermediate level of processing which maintains the salience of objects in the visual field.
Parietal neurons are sensitive to bottom-up parameters, such as abrupt changes in stimulus luminance, movement, or color (Balan and Gottlieb, 2006). In this primate study it seemed the abruptness of stimulus change recruited activity, as increases and decreases in stimulus luminance produced the same pattern of neural activation. The response of LIP neurons is also modulated by top down factors, such as reward probability (Sugrue et al., 2004). The LIP can be thought to accumulate evidence within a visuospatial task schema based on behavioral relevance and space of stimuli. For example, when monkeys detected the direction of a proportion of moving dots, LIP activity within the receptive field increased faster when the greatest proportion of dots moved. If the monkey must saccade away from the LIP receptive field, then the firing rate of that neuron decreases as a function of motion strength (Roitman and Shadlen,
2002). Thus, the general function of LIP may be to translate perceptual inputs and behavioral expectations into a dynamic spatial map in which individual neurons represent the location of salient stimuli, whereas the value of the attended location is coded by the changes in firing rate (Gottlieb, 2007). The increases in neuronal firing rate during the
125 detection of visual signals from Chapters IV and V indicate that rat PPC, like the primate
LIP, is also capable of producing representations of behaviorally relevant stimuli.
Acetylcholinergic transmission in the PPC has been shown to contribute to the conditioned responding to increases in valence. Rats that received greater rewards following compound stimuli relative to single stimuli produce greater amount of food cup behavior in the absence of reward. Rats with cholinergic deafferented PPC failed to demonstrate the conditioning response to compound stimuli (Bucci et al., 1998). In the current study, cholinergically transmission in the PPC contributed to signal-evoked increases in firing rate, suggesting a cellular mechanism by which ACh can modulate the representation of valence in the PPC.
The results in Experiment 2 support the general hypothesis that cholinergic neurotransmission modulates cortical processing by increasing the signal-related responses of PPC neurons relative to the background firing rate (i.e. the “SNR”).
Importantly, fewer neurons elicited signal-related activation during both standard and distractor sessions following cholinergic deafferentation, suggesting that cholinergic neurotransmission may modulate cortical processing under basal conditions, a finding which is substantiated by the impairments found in monkeys performing an attention task following local infusion of scopolamine in the parietal cortex (Davidson et al., 1999).
The enhancement of the SNR of cortical neurons in this study is consistent with findings of several studies recording evoke responses from brain slice and anesthetized preparations. Application of cholinergic agonists enhanced the cortical responsiveness to sensory stimuli in visual cortex (Sillito and Kemp, 1983), and auditory cortex (Metherate et al., 1990). Cholinergic transmission is also necessary for plasticity in the
126 somatosensory cortex (Alenda and Nunez, 2007). Recordings from brain slices of the anterior cingulate cortex (McCormick and Prince, 1986) demonstrated that exogenous application of ACh produced an initial, phasic hyperpolarization of neurons followed by a tonic depolarization. This may represent a mechanism by which cortical neurons can inhibit processing from neighboring neurons while enhancing the processing of sensory stimuli within a specific window of time. Studies in awake animals demonstrated that the auditory evoked response (Berntson et al., 2003b) of rats to arousal generating stimuli is significantly reduced following infusions of SAP into the nucleus basalis magnocellularis.
Results from this thesis simply add to this literature in that cholinergic neurotransmission contributes to cortical neuronal processing of stimuli in an operant task, in addition to conditioned responses found in these previous studies.
6.4. The role of the medial prefrontal cortex in attentional processing
Imagine driving a car to an approaching intersection. When the light changes from green to yellow, you must decide whether to drive through the intersection or to apply the brakes for a full stop. A full range of variables influence this decision, including current velocity, distance to the intersection, and your estimation of the yellow light interval. On an oft-traveled road, you may have had enough previous success such that this estimation can be carried out rather automatically. However, if you are unfamiliar with the intersection, you will slow your reaction time and produce more errors, resulting in unnecessary delays, accidents, or a traffic ticket. The prediction errors experienced, particularly as the number of errors or cost of each error increases, can recruit activation in the mPFC. On subsequent trials this part of the brain, which generally serves to
127 monitor performance, may be reactivated, in conjunction with the DLPFC, to select the appropriate response among alternatives.
The role of the PFC generally is to monitor and modulate activity in subcortical and posterior areas. Sensory representation or motor commands are not encoded in the
PFC, which is only recruited when task performance is not automated. The PFC is not required to encode and respond to apparent go signals (green light + no traffic) or stop signals (red light + backed up traffic). However, the warning light in the scenario outlined above can produce what is termed a “response conflict” in cognitive neuroscience. Current models hypothesize that PFC function is to produce the cognitive control required for the resolution of response conflicts (Miller and Cohen, 2001). In the current study the occurrence of the tone could initiate a right or left lever response depending on: 1.) whether a signal was presented and 2.) whether or not the signal was sufficiently registered by the rat. Signals that are easily discriminated from the background (e.g. 500 ms signals) elicit highly accurate responses at fast latencies. The visual distractor produces an increase in false alarm rates early in the distractor period, suggesting that the distractor light produces greater response conflict in normal rats when the tone is presented on nonsignal trials. The improvements in attentional performance over the course of the distractor trial block indicate an improved resolution of this response conflict, and may require increased attentional effort.
Acetylcholinergic input to the mPFC is hypothesized to mediate increases in attentional effort under detrimental conditions. Cortical ACh efflux is elevated in attention-task performing animals (Passetti et al., 2000; Dalley et al., 2001; Arnold et al.,
2002; McGaughy et al., 2002). More specifically, increases in PFC ACh efflux were
128 associated with motivated performance of pharmacologically impaired rats. Disruption of NMDA receptor signaling in the basal forebrain, which specifically impairs performance on signal trials in the sustained attention task (Turchi and Sarter, 2001b), produced increases in ACh efflux above that of normal performance (Kozak et al., 2006).
Experiment 3 was designed to test the specific hypothesis that this improvement in performance over the course of the distractor trial block requires cholinergic transmission in the mPFC. The lack of a difference in attentional performance between unilaterally and bilaterally lesion rats in the distractor trial block fails to support the specific predictions made in Experiment 3. However, any accuracy effects during the distractor may be confounded in part by the increase in nonsignal omissions following bilateral cholinergic lesions. Further, deficits in nonsignal performance in the final non-distractor trial blocks suggest a role for cholinergic transmission in the mPFC to regain baseline performance following the cessation of the visual distractor.
The physiological evidence produced in Chapter V requires a caveat. Only 38 neurons were collected from the animals that initially received the contralateral mPFC lesion, as opposed to larger populations gathered in the ipsilateral and bilateral conditions.
Because this was the control group of this experiment, one cannot strongly infer that the attenuation of the signal evoked responses were not due in part to group differences.
