Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Elucidating the neurobiology of problem gambling using a novel rodent slot machine task. Cocker, Paul J. 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_september_cocker_paul.pdf [ 4.23MB ]
Metadata
JSON: 24-1.0307411.json
JSON-LD: 24-1.0307411-ld.json
RDF/XML (Pretty): 24-1.0307411-rdf.xml
RDF/JSON: 24-1.0307411-rdf.json
Turtle: 24-1.0307411-turtle.txt
N-Triples: 24-1.0307411-rdf-ntriples.txt
Original Record: 24-1.0307411-source.json
Full Text
24-1.0307411-fulltext.txt
Citation
24-1.0307411.ris

Full Text

ELUCIDATING THE NEUROBIOLOGY OF PROBLEM GAMBLING USING A NOVEL RODENT SLOT MACHINE TASK by  Paul J. Cocker  B.Sc. (honors) Psychology, Sussex University, 2006  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Neuroscience)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2016  © Paul J. Cocker, 2016 ii  Abstract Gambling is an enjoyable and innocuous past-time for many, but for some it can become a maladaptive compulsion akin to drug or alcohol addiction.  Despite increasing recognition that the phenomenological process underlying both substance and behavioural addictions may be similar, treatment options for problem gambling remain limited, and of questionable efficacy.   Animal models offer an invaluable opportunity to not only study the underlying neurobiology of disorders such as gambling, but may also facilitate the development of novel pharmacotherapies. To that end we have developed a rodent slot machine task (rSMT) that suggests rats share key behavioural features with human gamblers. The dopamine D2-like receptor family is critically involved in modulating animals’ performance on the rSMT. Specifically, the D4 receptor appears to contribute to animals’ attributions of salience to game-related stimuli. D4 receptors are principally located within prefrontal cortical regions and consequently represent an intriguing target for modulating higher order cognitive processes. In addition to systemic pharmacology, we have demonstrated that disruption of neural regions that are relatively rich in D4 receptors, such as the anterior cingulate and insular cortex, impact animals’ ability to accurately respond to reward-related stimuli on the rSMT; further emphasising a role for these receptors in gambling-related decision making. Additionally, we have used the rSMT to try and model iatrogenic gambling observed in patients with Parkinson’s disease (PD). This form of compulsive gambling arises de-novo in a small but significant sub-set of patients following adjunctive therapy with D2-like agonists. Chronic administration of a D2/3 receptor agonist galvanized performance on the rSMT, a finding that could be considered translationally analogous to the compulsive play exhibited by some PD patients. These alterations in performance were accompanied by an increase in the transcription iii  factor pCREB in the nucleus accumbens. Administration of the β-adrenoreceptor blocker propranolol, which putatively attenuates this increase in pCREB, ameliorates the compulsive-like task engagement.   Ultimately, gambling is a heterogeneous disorder that is unlikely to have a single underlying aetiology.  However, these data indicate that aberrant dopaminergic signalling within the D2-like receptor family may underlie at least some of the cognitive perturbations observed in problem gambling.   iv  Preface  Introduction – (chapter 2, all sections) parts of this introduction have previously been published in similar form in reviews written by the candidate. Cocker, PJ & Winstanley, CA (2015) Towards a better understanding of disordered gambling: efficacy of animal paradigms in modelling aspects of gambling behaviour. Current Addiction Reports 2:240-248. And Cocker, PJ & Winstanley, CA (2015) Irrational beliefs, biases and gambling: exploring the role of animal models in elucidating vulnerabilities for the development of pathalogical gambling. Behavioral Brain Research, 279, 259-273. I wrote both of these reviews . Dr Catharine Winstanley edited the manuscripts.   Experiment 1 – (chapter 3, all sections) has previously been published in the manuscript Cocker, PJ, Le Foll, B, Rogers, RD, Winstanley, CA (2014) A selective role for dopamine D4 receptors in modulating reward expectancy in a rodent slot machine task. Biological psychiatry, 15;75(10): 817-24. I conducted all animal testing, prepared and administered all pharmacological agents, performed all data and statistical analysis, made all figures and wrote the manusciript. Dr’s Bernard Le Foll and Dr. Robert Rogers contributed to experimental design. Dr. Catharine Winstanley contributed to experimental design, manuscript preperation and revision.   Experiment 2- (chapter 4, all sections) has been accepted for publication; Cocker, PJ, Vonder Harr, C & Winstanley, CA (June, 2016) Elucidating the role of D4 receptors in mediating attributions of salience to incentive stimuli on Pavlovian conditioned approach and conditioned reinforcement paradigms. Behavioural Brain Research.  I conducted all animal testing, prepared v  and administered all pharmacological agents, performed all data and statistical analysis, made all figures and wrote the manusciript. Dr. Vonder Harr contributed to experimental design. Dr. Winstanley contributed to experimetnal design and edited the manuscript.   Experiment 3 – (chapter 5, all sections) has been published by Cocker PJ, Hosking JG, Murch, WS, Clark, L & Winstanley CA (2016) Activation of dopamine D4 receptors within the anterior cingulate cortex enhances the erroneous expectation of reward on a rat slot machine task, Neuropharmacology. I conducted all animal testing, performed surgery, prepared and administered all pharmacological agents, performed all data and statistical analysis, made all figures and wrote the manusciript. Dr. Jay Hosking also performed surgery. Spencer Murch aided with data collection. Dr Clark contributed to experimental design.  Dr. Catharine Winstanley contributed to both experimental design and manuscript revision.   Experiment 4 – (chapter 6, all sections) has been accepted for publication; Cocker, PJ, Lin, MY, Barrus, MB, Le Foll, B & Winstanley, CA (June, 2016) “The agranular and granular insula differentially contribute to gambling-like behavior on a rat slot machine task: effects of inactivation and local infusion of a dopamine D4 agonist on reward expectancy”. Psychopharmacology.  I performed surgery, prepared and administered all pharmacological agents, performed all data and statistical analysis, made all figures and wrote the manusciript. Amy Lin conducted all animal testing. Michael Barrus also performed surgery. Dr. Bernard Le Foll contributed to experimental design.  Dr. Catharine Winstanley contributed to both experimental design and manuscript revision.   vi  Experiment 5 – (chapter 7, all sections) has been submitted for publication by Cocker, PJ, Tremblay, M, Kaur, S & Winstanley, CA (May 2016) The dopamine D2/3 agonist ropinirole invigorates performance and induces compulse-like gambling behaviour on a rodent slot machine task. I conducted all animal testing, performed surgery, prepared and administered all pharmacological agents, prepared samples for Western blot analysis, conducted all PCR analysis, performed all statistical analysis, made all figures and wrote the manusciript. Melanie Trembaly also performed surgery. Sukhbir Kaur aided in molecualr biology techqnues. Dr. Winstanley contributed to experimental design and edited the manscript.   Experiment 6 – (chapter 8, all sections) Cocker PJ, Lin, MY., Tremblay M & Winstanley CA. The β-adrenoreceptor blocker propranolol ameliorates compulsive-like gambling behaviour on a rodent slot machine task; implications for iatrogenic gambling. I performed surgery, prepared and administered all pharmacological agents, performed all data and statistical analysis, made all figures and wrote the manusciript Amy Lin conducted all animal testing. Melanie Tremblay.  Dr. Catharine Winstanley contributed to both experimental design and manuscript revision.  General discussion – (Chapter 9, sections 9.2 & 9.3) A version of these sections have previously been published in reviews written by the candidate. Cocker, PJ & Winstanley, CA (2015) Towards a better understanding of disordered gambling: efficacy of animal paradigms in modelling aspects of gambling behaviour. Current Addiction Reports 2:240-248. And Cocker, PJ & Winstanley, CA (2015) Irrational beliefs, biases and gambling: exploring the role of animal models in elucidating vulnerabilities for the development of pathalogical gambling. Behavioral vii  Brain Research, 279, 25-273. I wrote both of these reviews . Dr Catharine Winstanley edited the manuscripts.  All animal testing was performed in accordance with the Canadian Council on Animal Care (CCAC) and received ethical approval by the University of British Columbia (UBC) Animal Care Committee (ACC), certificate numbers A-11-0123 & A-13-0011.  viii  Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents ....................................................................................................................... viii List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xiv List of Abbreviations ................................................................................................................. xvi Acknowledgements ................................................................................................................... xvii Chapter 1: General Introduction .................................................................................................1 1.1 A brief introduction to gambling disorder ...................................................................... 1 1.2 Modelling gambling-related decision making in the rat ................................................. 3 1.2.2 Impulsivity ...................................................................................................................... 4 1.2.3 Decision making deficits. ............................................................................................... 8 1.3 The role of cognitive biases and distortions in gambling ............................................. 12 1.4 The rSMT and experimental objectives ........................................................................ 16 Chapter 2: General Methods ......................................................................................................19 2.1 Subjects ......................................................................................................................... 19 2.2 Behavioural apparatus ................................................................................................... 19 2.3 rSMT: Habituation and pre-task training ...................................................................... 20 2.4 SMT: Task training ....................................................................................................... 20 2.5 The rodent Slot Machine Task (rSMT) ......................................................................... 22 2.6 Microinfusions (chapters 5 and 6) ................................................................................ 23 ix  2.7 Osmotic pump implantation (chapters 7 and 8) ............................................................ 24 2.8 Behavioural measures for the rSMT ............................................................................. 25 2.9 Statistical analysis ......................................................................................................... 25 Chapter 3: Experiment 1: A selective role for dopamine D4 receptors in modulating reward expectancy in a rodent slot machine task. .................................................................................30 3.1 Introduction ................................................................................................................... 30 3.2 Additional methods ....................................................................................................... 33 3.3 Results ........................................................................................................................... 34 3.4 Discussion ..................................................................................................................... 41 Chapter 4: Experiment 2: Elucidating the role of D4 receptors in mediating attributions of salience to incentive stimuli on Pavlovian conditioned approach and conditioned reinforcement paradigms ............................................................................................................55 4.1 Introduction ................................................................................................................... 55 4.2 Additional methods ....................................................................................................... 58 4.3 Results ........................................................................................................................... 62 4.4 Discussion ..................................................................................................................... 65 Chapter 5: Experiment 3: Activation of dopamine D4 receptors within the anterior cingulate cortex enhance the erroneous expectation of reward on a rat slot machine task. .77 5.1 Introduction ................................................................................................................... 77 5.2 Additional methods ....................................................................................................... 79 5.3 Results ........................................................................................................................... 80 5.4 Discussion ..................................................................................................................... 85 x  Chapter 6: Experiment 4: The agranular and granular insula differentially contribute to gambling-like behaviour on a rat slot machine task: effects of inactivation and local infusion of a dopamine D4 agonist on reward expectancy. .......................................................99 6.1 Introduction ................................................................................................................... 99 6.2 Additional methods ..................................................................................................... 102 6.3 Results ......................................................................................................................... 103 6.4 Discussion ................................................................................................................... 107 Chapter 7: Experiment 5: The dopamine D2/3 agonist ropinirole invigorates performance and induces compulsive-like gambling behaviour on a rodent slot machine task. ..............123 7.1 Introduction ................................................................................................................. 123 7.2 Additional methods ..................................................................................................... 126 7.3 Results ......................................................................................................................... 129 7.4 Discussion ................................................................................................................... 134 Chapter 8: Experiment 6: The β-adrenoreceptor blocker propranolol ameliorates compulsive-like gambling behaviour on a rodent slot machine task: implications for iatrogenic gambling. ..................................................................................................................152 8.1 Introduction ................................................................................................................. 152 8.2 Additional methods ..................................................................................................... 155 8.3 Results ......................................................................................................................... 158 8.4 Discussion ................................................................................................................... 164 Chapter 9: General Discussion .................................................................................................179 9.1 Summary of experimental findings ............................................................................. 179 9.2 Theoretical implications and considerations for future studies .................................. 183 xi  9.2.1 Investigate any role for the cholinergic and noradrenergic systems on the rSMT. ... 183 9.2.2 Optogenetic examination of the putative circuitry underlying rSMT performance ... 184 9.2.3 Explore the relationship between dysfunctional reward expectancy and propensity toward drug self-administration ......................................................................................... 185 9.2.4 Sex differences in gambling ....................................................................................... 185 9.3 Limitations and critical considerations ....................................................................... 186 9.4 Concluding remarks .................................................................................................... 190 Bibliography ...............................................................................................................................191  xii  List of Tables  Table 3.1 Latency to respond on the left lever by trial type at baseline and during pharmacological challenges. ......................................................................................................... 49 Table 3.2 Latency to respond at subsequent hole based on the status of the previous hole for baseline and pharmacological challenges. .................................................................................... 51 Table 3.3 Trials completed at baseline and following pharmacological challenges ..................... 53 Table 3.4 Relative binding sites and affinities for drugs. ............................................................. 54 Table 5.1 Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges .......................................................................................................... 