Nevertheless, the signal-evoked correlates of the smaller control group did not vary their response pattern, whereas animals with ipsilateral mPFC lesions produced attenuated neurophysiological responses in the distractor and subsequent blocks, indicating a role of mPFC cholinergic transmission in modulating parietal representation of the visual signal under detrimental conditions. Thus, a lack of mPFC cholinergic transmission results in
129 impaired parietal processing of the visual signal under detrimental conditions. That the deficits in signal processing are accompanied by deficits in correct rejection were surprising. However, a bias toward a response lever may be one adaptation motivated lesioned rats can make to maximize their reward intake. Such an adaptation would minimize response conflict by predominantly optimizing one stimulus-response association. In tasks which requires one stimulus-response association, ACh has been shown to minimally influence performance (Chiba et al., 1999; Risbrough et al., 2002;
Conner et al., 2003).
In conclusion, the present studies indicate that the PPC of rats produces several correlates to attentional performance similar to those seen in primates and humans.
Synaptic input to the PPC, as measured by changes in components of the LFP, is modulated by changes in signal duration and distractors. The contingent negative variation and increases in PPC firing rate represents maintenance of the stimulus representation throughout the retention interval. Cholinergic transmission in the PPC contributes to signal evoked activity under standard and distractor conditions. Further, cholinergic input may optimize the contrast between signal-evoked activity and background activity by suppressing the overall firing rate of PPC neurons. Cholinergic transmission in the mPFC may contribute to the improvements in performance following the cessation of the distractor condition and may optimize the parietal representation of behaviorally relevant signals under conditions that require increased attentional effort.
130
LIST OF REFERENCES
Alenda A, Nunez A (2007) Cholinergic modulation of sensory interference in rat primary somatosensory cortical neurons. Brain Res 1133:158-167.
Andersen RA, Buneo CA (2002) Intentional maps in posterior parietal cortex. Annu Rev Neurosci 25:189-220.
Aoki F, Fetz EE, Shupe L, Lettich E, Ojemann GA (2001) Changes in power and coherence of brain activity in human sensorimotor cortex during performance of visuomotor tasks. Biosystems 63:89-99.
Apparsundaram S, Martinez V, Parikh V, Kozak R, Sarter M (2005) Increased capacity and density of choline transporters situated in synaptic membranes of the right medial prefrontal cortex of attentional task-performing rats. J Neurosci 25:3851- 3856.
Arnold HM, Burk JA, Sarter M, Bruno JP (2002) Cortical acetylcholine release in rats performing an operant sustained attention task or operant control procedures. Neuroscience 114:451-460.
Baker SN, Kilner JM, Pinches EM, Lemon RN (1999) The role of synchrony and oscillations in the motor output. Exp Brain Res 128:109-117.
Balan PF, Gottlieb J (2006) Integration of exogenous input into a dynamic salience map revealed by perturbing attention. J Neurosci 26:9239-9249.
Bastiaansen MC, Posthuma D, Groot PF, de Geus EJ (2002) Event-related alpha and theta responses in a visuo-spatial working memory task. Clin Neurophysiol 113:1882-1893.
131 Berger H (1929) Über das Elektroenkephalogramm des Menschen. Archiv für Psychiatrie und Nervenkrankheiten 87:527-570.
Berntson GG, Shafi R, Sarter M (2002) Specific contributions of the basal forebrain corticopetal cholinergic system to electroencephalographic activity and sleep/waking behavior. European Journal of Neuroscience 16:2453-2461.
Berntson GG, Sarter M, Cacioppo JT (2003a) Ascending visceral regulation of cortical affective information processing. Eur J Neurosci 18:2103-2109.
Berntson GG, Shafi R, Knox D, Sarter M (2003b) Blockade of epinephrine priming of the cerebral auditory evoked response by cortical cholinergic deafferentation. Neuroscience 116:179-186.
Berridge KC (2004) Motivation concepts in behavioral neuroscience. Physiol Behav 81:179-209.
Bichot NP, Rossi AF, Desimone R (2005) Parallel and serial neural mechanisms for visual search in macaque area V4. Science 308:529-534.
Birrell JM, Brown VJ (2000) Medial frontal cortex mediates perceptual attentional set shifting in the rat. J Neurosci 20:4320-4324.
Bisley JW, Goldberg ME (2006) Neural correlates of attention and distractibility in the lateral intraparietal area. J Neurophysiol 95:1696-1717.
Bisley JW, Krishna BS, Goldberg ME (2004) A rapid and precise on-response in posterior parietal cortex. J Neurosci 24:1833-1838.
Boksem MA, Meijman TF, Lorist MM (2005) Effects of mental fatigue on attention: an ERP study. Brain Res Cogn Brain Res 25:107-116.
Boudreau CE, Williford TH, Maunsell JH (2006) Effects of task difficulty and target likelihood in area V4 of macaque monkeys. J Neurophysiol 96:2377-2387.
132 Broussard J, Sarter M, Givens B (2006) Neuronal correlates of signal detection in the posterior parietal cortex of rats performing a sustained attention task. Neuroscience 143:407-417.
Bucci DJ, Holland PC, Gallagher M (1998) Removal of cholinergic input to rat posterior parietal cortex disrupts incremental processing of conditioned stimuli. J Neurosci 18:8038-8046.
Bucci DJ, Conley M, Gallagher M (1999) Thalamic and basal forebrain cholinergic connections of the rat posterior parietal cortex. NeuroReport 10:941-945.
Buffalo EA, Bertini G, Ungerleider LG, Desimone R (2005) Impaired filtering of distracter stimuli by TE neurons following V4 and TEO lesions in macaques. Cereb Cortex 15:141-151.
Bunge SA, Hazeltine E, Scanlon MD, Rosen AC, Gabrieli JDE (2002) Dissociable contributions of prefrontal and parietal cortices to response selection. NeuroImage 17:1562-1571.
Burcham KJ, Corwin JV, Stoll ML, Reep RL (1997) Disconnection of medial agranular and posterior parietal cortex produces multimodal neglect in rats. Behavioural Brain Research 86:41-47.
Burk JA, Herzog CD, Porter MC, Sarter M (2002) Interactions between aging and cortical cholinergic deafferentation on attention. Neurobiology of Aging 23:467- 477.
Bushnell PJ, Kelly KL, Crofton KM (1994) Effects of toluene inhalation on detection of auditory signals in rats. Neurotoxicol Teratol 16:149-160.
Buzsaki G (2006) Rhythms of the Brain, 1st Edition. New York, NY: Oxford University Press.
Buzsaki G, Draguhn A (2004) Neuronal oscillations in cortical networks. Science 304:1926-1929.
133 Buzsaki G, Bickford RG, Ponomareff G, Thal LJ, Mandel R, Gage FH (1988) Nucleus basalis and thalamic control of neocortical activity in the freely moving rat. J Neurosci 8:4007-4026.