97 Table 5.2 Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges. .................................................................................... 98 Table 5.3 Trials completed at baseline and following pharmacological challenges ..................... 98 Table 6.1A Agranular insula; Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges. ...................................................................................... 120 Table 6.1B Granular insula; Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges ....................................................................................... 120 Table 6.2A Agranular insula; Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges ....................................................... 121 Table 6.2B Granular insula; Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges. ...................................................... 121 Table 6.3A Agranular insula; Trials completed at baseline and following pharmacological challenges .................................................................................................................................... 122 xiii  Table 6.3B Granular insula; Trials completed at baseline and following pharmacological challenges. ................................................................................................................................... 122 Table 7.1 The sequence of the primers used to detect mRNA for target genes using qPCR ..... 149 Table 7.2A Latency to respond on the collect lever by trial type at baseline and for different time-points during saline administration .................................................................................... 149 Table 7.2B Latency to respond on the collect lever by trial type at baseline and for different time-points during ropinirole administration. ..................................................................................... 150 Table 7.3A Latency to respond at subsequent hole based on the statues of the previous hole for baseline and different time-points during saline administration ................................................. 150 Table 7.3b Latency to respond at subsequent hole based on the statues of the previous hole for baseline and different time-points during ropinirole administration. ......................................... 151 Table 7.4 Changes in mRNA or protein in rats treated with chronic ropinirole or saline from both early and late time points. Data are expressed as fold change from control ± SEM .................. 151 Table 8.1 Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges. ....................................................................................................... 176 Table 8.2 Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges ................................................................................... 177  xiv  List of Figures Figure 2.1 The standard five-hole operant chamber used for the rSMT. ...................................... 27 Figure 2.2 Schematic diagram showing the trial structure of the rSMT. ...................................... 29 Figure 3.1 Baseline performance of rSMT. .................................................................................. 46 Figure 3.2 Effects of the D2-like agonist quinpirole, the D4 agonist PD168077 and the D4 antagonist L-745,870 on performance of the rSMT. .................................................................... 47 Figure 3.3 Effect of the D2 antagonist L-741,626, the D3 antagonist SB27701A and the D3 agonist PD129807 on performance on the rSMT. ........................................................................ 48 Fig 3.4 Effect of pre-treatment with antagonists selective for different D2- receptor subtypes on the response to quinpirole. ............................................................................................................ 49 Figure 4.1 Summary of goal- and sign-tracking responses during training and following PD168077 administration ............................................................................................................. 73 Figure 4.2 Head entry responses during training sessions during CRf training. .......................... 74 Figure 4.3 Summary of behavioural responses on single CRf test day following administration nof either saline or PD168077. ..................................................................................................... 76 Figure 5.1 Histological analysis of cannula .................................................................................. 93 Figure 5.2 Baseline rSMT performance ........................................................................................ 93 Figure 5.3 Effects of ACC inactivation on rSMT performance.................................................... 94 Figure 5.4 Effect of activation of local D4 receptors within the ACC via infusion of the D4 agonist PD168077 on rSMT performance. ................................................................................... 95 Figure 5.5 Effect of preceding trial type on subsequent task performance .................................. 97 Figure 6.1 Histological analysis of cannula ................................................................................ 117 Figure 6.2 Baseline rSMT performance ...................................................................................... 117 xv  Figure 6.3 Effects of insula inactivation on rSMT performance ................................................ 118 Figure 6.4 Effect of activation of local D4 receptors within the insular cortex via infusion of the D4 agonist PD168077 on rSMT performance. ........................................................................... 119 Figure 7.1 Baseline rSMT performance ...................................................................................... 142 Figure 7.2 Effect of chronic ropinirole on reward expectancy. .................................................. 144 Figure 7.3 Effect of chronic ropinirole on the number of trials completed. ............................... 144 Figure 7.4 Long term behavioural effects of chronic ropinirole administration. ........................ 145 Figure 7.5 Summary of changes in protein level in the dorsal striatum in animals treated with chronic ropinirole in comparison to saline .................................................................................. 147 Figure 7.6 Summary of changes in protein level in the nucleus accumbens in animals treated with chronic ropinirole in comparison to saline. ................................................................................. 149 Figure 8.1 Baseline rSMT performance ...................................................................................... 171 Figure 8.2 Effect of dopamine depletions in the dorsal striatum via bilateral infusion of 6-OHDA..................................................................................................................................................... 172 Figure 8.3 Effects of dietary lithium and chronic ropinirole on reward expectancy .................. 174 Figure 8.4 Effects of repeated injections of the β-adrenoreceptor blocker propranolol and chronic ropinirole on reward expectancy ................................................................................................. 174 Figure 8.5 Effects of chronic ropinirole, dietary lithium and propranolol on the number of trials completed .................................................................................................................................... 175 Figure 8.6 Effects of 6-OHDA and chronic ropinirole on the number of adjusting steps made during the forelimb adjusting step test procedure ....................................................................... 176  xvi  List of Abbreviations rSMT – rodent Slot Machine Task  5-CSRTT – Five-Choice Serial Reaction-Time Task IGT – Iowa Gambling Task  rGT – rat Gambling Task ACC – Anterior cingulate cortex BLA – Basolateral amygdala NAc – Nucleus accumbens  OFC – Orbitofrontal cortex  PFC – Prefrontal cortex PrL – Prelimbic Cortex IL – Infralimbic Cortex VMPFC – Ventromedial prefrontal cortex 5-HT – Serotonin  GD – Gambling Disorder PD – Parkinson’s Disease Li+ - Lithium chloride  EGM’s – Electronic Gaming Machines  xvii  Acknowledgements  Thank you to Dr. Catharine Winstanley. Cath, you gave me a job when I first came to Canada (I’m still not really sure why), you’ve been an incredible role model and mentor throughout my time in the lab. Your encouragement and support throughout this process has been instrumental in me making it this far. Any chance I have to succeed as a scientist I owe to you.  Cath, thanks.  Thank you to the lab, particularly those whom without their meaningfully contributions this dissertation would not exist. Most importantly Amy Lin, as dependable and enthusiastic a volunteer as one could possibly hope for. Amy – thank you for all your work, you not only made this thesis possible but also gave me time to be a deadbeat dad for a while. I hope you enjoy being a ‘real’ doctor. Melanie Tremblay – thanks for always forging a path and all your help with PC6 and AL2, it’s almost enough to forgive you for your inexplicable love of Nickelback. Michael Barrus – Barry, thanks for you work on PC5 and for occasionally making me go out into the wilderness to be whimsical. Wendy Adams, for being ‘lab mum’ and your unfailing enthusiasm. And to Jay Hosking, not only for your scientific contributions to this thesis and being unfailing amiable, but mostly for persuading me that ‘any idiot can do this’. Thank you to Dr. Christian Schutz for giving me a chance to try my hand at human research, you support and guidance. Thanks also to the other members of my supervisory committee, Dr. Jon Stoessl and Dr. Alasdair Barr.  Thank you to both my own and Isabelle’s family. Nicole and Wolfgang, thanks for making me feel a part of your family, your advice and support throughout my time here. To my parents, Karen and John, I realise moving to the other side of the world to do this wasn’t ‘ideal’, xviii  thank you for the encouragement and always being there.  I don’t think I can adequately articulate how grateful I am to you both; I am unfathomably lucky to have you as parents.  Finally, thank you to Isabelle Linden, for being my wife, my best friend and my voice of reason. You are nothing short of incredible, your understanding and patience throughout this entire process has been awesome. I couldn’t have done this without you.  And lastly, to Thomas John Cocker, your smiles have made the last few months bearable. You are my most significant outcome.    1  Chapter 1: General Introduction    1.1 A brief introduction to gambling disorder        Gambling or wagering on uncertain outcomes has been reported throughout human history. Present in virtually every culture, gambling is undertaken and enjoyed by the majority of individuals within society (Gerstein 1999; Wardle. H 2010). For most people gambling is simply a harmless pastime, but for some it can become a maladaptive compulsion with severe repercussions on an individuals’ quality of life in a manner akin to drug or alcohol addiction (Potenza 2006).  Legislation surrounding gambling has been progressively relaxed (in western countries) over the last several decades, which, coupled with the proliferation of on-line gambling, has resulted in unprecedented access to gambling opportunities (Griffiths 1999). Gambling disorder (GD) is consequently a growing public health concern (Shaffer and Korn 2002), with societal costs estimated to exceed $5 billion per year in the United States alone (Gerstein 1999).  Prevalence rates for GD have been estimated to be between 0.5-2.5% (Ladouceur et al. 1999; Petry et al. 2005); with an additional 5-15% engaging with gambling at a problematic, but sub-clinical level (Shaffer and Hall 2001).  These numbers belie the relative ubiquity of gambling within the general population. Representative samples in the United States, Canada & the United Kingdom have recently suggested some level of gambling participation in 72-86% of the population (Gerstein 1999; J 2000; Wardle. H 2010).  Given that gambling is widespread throughout society, the question arises as to why some individuals develop clinically significant problems with gambling whereas to most it is simply a harmless recreational pastime.  The reasons people gamble may be multifarious and complex, yet monetary gain is undoubtedly 2  a major component (Neighbors et al. 2002; Walker 1992).  However, the allure of financial benefit is insufficient to account for the persistence of gambling, especially as the vast majority of gamblers are aware that the odds of winning are against them (Rachlin 1990).  Thus it could be argued that the commencement of any sort of gamble is somewhat irrational, in that participants are knowingly placing themselves at a disadvantage.  At the heart of these apparently irrational decisions is the finding that people do not value potential gains and losses equally.  Daniel Bernouli first proposed that utility is not linear for gains and losses (Bernoulli 1954). This was formalized by Kahneman and Tversky in prospect theory which stated that the value function, although roughly ‘S’ shaped, is concave in the gain domain and convex in the loss domain (Kahneman and Tversky 1979).  In other words, losses loom larger than gains. Interest in the etiology of GD is rising rapidly, particularly following the recognition that GD and more ‘traditional’ forms of addition, such as drug or alcohol addiction, may best be considered as equivalent (see Potenza 2006 for review; Potenza 2008). This has culminated in GD being reclassified as an addiction disorder in the DSM V.  Despite purportedly overlapping aetiologies, gambling addiction differs from drug or alcohol dependence in that there is no ingestion of psychoactive substances.  Problem gambling may therefore offer an ideal platform from which to make inferences about the development of the cycle of addiction, both cognitively and neurobiologically, independent of any changes induced by the pharmacological actions of drugs themselves (Bechara 2003).  However, problematic gambling is often concurrent with affective disorders, as well as substance dependence (Martin et al. 2013), making it more difficult to truly remove confounds relating to drug use and other psychiatric issues from the analysis of addictive behaviour.  Animal studies offer a potential solution, in that preclinical behavioural paradigms with strong face and construct validity can inform research into the 3  phenomenological and neurobiological underpinnings of behaviour without the problematic issue of causality or comorbidity that is inherent within human studies.  Moreover treatment options for GD are currently limited (Grant et al. 2012) and only partial success has been achieved with pharmacological (Grant and Kim 2006) and behavioural interventions (Ladouceur et al. 2001). Therefore, animal models offer real potential to not only facilitate a more comprehensive understanding of the individual differences that contribute toward vulnerability to addictive disorders, but may meaningfully inform future treatment strategies.    1.2 Modelling gambling-related decision making in the rat  Although animal models cannot mimic GD per se, they can model aspects of gambling-related behaviour and thus make a critical contribution to our understanding of the cognitive processes underlying engagement in gambling. Considering different gambling behaviours as potentially subject to independent expression and regulation, rather than assuming a universal “pro-gambling” phenotype from the outset, may be more appropriate when trying to elucidate risk factors for disorders like GD in which the etiology is complex and likely multifactorial.   Moreover, such an approach is in-line with emerging diagnostic frameworks (Kirmayer and Crafa 2014; Morris and Cuthbert 2012).  To that end gambling in humans could be considered to encompass dysfunction within three core arenas; namely increased impulsivity (Dixon et al. 2003; Lawrence et al. 2009; Michalczuk et al. 2011; Petry 2001; Potenza 2007; Rodriguez-Jimenez et al. 2006), disadvantageous cost/benefit decision making (Brand et al. 2005; Goudriaan et al. 2005; Linnet et al. 2011) and increased endorsement of irrational beliefs or cognitive biases pertaining to gambling (Billieux et al. 2012; Gaboury and Ladouceur 1989; Michalczuk et al. 2011).  Of these a great deal of excellent preclinical work in rodent models has 4  been conducted on both impulsivity and cost benefit decision making, greatly informing our understanding about the neurobiological basis of both traits. However, investigations examining the role of cognitive biases in in the development of GD have, up to now, been virtually non-existent.   1.2.2 Impulsivity Impulsivity can broadly be defined as acting or making decisions without appropriate forethought (Winstanley et al. 2006). Although some level of impulsivity can be adaptive in both human and animal populations, high levels of impulsivity inevitably result in deleterious consequences and are associated with a wide-range of behavioural disorders, including problem gambling (Chamberlain and Sahakian 2007; Verdejo-Garcia et al. 2008).   Factor-analysis of self-report questionnaires, as well as analyses of intra-individual behavioural variation, indicate that impulsivity is a non-unitary construct.  While there is still some debate over the most appropriate way to define and measure different aspects of impulsivity, the Barratt Impulsivity Scale (BIS-11) remains one of the most commonly-used assessments scale (Patton et al. 1995).  This metric typically identifies three sub-components of motor, non-planning and attentional impulsivity.  Most of the established rodent models of impulsive behaviours have been designed to model one of these sub-types (see Evenden 1999 for review). Of these paradigms, two that are widely employed and have arguably the most face, construct and predictive validity are the 5-choice serial-reaction time-task (5-CSRTT) and delay discounting paradigms. The 5-CSRTT is loosely analogous to the continuous performance task (CPT) in humans, and has recently been successfully back-translated into humans, further confirming its validity (Young et al. 2013). The CPT requires participants to scan a 5-digit sequence and respond only when that sequence matches a ‘target’ sequence. Errors of commission occur when the subject responds prematurely to a sequence that matches the target 5  stimulus in all but the last number.  