Cabeza R, Anderson ND, Houle S, Mangels JA, Nyberg L (2000) Age-related differences in neural activity during item and temporal-order memory retrieval: a positron emission tomography study. J Cogn Neurosci 12:197-206.
Canolty RT, Edwards E, Dalal SS, Soltani M, Nagarajan SS, Kirsch HE, Berger MS, Barbaro NM, Knight RT (2006) High gamma power is phase-locked to theta oscillations in human neocortex. Science 313:1626-1628.
Cao YJ, Surowy CS, Puttfarcken PS (2005) Different nicotinic acetylcholine receptor subtypes mediating striatal and prefrontal cortical [3H]dopamine release. Neuropharmacology 48:72-79.
Cavada C, Goldman-Rakic PS (1989) Posterior parietal cortex in rhesus monkey: II. Evidence for segregated corticocortical networks linking sensory and limbic areas with the frontal lobe. J Comp Neurol 287:422-445.
Cavada C, Goldman-Rakic PS (1991) Topographic segregation of corticostriatal projections from posterior parietal subdivisions in the macaque monkey. Neuroscience 42:683-696.
Chafee MV, Goldman-Rakic PS (2000) Inactivation of parietal and prefrontal cortex reveals interdependence of neural activity during memory-guided saccades. J Neurophysiol 83:1550-1566.
Chandler HC, King V, Corwin JV, Reep RL (1992) Thalamocortical connections of rat posterior parietal cortex. Neuroscience Letters 143:237-242.
Chelazzi L, Miller EK, Duncan J, Desimone R (2001) Responses of neurons in macaque area V4 during memory-guided visual search. Cereb Cortex 11:761-772.
Chen LL, Lin L-H, Green EJ, Barnes CA, McNaughton BL (1994a) Head direction cells in the rat posterior cortex. II. Contributions of visual and ideothetic information to the directional firing. Experimental Brain Research 101:24-34.
134 Chen LL, Lin L-H, Green EJ, Barnes CA, McNaughton BL (1994b) Head-direction cells in the rat posterior cortex. I. Anatomical distribution and behavioral modulation. Experimental Brain Research 101:8-23.
Chiba AA, Bushnell PJ, Oshiro WM, Gallagher M (1999) Selective removal of cholinergic neurons in the basal forebrain alters cued target detection. NeuroReport 10:3119-3123.
Clayton EC, Rajkowski J, Cohen JD, Aston-Jones G (2004) Phasic activation of monkey locus ceruleus neurons by simple decisions in a forced-choice task. J Neurosci 24:9914-9920.
Colby CL, Goldberg ME (1999) Space and attention in parietal cortex. Annu Rev Neurosci 22:319-349.
Conner JM, Chiba AA, Tuszynski MH (2005) The basal forebrain cholinergic system is essential for cortical plasticity and functional recovery following brain injury. Neuron 46:173-179.
Conner JM, Culberson A, Packowski C, Chiba AA, Tuszynski MH (2003) Lesions of the basal forebrain cholinergic system impair task acquisition and abolish cortical plasticity associated with motor skill learning. Neuron 38:819-829.
Corbetta M, Miezin FM, Shulman GL, Peterson SE (1993) A PET study of visuospatial attention. The Journal of Neuroscience 13:1202-1226.
Costa RM, Lin SC, Sotnikova TD, Cyr M, Gainetdinov RR, Caron MG, Nicolelis MA (2006) Rapid alterations in corticostriatal ensemble coordination during acute dopamine-dependent motor dysfunction. Neuron 52:359-369.
Coull JT, Frackowiak RSJ, Frith CD (1998) Monitoring for target objects: activation of right frontal and parietal cortices with increasing time on task. Neuropsychologia 36:1325-1334.
Critchley M (1953) The Parietal Lobes. New York, N.Y.: Hafner Publishing Company.
Crone EA, Donohue SE, Honomichl R, Wendelken C, Bunge SA (2006) Brain regions mediating flexible rule use during development. J Neurosci 26:11239-11247. 135 Dalley JW, Cardinal RN, Robbins TW (2004a) Prefrontal executive and cognitive functions in rodents: neural and neurochemical substrates. Neurosci Biobehav Rev 28:771-784.
Dalley JW, McGaughy J, O'Connell MT, Cardinal RN, Levita L, Robbins TW (2001) Distinct changes in cortical acetylcholine and noradrenaline efflux during contingent and noncontingent performance of a visual attentional task. J Neurosci 21:4908-4914.
Dalley JW, Theobald DE, Bouger P, Chudasama Y, Cardinal RN, Robbins TW (2004b) Cortical cholinergic function and deficits in visual attentional performance in rats following 192 IgG-saporin-induced lesions of the medial prefrontal cortex. Cereb Cortex 14:922-932.
Dani JA, Bertrand D (2007) Nicotinic acetylcholine receptors and nicotinic cholinergic mechanisms of the central nervous system. Annu Rev Pharmacol Toxicol 47:699- 729.
Davidson MC, Cutrell EB, Marrocco RT (1999) Scopolamine slows the orienting of attention in primates to cued visual targets. Psychopharmacology 142:1-8.
De Weerd P, Desimone R, Ungerleider LG (2003a) Impairments in spatial generalization of visual skills after V4 and TEO lesions in macaques (Macaca mulatta). Behavioral Neuroscience 117:1441-1447.
De Weerd P, Desimone R, Ungerleider LG (2003b) Generalized deficits in visual selective attention after V4 and TEO lesions in macaques. The European Journal of Neuroscience 18:1671-1691.
Demiralp T, Ademoglu A, Istefanopulos Y, Basar-Eroglu C, Basar E (2001) Wavelet analysis of oddball P300. International Journal of Psychophysiology 39:221-227.
Destexhe A, Sejnowski TJ (2001) Thalamocortical Assemblies. New York, NY: Oxford University Press.
Detari L, Rasmussen DD, Semba K (1999) The role of basal forebrain neurons in tonic and phasic activation of the cerebral cortex. Progress in Neurobiology 58:249-277.
136 DiMattia BV, Posley KA, Fuster JM (1990) Crossmodal short-term memory of haptic and visual information. Neuropsychologia 28:17-33.
Disney AA, Domakonda KV, Aoki C (2006) Differential expression of muscarinic acetylcholine receptors across excitatory and inhibitory cells in visual cortical areas V1 and V2 of the macaque monkey. J Comp Neurol 499:49-63.
Driver J, Mattingly JB (1998) Parietal neglect and visual awareness. Nature Neuroscience 1:17-22.
Driver J, Vuilleumier P (2001) Perceptual awareness and its loss in unilateral neglect and extinction. Cognition 79:39-88.