To avoid such impulsive responses, subjects must therefore wait to fully identify the target sequence. Subjects with GD make more of these premature responses on the CPT than healthy controls, indicative of greater motor impulsivity (Kertzman et al. 2008). The 5-CSRTT requires animals to scan the 5-hole array in order to accurately detect the brief illumination (typically 0.5s) of one of the apertures.  The animal must make a “nosepoke” response in the hole that was illuminated in order to gain food reward, thereby providing a measure of animals’ visuospatial attention.  Responses made prematurely, before the stimulus light is illuminated, generates an index of motor impulsivity (Robbins 2002).   The 5-CSRTT has been used extensively in numerous labs around the world, such that a comprehensive review of both the pharmacological and neurobiological regulation of task performance will not be undertaken here (see Robbins 2002 for review). Yet, what does appear clear is that not only are various aspects of performance behaviourally and pharmacologically dissociable, but they also depend on distinct core neural loci.  With specific regard to premature responses on the 5-CSRTT, amphetamine has particularly prominent and robust effects (Cole and Robbins 1987).  Amphetamine is a psychostimulant that potentiates the release of the monoamines dopamine, serotonin (5-HT) and noradrenaline (Sulzer et al. 2005).  Amphetamine-induced elevations in premature responding have been independently replicated by numerous groups, and can be attenuated by administration of both systemic and intra-accumbens dopaminergic antagonists, as well as selective ablation of dopaminergic terminals in the striatum (Cole and Robbins 1987).  Critically, similar elevations in impulsivity are also observed after administration of the selective dopamine reuptake inhibitor GBR12909, as well as other psychostimulants such as cocaine and methylphenidate (van Gaalen et al. 2006a).  Although serotonergic and noradrenergic compounds, such as the selective 5-HT2C receptor antagonist 6  SB242084 (Winstanley et al. 2004b), and alpha-2 receptor antagonist yohimbine (Sun et al. 2010), also increase premature responding, neither 5-HT or noradrenaline-specific reuptake inhibitors mimic this action of amphetamine (Baarendse and Vanderschuren 2012).  As such, it would appear that, in the most general terms, potentiation of dopaminergic signaling has a particularly prominent role to play in mediating this form of behavioural disinhibition.  In a similar manner to systemic pharmacological manipulations, differing cortico-striatal circuitry appears to subserve distinct aspects of 5-CSRTT performance.  With regards to the prefrontal cortex (PFC), whereas accuracy of attentional performance appears to depend on the integrity of the prelimbic area (PrL), lesions to either the anterior cingulate cortex (ACC) or infralimbic cortex (IL) selectively increase impulsive responding (Muir et al. 1996).  Elevations in both premature and perseverative responding have also been reported after lesions or inactivation of the orbitofrontal cortex (OFC) (Chudasama and Robbins 2003).  It should also be noted that lesions of the nucleus accumbens (NAc) produced more selective impairments in impulse control, with increases in premature responding only observed on trials directly following an incorrect response (Christakou et al. 2004).   Delay discounting tasks measure impulsive choice as the selection of a smaller immediately available reward over a larger delayed one which is thought to represent intolerance to delay of gratification.  Various delay-discounting tasks have been used widely in both humans and animals (Ainslie 1975). The size of the reward and/or the length of the delays are varied in order to generate a hyperbolic discounting curve. Steeper discounting curves i.e. increased preference for smaller-sooner rewards have been repeatedly shown in subjects with GD (Alessi and Petry 2003; Dixon et al. 2003; Michalczuk et al. 2011).  7  Although a plethora of rodent delay discounting tasks have been developed (see Hamilton et al. 2015 for review), perhaps the most widely used methodology is that based on Evenden and Ryan’s original paradigm which was developed specifically to enable the efficacy of pharmacological challenges to be assessed, and therefore incorporates standard within-session shifts in delay (Evenden and Ryan 1996).  This method has the advantage of simplicity (see Cardinal et al. 2002a for potential confounds present in adjusting delay/amount tasks), although it may not allow for a true assessment of delay-sensitivity independent of subjective reward evaluation (see Ho et al. 1999 for multiplicative hyperbolic discounting model discussion).  In the Evenden and Ryan model, the animal chooses between a small reward (typically 1 sugar pellet) delivered immediately, or a larger reward (typically 4 pellets) that is delivered after a delay.  This delay increases in a step-wise session across trial blocks, for example from 0 to 10, 20, 40 60 s.  All trials are of equivalent length, such that selection of the larger reward always results in the most reward at any point in the session. Similar to premature responding on the 5-CSRTT, impulsive choice on delay discounting tasks appears sensitive to pharmacological agents that potentiate the effects of dopamine.  Yet, in contrast to the increased premature responding reliably observed on the 5-CSRTT, most papers report that administration of amphetamine decreases impulsive choice on delay discounting tasks, increasing choice of the larger but delayed reward (see Winstanley 2011 for discussion). This increase can be blocked with prior administration of a D2 but not a D1 antagonist – moreover, the effects of amphetamine are mimicked by GBR12909 but not the norepinephrine reuptake inhibitor desipramine (van Gaalen et al. 2006b).  Although broadly similar, there are differences in the cortico-striatal circuits that mediate behaviour on the 5-CSRTT and delay discounting.  For instance, ACC lesions increase motor 8  impulsivity on the 5-CSRTT, but do not affect impulsive decision making on delay discounting (Cardinal et al. 2001). Likewise, excitotoxic lesions to the OFC increase perseverative and premature responding on the 5-CSRTT (Chudasama et al. 2004), but can produce bidirectional effects on impulsive decision making on delay discounting (Winstanley et al. 2004a), dependent upon task contingencies and subjective baseline behaviour (see Floresco et al. 2008; and Zeeb et al. 2010 for discussion). In contrast to these differences, both the 5-CSRTT and delay discounting are affected by lesions to the PrL cortex, whereby animals are less able to respond appropriately to task contingencies but are not specifically more or less impulsive (Cardinal et al. 2001; Muir et al. 1996). Similarly, impulsivity on both tasks is increased following lesions to the NAc (Cardinal et al. 2001; Christakou et al. 2004) and ventral hippocampus (Abela and Chudasama 2013; Abela et al. 2013). 1.2.3 Decision making deficits. Problem gambling could be conceptualized as increased risky financial decision making. Indeed, human gamblers consistently demonstrate impaired performance on tasks measuring cost/benefit decision making under both risk and ambiguity, such as the Cambridge gambling task (Lawrence et al. 2009), game of dice task (Brand et al. 2005) and the Iowa Gambling Task (IGT) (Cavedini et al. 2002); deficits that cannot exclusively be accounted for by increased impulsivity or deficits in cognitive ability (Cavedini et al. 2002).  Amongst the laboratory tasks used to probe aberrant decision making, the IGT has been the most widely used and several rodent analogues have been developed.  The IGT requires participants to choose cards from four decks in order to accumulate points.  Two decks are advantageous, associated with smaller immediate gains but also smaller losses. In contrast, the two disadvantages decks are associated with larger gains but also disproportionately larger long-term losses.  Ergo, the optimal strategy is to avoid the tempting “high-risk high-reward” options and 9  instead choose from the less risky decks and steadily accrue smaller amounts (Bechara et al. 1994). Persistent choice of the former, disadvantageous decks is commonly observed in both gambling and substance use disorder (Cavedini et al. 2002; Power et al. 2012; Roca et al. 2008), further indicating that addictive disorders may share a common etiology.  Yet, whether aberrant decision making on these tasks arises subsequently or antecedent to chronic exposure to gambling games is unclear. Animal models of such decision making may help in elucidating vulnerabilities to such cognitive perturbations, and the relationship of these deficits to other areas of dysfunction such as impulsivity.  Although multiple versions of a rodent IGT have been developed with differing strengths and weaknesses (see de Visser et al. 2011 for discussion), the rat gambling task (rGT) has been the most widely adopted  (Zeeb et al. 2009) The rGT, consistent with the IGT, requires animals to choose between four options with established contingencies. Again, two options are disadvantageous - associated with larger gains (food reward) but more frequent and larger punishments (time-out periods), whereas the other two options are advantageous – associated with smaller gains but smaller and less frequent punishments. Animals have 30-minutes to maximise their ‘earnings’, therefore these time-out periods reduce the opportunity to earn reward. Thus, analogous to the human version, animals must learn to choose the low-reward low-punishment ‘decks’ more often than the superficially alluring but ultimately disadvantageous “high risk, high reward” ones (Zeeb et al. 2009). Results with the rGT demonstrate that animals show a very similar pattern of behavioural responses to humans playing the IGT, with selection of the tempting high-risk options declining with increased experience with the contingencies, resulting in a clear preference for options linked to smaller but safer rewards (Zeeb et al. 2009).  Optimal decision making on the rGT is modulated by multiple pharmacological systems.  Systemic administration of amphetamine or 10  the 5-HT1A receptor agonist 8-OH-DPAT both resulted in similar impairments, with animals selecting the optimal options less following administration of either compound.  In contrast, the dopamine D2-like receptor antagonist eticlopride improved optimal choice behaviour (Zeeb et al. 2009).  Despite the similar effects of 8-OH-DPAT and amphetamine, subsequent investigations have indicated that the effects of amphetamine on choice do not appear to be due to its actions at a single monoamine transmitter, but rather the additive effects on multiple systems. Systemic administration of reuptake inhibitors for 5-HT, dopamine or norepinephrine in isolation produces only mild effects on task performance. In contrast, any combination of two reuptake inhibitors significantly impaired behaviour on the rGT in a similar manner to that observed following systemic amphetamine (Baarendse et al. 2013).  Interestingly, amphetamine’s effects on choice behaviour appear to be somewhat dissociable from those on motor impulsivity.  The rGT, in a similar manner to the 5-CSRTT, also measures premature responding.  Consistent with pharmacological data from the 5-CSRTT discussed above, motor impulsivity is increased following administration of amphetamine, an effect which appears contingent on increasing extra-synaptic levels of dopamine, but not 5-HT or norepinephrine (Baarendse et al. 2013; Zeeb et al. 2013).  These data imply that risky decision making and impulsivity are somewhat dissociable. However, a recent meta-analysis revealed that at a population level, motor impulsivity on the rGT and choice of the risky disadvantageous options are well correlated. Thus although these two constructs may be pharmacologically dissociable, indicating discrete traits, they may have synergistic effects on one another when both are present to a greater degree (see Barrus et al. 2015 for discussion).  Given that high impulsivity and risky decision-making coalesce in clinical disorders, the presence of such a population-level correlation supports the 11  study of poor decision-making and high impulsivity in healthy, non-clinical populations as endophenotypes for vulnerability to impulse control and addiction disorders. Human performance on the IGT has been demonstrated to rely on the integrity of brain regions that underlie the formation and maintenance of addictive behaviours, particularly the ventromedial prefrontal cortex (VMPFC) and amygdala (Bechara 2005; Bechara et al. 1994; Bechara et al. 1999; Jentsch and Taylor 1999).  Hypoactivity of prefrontal regions is one of the most commonly reported findings from imaging studies of subjects with GD   (Balodis et al. 2012b; Potenza et al. 2003; Reuter et al. 2005), and the decision-making profile of problem gamblers on the IGT is characteristic of those with focal lesions to prefrontal cortical regions (Baarendse et al. 2013; Bechara et al. 2001; Cavedini et al. 2002).  Similarly, animals performing the rGT show impairments following inactivations to the basolateral amygdala (BLA) or OFC, and disconnection of these two areas also retards learning of the optimal task strategy (Zeeb and Winstanley 2011; 2013).  Lesions to the PFC can also reduce preference for the advantageous options on the rGT (Paine et al. 2013). Lastly, recent work has highlighted a key role for the agranular insula and medial PFC in mediating choice behaviour on the rGT, as either inactivations or lesions to these regions decreased optimal decision making by promoting the most risk-averse strategy (Pushparaj et al. 2015b; Zeeb et al. 2015).  These data are congruent with the notion that increased insula activity may facilitate detrimental gambling-related decision making, as patients with focal lesions to the insula cortex appear immune to  various gambling-related cognitions (Clark et al. 2014).  In summary there are several well-validated tasks with which to investigate neural loci and pharmacology that mediate perturbations in impulsive decision-making/action and cost-benefit decision making.  Interestingly, despite some superficial differences, the tasks discussed herein 12  tend to implicate common neural substrates and neuromodulatory systems in mediating gambling-like behaviour- specifically, the PFC and dopamine.  Poor decision-making and high impulsivity may therefore increase susceptibility to GD via common neurobiological mechanisms. In contrast to the large bodies of work on impulsive- and cost/benefit decision-making, there is a relative dearth of empirical examinations aiming to elucidate the neurobiological basis of cognitive biases and their supposed role in the manifestation and maintenance of problem gambling.   1.3 The role of cognitive biases and distortions in gambling  Cognitive biases can roughly be grouped into three main categories: personal skill or knowledge, interpretative biases and personal control (Raylu and Oei 2004). Distortions of personal skill can include superstitious or illusory ideas, such as a ‘lucky shirt’, or irrational behaviours that belie the sense that the player can influence the occurrence of random events, e.g. craps players who throw the dice harder when trying to roll a higher number (Henslin 1967). Interpretive biases refer to gamblers tendency to attribute their own successes to skill and failures to luck (Delfabbro 2004).  Personal control pertains to a gamblers belief that they can predict outcomes based on salient features of the game, for instance believing that a win is due following a series of losses (Terrell 1994).  Near-misses (unsuccessful outcomes that are structurally proximal to a win) have been found to foster the belief of personal control and potentiate the illusion that a win is imminent (Reid 1986).  The strong association of cognitive biases and distortions with the presentation of GD has led to the theory that distorted beliefs play a causative role in establishing and maintaining problematic gambling (Ladouceur et al. 1988; Toneatto et al. 1997).  Indeed, some of the most widely used psychometric measures such as the 13  South-Oaks Gambling Screen (SOGS) (Lesieur and Blume 1987), the Gambling-Related Cognitions Scale (GRCS) (Raylu and Oei 2004) and the Gambling Beliefs Questionnaire (Steenbergh et al. 2002) rely on the strength of cognitive distortions in order to asses problematic gambling.  The assumption of causality is primarily based on the increased presence or magnitude of irrational beliefs as gambling severity increases.  For instance, social gamblers demonstrate greater endorsement of irrational beliefs in relation to gambling than non-gamblers (Kallmen et al. 2008).  In turn, pathological gamblers endorse more irrational beliefs then social gamblers (Joukhador et al. 2003).  Additionally, behavioural treatments aimed at correcting these irrational beliefs are somewhat efficacious in promoting gambling cessation (Gonzalez-Ibanez et al. 2005; Ladouceur et al. 2001; Leung and Cottler 2009) and the persistence of beliefs following treatment has been linked to a higher likelihood of relapse (Oei and Gordon 2008).  Collectively, these data imply that cognitive biases are associated with gambling acquisition, maintenance and severity.   Amongst these biases, the near-miss effect has arguably garnered the most attention.  Near-misses are present throughout multiple forms of gambling, but are particularly prominent in electronic gambling machines (EGM’s) such as slot machines.  Outcomes in these games are dictated by chance, yet the presence of near-misses can foster beliefs of mastery and make participants believe they are ‘getting the hang of the game’(Walker 1992). Indeed, the presence of near-misses in slot machines may account for the particular virulence with which this sort of game has taken hold.  For instance, the majority of individuals seeking treatment for GD report slot machines as their primary game of choice (Breen and Zimmerman 2002) and these individuals have the shortest latency between commencement of gambling and the development of problematic gaming (Breen and Zimmerman 2002; Choliz 2010; Dowling et al. 2005).   