Dykes RW, Dancause N, Miasnikov AA, Agueev V (2001) Rapid differential conditioning of the somatosensory evoked potential by changed patterns of brief innocuous tactile stimuli in waking rats is altered by atropine sulfate. Brain Research 910:74-80.
Edeline J-M (1993) Receptive field plasticity in the auditory cortex during frequency discrimination training: selective retuning independent of task difficulty. Behavioral Neuroscience 107:82-103.
Eggermann E, Serafin M, Bayer L, Machard D, Saint-Mleux B, Jones BE, Muhlethaler M (2001) Orexins/hypocretins excite basal forebrain cholinergic neurones. Neuroscience 108:177-181.
Ellison A, Rushworth M, Walsh V (2003) The parietal cortex in visual search: a visuomotor hypothesis. Suppl Clin Neurophysiol 56:321-330.
Ellison A, Schindler I, Pattison LL, Milner AD (2004) An exploration of the role of the superior temporal gyrus in visual search and spatial perception using TMS. Brain 127:2307-2315.
Ergenoglu T, Demiralp T, Bayraktaroglu Z, Ergen M, Beydagi H, Uresin Y (2004) Alpha rhythm of the EEG modulates visual detection performance in humans. Brain Res Cogn Brain Res 20:376-383.
137 Erisir A, Levey AI, Aoki C (2001) Muscarinic receptor M(2) in cat visual cortex: laminar distribution, relationship to gamma-aminobutyric acidergic neurons, and effect of cingulate lesions. J Comp Neurol 441:168-185.
Fellous JM, Sejnowski TJ (2000) Cholinergic induction of oscillations in the hippocampal slice in the slow (0.5-2 Hz), theta (5-12 Hz), and gamma (35-70 Hz) bands. Hippocampus 10:187-197.
Flynn DD, Ferrari-DiLeo G, Mash DC, Levey AI (1995) Differential regulation of molecular subtypes of muscarinic receptors in Alzheimer's disease. J Neurochem 64:1888-1891.
Fournier GN, Materi LM, Semba K, Rasmussen DD (2004) Cortical acetylcholine release and electroencephalogram activation evoked by ionotropic glutamate receptor agonists in the rat basal forebrain. Neuroscience 123:785-792.
Freund TF, Gulyas AI (1991) GABAergic interneurons containing calbindin D28K or somatostatin are major targets of GABAergic basal forebrain afferents in the rat neocortex. J Comp Neurol 314:187-199.
Friedman-Hill SR, Robertson LC, Desimone R, Ungerleider LG (2003) Posterior parietal cortex and the filtering of distractors. Proceedings of the National Academy of Sciences of the United States of America 100:4263-4268.
Fries P, Reynolds JH, Rorie AE, Desimone R (2001) Modulation of oscillatory neuronal synchronization by selective visual attention. Science 291:1560-1563.
Fuster JM (1997) The Prefrontal Cortex: Anatomy, Physiology, and the Neuropsychology of the Frontal Lobe, Third Edition. New York, NY: Lippincott-Raven.
Ge S, Dani JA (2005) Nicotinic acetylcholine receptors at glutamate synapses facilitate long-term depression or potentiation. J Neurosci 25:6084-6091.
Gehring WJ, Knight RT (2002) Lateral prefrontal damage affects processing selection but not attention switching. Brain Res Cogn Brain Res 13:267-279.
138 Gill TM, Sarter M, Givens B (2000) Sustained visual attention performance-associated prefrontal neuronal activity: evidence for cholinergic modulation. J Neurosci 20:4745-4757.
Gill TM, Masters J, Sarter M, Givens B (1999) The role of acetylcholine within the medial prefrontal and posterior parietal cortices during sustained visual attention in the rat. Society of Neuroscience Abstracts 25.
Gioanni Y, Rougeot C, Clarke PB, Lepouse C, Thierry AM, Vidal C (1999) Nicotinic receptors in the rat prefrontal cortex: increase in glutamate release and facilitation of mediodorsal thalamo-cortical transmission. Eur J Neurosci 11:18-30.
Gold PE (2003) Acetylcholine modulation of neural systems involved in learning and memory. Neurobiol Learn Mem 80:194-210.
Golmayo L, Nunez A, Zaborszky L (2003) Electrophysiological evidence for the existences of a posterior cortical-prefrontal-basal forebrain circuitry in modulating sensory responses in visual and somatosensory rat cortical areas. Neuroscience 119:597-609.
Gotti C, Zoli M, Clementi F (2006) Brain nicotinic acetylcholine receptors: native subtypes and their relevance. Trends Pharmacol Sci 27:482-491.
Gottlieb J (2007) From Thought to Action: The Parietal Cortex as a Bridge between Perception, Action, and Cognition. Neuron 53:9-16.
Gottlieb J, Kusunoki M, Goldberg ME (1998) The representation of visual salience in monkey parietal cortex. Nature 391:481-484.
Greuel JM, Luhmann HJ, Singer W (1988) Pharmacological induction of use-dependent receptive field modifications in the visual cortex. Science 242:74-77.
Gu Q (2003) Contribution of acetylcholine to visual cortex plasticity. Neurobiol Learn Mem 80:291-301.
Gu Q, Singer W (1989) The role of muscarinic acetylcholine receptors in ocular dominance plasticity. Exs 57:305-314.
139 Gu Q, Singer W (1993) Effects of intracortical infusion of anticholinergic drugs on neuronal plasticity in kitten striate cortex. Eur J Neurosci 5:475-485.
Gulledge AT, Stuart GJ (2005) Cholinergic inhibition of neocortical pyramidal neurons. J Neurosci 25:10308-10320.
Hasselmo ME (2006) The role of acetylcholine in learning and memory. Curr Opin Neurobiol 16:710-715.
Hasselmo ME, Bower JM (1992) Cholinergic suppression specific to intrinsic not afferent fiber synapses in rat piriform (olfactory) cortex. J Neurophysiol 67:1222- 1229.
Hasselmo ME, Wyble BP (1997) Free recall and recognition in a network model of the hippocampus: simulating effects of scopolamine on human memory function. Behav Brain Res 89:1-34.
Hasselmo ME, McGaughy J (2004) High acetylcholine levels set circuit dynamics for attention and encoding and low acetylcholine levels set dynamics for consolidation. Progress in Brain Research 145:207-231.
Hasselmo ME, Anderson BP, Bower JM (1992) Cholinergic modulation of cortical associative memory function. J Neurophysiol 67:1230-1246.
Hazeltine E, Poldrack R, Gabrieli JD (2000) Neural activation during response competition. J Cogn Neurosci 12 Suppl 2:118-129.
Himmelheber AM, Sarter M, Bruno JP (2000) Increases in cortical acetylcholine release during sustained attention performance in rats. Brain Research Cognitive Brain Research 9:313-325.
Himmelheber AM, Sarter M, Bruno JP (2001) The effects of manipulations of attentional demand on cortical acetylcholine release. Brain Research Cognitive Brain Research 12:353-370.
Hopfinger JB, West VM (2006) Interactions between endogenous and exogenous attention on cortical visual processing. Neuroimage 31:774-789.
140 Hopfinger JB, Buonocore MH, Mangun GR (2000) The neural mechanisms of top-down attentional control. Nat Neurosci 3:284-291.
Hughes SW, Crunelli V (2005) Thalamic mechanisms of EEG alpha rhythms and their pathological implications. Neuroscientist 11:357-372.
Hung J, Driver J, Walsh V (2005) Visual selection and posterior parietal cortex: effects of repetitive transcranial magnetic stimulation on partial report analyzed by Bundesen's theory of visual attention. J Neurosci 25:9602-9612.
Huxter J, Burgess N, O'Keefe J (2003) Independent rate and temporal coding in hippocampal pyramidal cells. Nature 425:828-832.
Ipata AE, Gee AL, Gottlieb J, Bisley JW, Goldberg ME (2006) LIP responses to a popout stimulus are reduced if it is overtly ignored. Nat Neurosci 9:1071-1076.
Itti L, Koch C (2001) Computational modelling of visual attention. Nat Rev Neurosci 2:194-203.
Jimenez-Capdeville M, Dykes RW, Myasnikov AA (1997) Differential control of cortical activity by the basal forebrain in rats: a role for both cholinergic and inhibitory influences. The Journal of Comparative Neurology 381:53-67.
Jonides J, Schumacher EH, Smith EE, Koeppe RA, Edward A, Reuter-Lorenz PA, Marchuetz C, Willis CR (1998) The role of parietal cortex in verbal working memory. The Journal of Neuroscience 18:5026-5034.
Kalaska JF (1996) Parietal cortex area 5 and visuomotor behavior. Canadian Journal of Physiology and Pharmacology 74:483-498.
Kastner S, Pinsk MA, De Weerd P, Desimone R, Ungerleider LG (1999) Increased activity in human visual cortex during directed attention in the absence of visual stimulation. Neuron 22:751-761.
Kesner RP, Holbrook T (1987) Dissociation of item and order spatial memory in rats following medial prefrontal cortex lesions. Neuropsychologia 25:653-664.
141 Kimura F (2000) Cholinergic modulation of cortical function: A hypothetical role in shifting the dynamics in cortical network. Neuroscience Research 38:19-26.
Kimura F, Fukuda M, Tsumoto T (1999) Acetylcholine suppresses the spread of excitation in the visual cortex revealed by optical recording: possible differential effect depending on the source of input. European Journal of Neuroscience 11:3597-3609.
King V, Corwin JV (1993) Comparisons of hemi-inattention produced by unilateral lesions of the posterior parietal cortex or medial agranular prefrontal cortex in rats: neglect, extinction, and the role of stimulus distance. Behavioural Brain Research 54:117-131.
Klein SB (1996) Learning: principles and application, 3rd Edition. New York, NY: McGraw-Hill, Inc.
Klimesch W (1999) EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res Brain Res Rev 29:169-195.
Klimesch W, Sauseng P, Hanslmayr S (2007) EEG alpha oscillations: the inhibition- timing hypothesis. Brain Res Rev 53:63-88.
Klimesch W, Doppelmayr M, Russegger H, Pachinger T, Schwaiger J (1998) Induced alpha band power changes in the human EEG and attention. Neurosci Lett 244:73-76.
Knox D (2005) Nucleus basalis cholinergic lesions and defense responses. In: Psychology, p 115. Columbus, OH: The Ohio State University.
Kolb B, Walkey J (1987) Behavioral and anatomical studies of the posterior parietal cortex in the rat. Behavioural Brain Research 23:127-145.
Kolb B, Buhrmann K, McDonald R, Sutherland RJ (1994) Dissociation of the medial prefrontal, posterior parietal, and posterior temporal cortex for spatial navigation and recognition memory in the rat. Cereb Cortex 4:664-680.
Kozak R, Bruno JP, Sarter M (2006) Augmented prefrontal acetylcholine release during challenged attentional performance. Cereb Cortex 16:9-17. 142 Le Dantec C, Gontier E, Paul I, Charvin H, Bernard C, Lalonde R, Rebai M (2007) ERPs associated with visual duration discriminations in prefrontal and parietal cortex. Acta Psychol (Amst) 125:85-98.
Lee KH, Choi YY, Gray JR, Cho SH, Chae J-H, Lee S, Kim K (2006) Neural correlates of superior intelligence: stronger recruitment of posterior parietal cortex. NeuroImage 29:578-586.
Lee MG, Hassani OK, Alonso A, Jones BE (2005) Cholinergic basal forebrain neurons burst with theta during waking and paradoxical sleep. J Neurosci 25:4365-4369.
Leichnetz GR (2001) Connections of the medial posterior parietal cortex (area 7m) in the monkey. Anat Rec 263:215-236.
Levy RB, Reyes AD, Aoki C (2006) Nicotinic and muscarinic reduction of unitary excitatory postsynaptic potentials in sensory cortex; dual intracellular recording in vitro. J Neurophysiol 95:2155-2166.
Lin SC, Gervasoni D, Nicolelis MA (2006) Fast modulation of prefrontal cortex activity by basal forebrain non-cholinergic neuronal ensembles. J Neurophysiol.
Linden DE (2005) The p300: where in the brain is it produced and what does it tell us? Neuroscientist 11:563-576.
Linster C, Hasselmo ME (2001) Neuromodulation and the functional dynamics of piriform cortex. Chem Senses 26:585-594.
Logothetis NK (2003a) The underpinnings of the BOLD functional magnetic resonance imaging signal. J Neurosci 23:3963-3971.
Logothetis NK (2003b) The Underpinnings of the BOLD Functional Magnetic Resonance Imaging Signal. J Neurosci 23:3963-3971.
Long JM, Kesner RP (1998) Effects of hippocampal and parietal cortex lesions on memory for egocentric distance and spatial location information in rats. Behav Neurosci 112:480-495.
143 Lukatch HS, MacIver MB (1997) Physiology, pharmacology, and topography of cholinergic neocortical oscillations in vitro. J Neurophysiol 77:2427-2445.
Maalouf M, Miasnikov AA, Dykes RW (1998) Blockade of cholinergic receptors in rat barrel cortex prevents long-term changes in the evoked potential during sensory preconditioning. Journal of Neurophysiology 80:529-545.
Maddux JM, Kerfoot EC, Chatterjee S, Holland PC (2007) Dissociation of attention in learning and action: effects of lesions of the amygdala central nucleus, medial prefrontal cortex, and posterior parietal cortex. Behav Neurosci 121:63-79.