Near-14  misses in slot machines have been so pervasively linked to the manifestation of problem gambling that the disproportionate presence of these types of trials has led to legal proceedings against the machines’ manufactures (Harrigan 2007).  Laboratory-based studies support the conclusion that near-misses can facilitate game involvement.  When participants were allowed to play a three wheel slot machine for money, the presence of near-miss trials increased the duration of time subjects were willing to play relative to a condition in which near-misses were absent (Cote et al. 2003).  Additionally, the persistence exhibited by some participants in response to near-misses correlates with their endorsement of gambling-related irrational beliefs (Billieux et al. 2012).  Near-misses appear to facilitate continued game engagement through generating the expectation of imminent reward. Imaging studies have consistently demonstrated that near-misses generate a pattern of neural activation that is similar to that seen during win trials (Chase and Clark 2010; Clark et al. 2009). Put more simply, near-misses may evoke a representation of reward, even in the absence of any actual reward. The ability of near miss trials to generate this reward expectancy is potentiated in gamblers in comparison to controls (Habib and Dixon 2010), and the degree to which near-misses elicit activation within these areas has been correlated with gambling severity (Chase and Clark 2010; van Holst et al. 2014). The areas activated by near-misses include the striatum, insula and ACC, all areas that are heavily innervated by dopamine (Gaspar et al. 1989).  Dysfunction within the dopaminergic system has been canonically linked with substance addiction (Koob and Volkow 2010; Volkow 2005).  Given the reputed similarities between substance and behavioural addictions, it is likely that dopamine may also be involved in GD.  Dopamine neurons within the mid-brain reliably fire in response to primary rewards (Schultz and Romo 1990).  Importantly, if the occurrence of these rewards is predicted by an environmental 15  cue, the phasic firing of the dopamine neurons shifts towards the conditioned stimulus that predicts the reward, rather than the reward itself (see Schultz 1998 for review).  The phasic activation of dopamine neurons in response to such stimuli varies according to the likelihood of reward (Fiorillo et al. 2003).  In regards to slot machines, it could be argued that near-misses are just such a conditioned stimulus; due to their structural similarity to a win these trials may evoke positive prediction errors.  In support of a role for dopamine transmission in potentially mediating dysfunctional slot machine play, Zack and Poulous (2007) noted that administration of a low dose of the D2 receptor antagonist haloperidol increased the subjective enjoyment of pathological gamblers, but not control subjects, playing a slot machine (Zack and Poulos 2007).  Subsequent studies also showed that this enjoyment facilitated increased and continued betting (Tremblay et al. 2011). The dose used in the above study was relatively low and therefore assumed to increase dopaminergic activity via inhibition of autoreceptors (Frank and O'Reilly 2006; Zack and Poulos 2007).  Thus, augmented dopaminergic activity appears to facilitate maladaptive gambling behaviour, a theory that is bolstered by reports of iatrogenic gambling following therapeutic administration of dopamine agonists (Dodd et al. 2005).  This particularly compulsive form of gambling can arise de-novo following treatment with agonists with a high affinity for the dopamine D2-like receptor family (Voon et al. 2009).  Although most often reported in patients with Parkinson’s Disease (PD), compulsive behavioural disorders, such as GD, have also been reported in patients with restless leg syndrome (RLS), fibromyalgia and prolactinoma, following treatment with dopamine agonists (Clark and Dagher 2014; Voon et al. 2007a; Weintraub and Potenza 2006). These findings indicate a key role for dopamine dysfunction in the pathogenesis of GD.  However, idiopathic GD patients do not appear to exhibit gross alterations in dopamine 16  receptor density in comparison to controls (Boileau et al. 2013) and imaging studies have shown inconsistent results in regards to hyper- or hypo-activation of dopamine rich regions (Balodis et al. 2012b; van Holst et al. 2012) In sum, subjective responses to near-misses may confer vulnerability to the commencement and maintenance of problematic gambling.  Near-misses induce activity in reward related areas in gamblers and dopaminergic agents potentiate these responses.  Thus if near-misses are acting as incentive conditioned stimuli, promoting dopamine release as if they are predictive of reward delivery (Robinson and Berridge 1993), it may explain the particular potency they exhibit in prolonging gambling. Still, this remains to be confirmed, as does the specific role of dopamine.  As mentioned above, few animal paradigms to date have successfully modelled cognitive biases in rodents. Whereas many of the biases covered by the GRCS and SOGS questionnaires are difficult to model in animals, rats are capable of quite complex conditional discrimination tasks, and as such could be susceptible to a near-miss effect. To that end, we developed a rodent slot machine task (rSMT) that suggests rats share key behavioural features with human gamblers (Winstanley et al. 2011)  1.4 The rSMT and experimental objectives The rat slot machine task (rSMT) was designed in order to provide a facsimile of slot machine play in rodents (Winstanley et al. 2011) (task schematic figure 2.2).  Each self-initiated trial within the rSMT requires the rat to respond to three flashing apertures analogous to the spinning wheels of a slot machine.  The rat has to respond to each hole in order, setting the light inside to on or off.  Once the rat has responded to these three apertures, it must choose between two levers.  One lever, the ‘cash-out’ lever, delivers sugar pellets on winning trials, but a time-17  out punishment on losing outcomes, whereas the other, the ‘roll’ lever, allows the animal to begin a new trial straight away. The rats’ optimal strategy is dependent on the number of lights illuminated, in that three illuminated apertures is a win, any other combination of lights is a loss.  The rSMT thus provides a behavioural index of ‘reward expectancy’.  Additionally, the task provides a number of useful ancillary measures, in that it allows for the indirect measurement of motivation or task engagement via the number of trials animals complete, and also the incentive salience animals attribute to reward-related stimuli, measured via the latency to respond at the apertures based on the status of the preceding hole.   The first experiment using this task (completed by the candidate whilst working as a technician in Dr. Winstanley’s lab) not only validated the task but also provided initial pharmacological data demonstrating a crucial role for the dopamine D2-like receptor family in mediating reward expectancy during gambling-like decision making.  The experiments presented in this thesis represent my work using the rSMT to probe the neurobiological mechanism underlying the near-miss effect and slot machine gambling.  More specifically, the aims of the six experiments within this dissertation are as follows: Experiment 1 (chapter 3) examines the effects of systemic administration of pharmacological agents with high selectivity for the dopamine D2-like receptor subtypes on rSMT performance, it shows a novel role for the dopamine D4 receptor in mediating attributions of salience to reward-related stimuli.  Experiment 2 (chapter 4) explores the role of D4 receptor stimulation in modulating the incentive motivational properties of cues on simple behavioural tasks, in order to ascertain if augmentation of these cue-driven behaviours could contribute to the effects of D4 receptor 18  agonists in the rSMT. Specifically, I examined the effects of a selective D4 agonist on autoshaping (sign-tracking) and conditioned reinforcement.  Experiment 3 (chapter 5) investigates the role of the anterior cingulate cortex on rSMT performance via temporary inactivations. Secondly, this study also assessed any selective role for D4 receptors within the ACC in contributing to rSMT performance.  Experiment 4 (chapter 6) in a similar manner to experiment 3 temporary inactivations were used to delineate the role of the granular and agranular insular cortex in rSMT performance. The study provides novel insight into differential roles for the insula sub-regions in gambling-related decision making. Again, we also investigated the contribution of D4 receptors to via local infusion of a D4 agonist.  Experiment 5 (chapter 7) presents a model of ‘problem gambling’ using chronic administration of the D2–like agonist ropinirole during rSMT play. This study also utilized molecular biology techniques to link intra-cellular signaling changes with behavioural alterations.  