Manns ID, Alonso A, Jones BE (2000) Discharge properties of juxtacellularly labeled and immunohistochemically identified cholinergic basal forebrain neurons recorded in association with the electroencephalogram in anesthetized rats. The Journal of Neuroscience 20:1505-1518.
Marois R, Chun MM, Gore JC (2000) Neural correlates of the attentional blink. Neuron 28:299-308.
Martinez V, Sarter M (2004) Lateralized attentional functions of cortical cholinergic inputs. Behavioral Neuroscience 118:984-991.
Matsuyama S, Matsumoto A (2003) Epibatidine induces long-term potentiation (LTP) via activation of alpha4beta2 nicotinic acetylcholine receptors (nAChRs) in vivo in the intact mouse dentate gyrus: both alpha7 and alpha4beta2 nAChRs essential to nicotinic LTP. J Pharmacol Sci 93:180-187.
McAdams CJ, Maunsell JH (1999) Effects of attention on orientation-tuning functions of single neurons in macaque cortical area V4. J Neurosci 19:431-441.
McCormick DA, Prince DA (1985) Two types of muscarinic response to acetylcholine in mammalian cortical neurons. Proc Natl Acad Sci U S A 82:6344-6348.
McCormick DA, Prince DA (1986) Mechanisms of action of acetylcholine in the guinea- pig cerebral cortex in vitro. J Physiol 375:169-194.
144 McCormick DA, Pape HC (1990) Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in thalamic relay neurones. J Physiol 431:291-318.
McDaniel W, F, Williams LB, Cullen MA, Compton DM (1998) Turn-signal utilization by rats with either unilateral or bilateral posterior parietal cortex injuries. Psychobiology 26:143-152.
McDaniel W, F, Via J, D, Smith J, S, Wells D, L, Fu J, J, Bishop J, F, Boyd P, A, Ledesma HM (1995) Unilateral injury of posterior parietal cortex and spatial learning in hooded rats. Behavioural brain research 70:165-179.
McGaughy J, Sarter M (1995) Behavioral vigilance in rats: task validation and effects of age, amphetamine and benzodiazepine receptor ligands. Psychopharmacology 117:340-357.
McGaughy J, Sarter M (1998) Sustained attention performance in rats with intracortical infusions of 192 IgG-saporin-induced cortical cholinergic deafferentation: effects of physostigmine and FG 7142. Behav Neurosci 112:1519-1525.
McGaughy J, Kaiser T, Sarter M (1996) Behavioral vigilance following infusions of 192 IgG-saporin into the basal forebrain: selectivity of the behavioral impairment and relation to cortical AChE-positive fiber density. Behavioral Neuroscience 110:247-265.
McGaughy J, Decker MW, Sarter M (1999) Enhancement of sustained attention performance by the nicotinic acetylcholine receptor agonist ABT-418 in intact but not basal forebrain-lesioned rats. Psychopharmacology (Berl) 144:175-182.
McGaughy J, Dalley JW, Morrison CH, Everitt BJ, Robbins TW (2002) Selective behavioral and neurochemical effects of cholinergic lesions produced by intrabasalis infusions of 192 IgG-saporin on attentional performance in a five- choice serial reaction time task. J Neurosci 22:1905-1913.
McGehee DS, Heath MJ, Gelber S, Devay P, Role LW (1995) Nicotine enhancement of fast excitatory synaptic transmission in CNS by presynaptic receptors. Science 269:1692-1696.
145 McLin III DE, Miasnikov AA, Weinberger NM (2002) Induction of behavioral associative memory by stimulation of the nucleus basalis. Proceedings of the National Academy of Sciences 99:4002-4007.
Mesulam M-M (1981) A cortical network for directed attention and unilateral neglect. Annals of Neurology 10:309-325.
Metherate R, Ashe JH, Weinberger NM (1990) Acetylcholine modifies neuronal acoustic rate-level functions in guinea pig auditory cortex by an action at muscarinic receptors. Synapse 6:364-368.
Miller EK, Cohen JD (2001) An integrative theory of prefrontal cortex function. Annu Rev Neurosci 24:167-202.
Miner LA, Ostrander M, Sarter M (1997) Effects of ibotenic acid-induced loss of neurons in the medial prefrontal cortex of rats on behavioral vigilance: evidence for executive dysfunction. J Psychopharmacol 11:169-178.
Mountcastle VB, Lynch JC, A G, Sakata H, Acuna C (1974) Posterior parietal association cortex of the monkey: command functions for operations within extrapersonal space. Journal of Neurophysiology.
Mowrer OH (1960) Learning Theory and Behavior. New York, NY: John Wiley & Sons, Inc.
Mowrer OH, Jones HM (1943) Extinction and behavior variability as functions of effortfulness of task. J exp Psychol 33:369-386.
Mrzljak L, Levey AI, Goldman-Rakic PS (1993) Association of m1 and m2 muscarinic receptor proteins with asymmetric synapses in the primate cerebral cortex: morphological evidence for cholinergic modulation of excitatory neurotransmission. Proc Natl Acad Sci U S A 90:5194-5198.
Mrzljak L, Levey AI, Rakic P (1996) Selective expression of m2 muscarinic receptor in the parvocellular channel of the primate visual cortex. Proc Natl Acad Sci U S A 93:7337-7340.
146 Mrzljak L, Levey AI, Belcher S, Goldman-Rakic PS (1998) Localization of the m2 muscarinic acetylcholine receptor protein and mRNA in cortical neurons of the normal and cholinergically deafferented rhesus monkey. J Comp Neurol 390:112- 132.
Muir JL, Everitt BJ, Robbins TW (1996) The cerebral cortex of the rat and visual attentional function: dissociable effects of mediofrontal, cingulate, anterior dorsolateral, and parietal cortex lesions on a five-choice serial reaction time task. Cereb Cortex 6:470-480.
Muller CM, Lewandowski MH, Singer W (1993) Structures mediating cholinergic reticular facilitation of cortical responses in the cat: effects of lesions in immunocytochemically characterized projections. Exp Brain Res 96:8-18.
Murphy ER, Dalley JW, Robbins TW (2005) Local glutamate receptor antagonism in the rat prefrontal cortex disrupts response inhibition in a visuospatial attentional task. Psychopharmacology (Berl) 179:99-107.
Murthy VN, Fetz EE (1996) Oscillatory activity in sensorimotor cortex of awake monkeys: synchronization of local field potentials and relation to behavior. J Neurophysiol 76:3949-3967.
Nakamura K (1999) Auditory spatial discriminatory and mnemonic neurons in rat posterior parietal cortex. Journal of Neurophysiology 82:2503-2517.
Nelson C, Sarter M, Bruno J (2005) Prefrontal cortical modulation of acetylcholine release in posterior parietal cortex. Neuroscience 132:347-359.
Nieto-Escamez FA, Sanchez-Santed F, de Bruin JP (2002) Cholinergic receptor blockade in prefrontal cortex and lesions of the nucleus basalis: implications for allocentric and egocentric spatial memory in rats. Behav Brain Res 134:93-112.
Ongur D, Price JL (2000) The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb Cortex 10:206-219.
Parasuraman R, Warm JS, Dember WN (1987) Vigilance: Taxonomy and utility. In: Ergonomics and Human Factors: Recent Research (Mark LS, Warm JS, Huston RL, eds). New York: Springer Verlag.
147 Pare M, Wurtz RH (1997) Monkey posterior parietal cortex neurons antidromically activated from superior colliculus. J Neurophysiol 78:3493-3497.
Passetti F, Dalley JW, O'Connell MT, Everitt BJ, Robbins TW (2000) Increased acetylcholine release in the rat medial prefrontal cortex during performance of a visual attentional task. Eur J Neurosci 12:3051-3058.
Paxinos G, Watson C (1998) The Rat Brain: in Stereotaxic Coordinates, 4th Edition. New York: Academic Press.
Pesaran B, Pezaris JS, Sahani M, Mitra PP, Andersen RA (2002) Temporal structure in neuronal activity during working memory in macaque parietal cortex. Nature Neuroscience 5:805-811.
Posner MI, Petersen SE (1990) The attention system of the human brain. Annu Rev Neurosci 13:25-42.
Posner MI, Dehaene S (1994) Attentional networks. Trends Neurosci 17:75-79.
Posner MI, Raichle ME (1994) Images of Mind: Scientific American Books.
Preuss T (1995) Do rats have a prefrontal cortex? The Rose-Woolsey-Akert program reconsidered. J Cogn Neurosci 7:1-24.
Rasband W (1997-2006) ImageJ. In. Bethesda, MA: U.S. National Institutes of Health.
Reep RL, Chandler HC, King V, Corwin JV (1994) Rat posterior parietal cortex: topography of corticocortical and thalamic connections. Experimental Brain Research 100:67-84.
Reynolds JH, Pasternak T, Desimone R (2000) Attention increases sensitivity of V4 neurons. Neuron 26:703-714.
Rich EL, Shapiro ML (2007) Prelimbic/infralimbic inactivation impairs memory for multiple task switches, but not flexible selection of familiar tasks. J Neurosci 27:4747-4755.
148 Rihs TA, Michel CM, Thut G (2007) Mechanisms of selective inhibition in visual spatial attention are indexed by alpha-band EEG synchronization. Eur J Neurosci 25:603- 610.
Risbrough V, Bontempi B, Menzaghi F (2002) Selective immunolesioning of the basal forebrain cholinergic neurons in rats: effect on attention using the 5-choice serial reaction time task. Psychopharmacology (Berl) 164:71-81.
Roberts MJ, Zinke W, Guo K, Robertson R, McDonald JS, Thiele A (2005) Acetylcholine dynamically controls spatial integration in marmoset primary visual cortex. J Neurophysiol 93:2062-2072.
Robinson TE, Berridge KC (2001) Incentive-sensitization and addiction. Addiction 96:103-114.
Rodriguez R, Kallenbach U, Singer W, Munk MH (2004) Short- and long-term effects of cholinergic modulation on gamma oscillations and response synchronization in the visual cortex. J Neurosci 24:10369-10378.
Roelfsema PR, Engel AK, Konig P, Singer W (1997) Visuomotor integration is associated with zero time-lag synchronization among cortical areas. Nature 385:157-161.
Roitman JD, Shadlen MN (2002) Response of neurons in the lateral intraparietal area during a combined visual discrimination reaction time task. J Neurosci 22:9475- 9489.
Rosner AL, Mittleman G (1996) Visuospatial attention in the rat and posterior parietal cortex lesions. Behav Brain Res 79:69-77.
Rushworth MF, Taylor PC (2006) TMS in the parietal cortex: Updating representations for attention and action. Neuropsychologia 44:2700-2716.
Sarter M, Bruno JP, Givens B (2003) Attentional functions of cortical cholinergic inputs: what does it mean for learning and memory? Neurobiology of Learning and Memory 80:245-256.
149 Sarter M, Gehring WJ, Kozak R (2006) More attention must be paid: the neurobiology of attentional effort. Brain Res Brain Res Rev 51:145-160.
Sarter M, Hasselmo M, Bruno JP, Givens B (2005) Unraveling the attentional functions of cortical cholinergic inputs: interactions between signal-driven and cognitive modulation of signal detection. Brain Res Rev 48:98-111.
Sato H, Hata Y, Masui H, Tsumoto T (1987) A functional role of cholinergic innervation to neurons in the cat visual cortex. J Neurophysiol 58:765-780.
Sauseng P, Klimesch W, Doppelmayr M, Pecherstorfer T, Freunberger R, Hanslmayr S (2005a) EEG alpha synchronization and functional coupling during top-down processing in a working memory task. Hum Brain Mapp 26:148-155.
Sauseng P, Klimesch W, Stadler W, Schabus M, Doppelmayr M, Hanslmayr S, Gruber WR, Birbaumer N (2005b) A shift of visual spatial attention is selectively associated with human EEG alpha activity. Eur J Neurosci 22:2917-2926.
Scheperjans F, Palomero-Gallagher N, Grefkes C, Schleicher A, Zilles K (2005) Transmitter receptors reveal segregation of cortical areas in the human superior parietal cortex: relations to visual and somatosensory regions. Neuroimage 28:362-379.
Scherberger H, Jarvis MR, Andersen RA (2005) Cortical local field potential encodes movement intentions in the posterior parietal cortex. Neuron 46:347-354.
Schurmann M, Basar-Eroglu C, Kolev V, Basar E (2001) Delta responses and cognitive processing: single-trial evaluations of human visual P300. Int J Psychophysiol 39:229-239.
Sereno AB, Amador SC (2006) Attention and memory-related responses of neurons in the lateral intraparietal area during spatial and shape-delayed match-to-sample tasks. J Neurophysiol 95:1078-1098.
Sher E, Chen Y, Sharples TJ, Broad LM, Benedetti G, Zwart R, McPhie GI, Pearson KH, Baldwinson T, De Filippi G (2004) Physiological roles of neuronal nicotinic receptor subtypes: new insights on the nicotinic modulation of neurotransmitter release, synaptic transmission and plasticity. Curr Top Med Chem 4:283-297.
150 Shomstein S, Yantis S (2004) Control of attention shifts between vision and audition in human cortex. J Neurosci 24:10702-10706.
Shomstein S, Yantis S (2006) Parietal cortex mediates voluntary control of spatial and nonspatial auditory attention. J Neurosci 26:435-439.
Sillito AM, Kemp JA (1983) Cholinergic modulation of the functional organization of the cat visual cortex. Brain Res 289:143-155.
Singer W, Gray CM (1995) Visual feature integration and the temporal correlation hypothesis. Annu Rev Neurosci 18:555-586.
Singh J, Knight RT (1990) Frontal lobe contribution to voluntary movements in humans. Brain Res 531:45-54.
Snyder L (2000) Moving forward by looking away. Nature 408:921-923.
Spencer KM, Polich J (1999) Poststimulus EEG spectral analysis and P300: attention, task, and probability. Psychophysiology 36:220-232.
Steinmetz MA, Constantinidis C (1995) Neurophysiological evidence for a role of posterior parietal cortex in redirecting visual attention. Cereb Cortex 5:448-456.
Steriade M, Descarries L (2006) Cholinergic Modulation of Cortical Activity. In: The Brain Cholinergic System in Health and Disease (Pepeu G, Giacobini E, eds). London, UK: Informa Healthcare.
Steriade M, McCormick DA, Sejnowski TJ (1993) Thalamocortical oscillations in the sleeping and aroused brain. Science 262:679-685.
Sugrue LP, Corrado GS, Newsome WT (2004) Matching behavior and the representation of value in the parietal cortex. Science 304:1782-1787.
Swick D, Knight RT (1998) Brain Lesions and Attention. In: The Attentive Brain (Parasuraman R, ed), pp 143-163. Boston, MA: MIT Press.
151 Tago H, Kimura H, Maeda T (1986) Visualization of detailed acetylcholinesterase fiber and neuron staining in rat brain by a sensitive histochemical procedure. J Histochem Cytochem 34:1431-1438.
Thut G, Nietzel A, Brandt SA, Pascual-Leone A (2006) Alpha-band electroencephalographic activity over occipital cortex indexes visuospatial attention bias and predicts visual target detection. J Neurosci 26:9494-9502.
Turchi J, Sarter M (2001a) Bidirectional modulation of basal forebrain N-methyl-D- aspartate receptor function differentially affects visual attention but not visual discrimination performance. Neuroscience 104:407-417.
Turchi J, Sarter M (2001b) Antisense oligodeoxynucleotide-induced suppression of basal forebrain NMDA-NR1 subunits selectively impairs visual attentional performance in rats. The European Journal of Neuroscience 14:103-117.
Ungerleider LG (1995) Functional brain imaging studies of cortical mechanisms for memory. Science 270:769-775.
Uylings HB, van Eden CG (1990) Qualitative and quantitative comparison of the prefrontal cortex in rat and in primates, including humans. Prog Brain Res 85:31- 62.
Uylings HB, Groenewegen HJ, Kolb B (2003) Do rats have a prefrontal cortex? Behav Brain Res 146:3-17.
Varela F, Lachaux JP, Rodriguez E, Martinerie J (2001) The brainweb: phase synchronization and large-scale integration. Nat Rev Neurosci 2:229-239.
Verdier D, Dykes RW (2001) Long-term cholinergic enhancement of evoked potentials in rat hindlimb somatosensory cortex displays characteristics of long-term potentiation. Experimental Brain Research 137:71-82.
Vidal C, Changeux JP (1993) Nicotinic and muscarinic modulations of excitatory synaptic transmission in the rat prefrontal cortex in vitro. Neuroscience 56:23-32.
Volpicelli LA, Levey AI (2004) Muscarinic acetylcholine receptor subtypes in cerebral cortex and hippocampus. Prog Brain Res 145:59-66. 152 von Stein A, Chiang C, Konig P (2000) Top-down processing mediated by interareal synchronization. Proc Natl Acad Sci U S A 97:14748-14753.
Wang YP, Kawai Y, Nakashima K (1999) Rabbit P300-Like Potential Depends on Cortical Muscarinic Receptor Activation. Neuroscience 89:423-427.
Wenk GL, Stoehr JD, Quintana G, Mobley S, Wiley RG (1994) Behavioral, biochemical, histological, and electrophysiological effects of 192 IgG-saporin injections into the basal forebrain of rats. The Journal of Neuroscience 14:5986-5995.
Whitehouse PJ (2004) Paying attention to acetylcholine: the key to wisdom and quality of life? Prog Brain Res 145:311-317.
Wiesel TN, Hubel DH (1963a) Effects of visual deprivation on morphology and physiology of cells in the cats lateral geniculate body. J Neurophysiol 26:978-993.
Wiesel TN, Hubel DH (1963b) Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol 26:1003-1017.
Wiesel TN, Hubel DH (1965) Extent of recovery from the effects of visual deprivation in kittens. J Neurophysiol 28:1060-1072.
Wiest MC, Nicolelis MA (2003) Behavioral detection of tactile stimuli during 7-12 Hz cortical oscillations in awake rats. Nat Neurosci 6:913-914.
Wilke M, Logothetis NK, Leopold DA (2006) Local field potential reflects perceptual suppression in monkey visual cortex. Proc Natl Acad Sci U S A 103:17507-17512.
Woertz M, Pfurtscheller G, Klimesch W (2004) Alpha power dependent light stimulation: dynamics of event-related (de)synchronization in human electroencephalogram. Brain Res Cogn Brain Res 20:256-260.
Yamaguchi S, Knight RT (1990) Gating of somatosensory input by human prefrontal cortex. Brain Res 521:281-288.
153 Yantis S, Schwarzbach J, Serences JT, Carlson RL, Steinmetz MA, Pekar JJ, Courtney SM (2002) Transient neural activity in human parietal cortex during spatial attention shifts. Nat Neurosci 5:995-1002.
Yordanova J, Kolev V, Polich J (2001) P300 and alpha event-related desynchronization (ERD). Psychophysiology 38:143-152.
Yu AJ, Dayan P (2002) Acetylcholine in cortical inference. Neural Netw 15:719-730.
Yu AJ, Dayan P (2005) Uncertainty, neuromodulation, and attention. Neuron 46:681-692.
Zaborszky L, Gaykema RP, Swanson DJ, Cullinan WE (1997) Cortical input to the basal forebrain. Neuroscience 79:1051-1078.
Zar JH (2007) Biostatistical Analysis. Englewood Cliffs, NJ: Prentiss Hall.
Zhang M, Barash S (2000) Neuronal switching of sensorimotor transformations for antisaccades. Nature 408:971-975.
Zhang M, Barash S (2004) Persistent LIP activity in memory antisaccades: working memory for a sensorimotor transformation. Journal of Neurophysiology 91:1424- 1441.
Zinke W, Roberts MJ, Guo K, McDonald JS, Robertson R, Thiele A (2006) Cholinergic modulation of response properties and orientation tuning of neurons in primary visual cortex of anaesthetized Marmoset monkeys. Eur J Neurosci 24:314-328.
Zoli M, Picciotto MR, Ferrari R, Cocchi D, Changeux JP (1999) Increased neurodegeneration during ageing in mice lacking high-affinity nicotine receptors. Embo J 18:1235-1244.
154