Experiment 6 (chapter 8) utilizing the model of problem gambling presented in chapter 7 and 6-OHDA lesions of the dorsal striatum we examine the role of dopamine replacement therapies in the etiology of impulse compulsive disorders in Parkinson’s disease. Additionally, we assessed the therapeutic potential of lithium chloride and the β-adrenoreceptor blocker propranolol in ameliorating compulsive gambling.   19  Chapter 2: General Methods   2.1 Subjects Subjects were male Long Evans rats (Charles River Laboratories, St. Constant, Canada) weighing 275-300g at the start of testing.  Subjects were food restricted to 85% of their free feeding weight and maintained on 14g rat chow given daily.  Water was available ad libitum.  All animals were pair-housed in a climate-controlled colony room maintained at 21˚C on a reverse 12hour light-dark schedule (lights off 8 am).  The testing and housing were in accordance with the Canadian Council of Animal Care and all experimental protocols were approved by the Animal Care Committee of the University of British Columbia.    2.2 Behavioural apparatus  Behavioural testing took place in eight standard five-hole operant chambers (figure 2.1), each enclosed within a ventilated sound-attenuating cabinet (Med Associates Inc, Vermont). Each chamber was fitted with an array of five response holes positioned 2 cm above a bar floor. A stimulus light was set at the back of each hole. Nose-poke responses into these apertures were detected by a horizontal infrared beam. A food magazine, also equipped with an infrared beam and a tray light, was located in the middle of the opposite wall, and sucrose pellets (45 mg; Bioserv, NJ, USA) could be delivered into it from an external pellet dispenser. Two retractable levers were located on either side of the food tray and chambers could be illuminated using a house light located above the food tray.   20  2.3 rSMT: Habituation and pre-task training  All training and testing sessions lasted 30 minutes, and each subject received one session per day.  Subjects were initially habituated to the testing chambers over 2 sessions, during which time sucrose pellets were placed in the nose-poke apertures and in the food tray.  Subjects then learned to respond on one of the retractable levers to earn food reward under a fixed ratio 1 schedule.  Only one lever was presented during each session.  After the animal had made >50 lever presses in a session, training was repeated on the other lever.  The order in which levers were presented (left/ right) was counterbalanced between subjects.    2.4 SMT: Task training  Training stage 1: In this training stage, rats learned to start each trial by pressing on the right (roll) lever and to respond to a flashing light in the array in order to earn reward.  At the start of the task, the roll lever was inserted into the operant chamber.  A response on the roll lever triggered the light inside hole 2 to flash at a frequency of 2Hz.  A nose-poke response in the aperture caused the light inside to set to on and a 20kHz tone to sound.  The tone acted as an additional cue indicating the status of the aperture.  A single pellet was then dispensed into the magazine and the magazine light illuminated.  Responding in the magazine for the food pellet extinguished the light and extended the roll lever back into the chamber to allow the rat to start another trial.  Animals were moved onto the next stage after meeting criteria of >50 trials completed within the 30-minute session.   Training stage 2: In this training stage, rats learned that the set status of hole 2 (on or off, summarized henceforth as “1” or “0”) determined whether a response on the collect lever was rewarded or punished.  Trials were initiated in the same manner and animals were again required 21  to nose-poke into hole 2.  Half of the trials were identical to those in the previous session, such that this response caused the light to set to on and a 20kHZ tone to sound.  These trials were designated as “win” trials.  The other 50% of trials were “loss” trials, in which the light inside hole 2 turned off following a nose-poke response and a 12Khz tone sounded.  Following a response in hole 2, either the ‘roll’ lever and/or left ‘collect’ lever were presented.  Once a lever-press was made, both levers retracted.  Regardless of the trial type (win or loss), responding on the roll lever forfeited the current trial and a new trial began straight away with the light flashing in hole 2.  The consequence of pressing the collect lever was dependent on the trial type.  On win trials, pressing the collect lever resulted in the delivery of two food pellets, after which the roll lever was again extended into the chamber allowing the animal to begin a new trial at will.  In contrast, pressing the collect lever on loss trials was punished by a 10-second time out period during which no reward could be earned or new trials initiated.  Following the time out period, the roll lever was extended into the chamber and the rat could start a new trial.  For the first 5 sessions, loss trials were forced choice to ensure animals were exposed to both contingencies, after which both trial types ended in presentation of both levers.  Training continued until rats responded on the collect lever on ≥80% of win trials and ≤ 20% of loss trials (10-20 sessions). Training stage 3: - In this training stage, another active hole is added such that a response is required in holes 2 and 3 before levers are presented.  The trial structure is identical to stage 2, except that after the tone had sounded following a response in hole 2, hole 3 began to flash.  The animal was required to make a nosepoke response in this aperture, after which the light set to on or off and the appropriate tone sounded.  Both the collect and roll levers were then presented.  In this stage, a win was signaled if both lights were set to on ([1,1]) , and a response on the collect lever on such win trials resulted in delivery of 5 sugar pellets.  All other trial types ([1,0]; [0,1]; 22  [0,0]) were classed as loss trials, and responding on the collect lever resulted in a 10 s time out penalty.  The incidence of the different trial types was distributed evenly throughout the session such that each trial type occurred at least once every 4 trials and not more than twice in every 8 trials.  The exact sequence of trials was randomized within these constraints.  Training continued for 5 sessions, after which animals were moved on to the full task.   2.5 The rodent Slot Machine Task (rSMT) A task schematic is provided in figure 2.2. The middle three holes within the five-hole array were used in the task (holes 2–4). The rat initiated each trial by pressing the roll lever. This lever then retracted and the light inside hole 2 began to flash at a frequency of 2 Hz (Figure 2.2a). Once, the rat made a nosepoke response at this aperture, the light inside set to on or off for the remainder of the trial. Depending on the illumination status of the light, either a 20 kHz (‘on') or 12 kHZ (‘off') tone sounded for 1 s, after which the light in hole 3 began to flash (Figure 2.2b). Again, a nosepoke response caused the light to set to on or off and triggered the presentation of a 1 s 20/12 kHZ tone, after which the light in hole 4 began to flash (Figure 2.2c). Once the rat had responded in hole 4 and the light inside set to on or off, again accompanied by the relevant tone, both the collect and roll levers were presented. The rat was then required to respond on one or other lever, and the optimum choice was indicated by the illumination status of the lights in holes 2–4. On win trials, all three lights were set to on (1,1,1), and a response on the collect lever delivered 10 sugar pellets (Figure 2.2d). If any of the lights had set to off (e.g. Figure 2.2e), then a response on the collect lever lead to a 10 s time-out period during which reward could not be earned. The use of three active holes resulted in eight possible trial types (Figure 2.2f, [1,1,1]; [1,1,0]; [1,0,1]; [0,1,1]; [1,0,0]; [0,1,0]; [0,0,1]; [0,0,0]), the incidence of which was pseudo-23  randomly distributed evenly throughout the session on a variable ratio 8 schedule. If the rat chose the roll lever on any trial, then the potential reward or time-out was canceled, and a new trial began. Hence, on win trials, the optimal strategy was to respond on the collect lever to obtain the scheduled reward, whereas on loss trials, the optimal strategy was to instead respond on the roll lever and start a new trial. If the rat chose to collect, both levers retracted until the end of the reward delivery/time-out period, after which the roll lever was presented and the rat could initiate the next trial. The task was entirely self-paced in that animals were not required to make any of the responses within a particular time window; if necessary, the program would continue to wait for the animal to make the next valid response in the sequence until the end of the session. The only point at which the rat could fail to complete a trial was therefore if the session ended partway through. Animals received 5-6 daily testing sessions per week until asymptotic statistically stable patterns of responding had been established over five sessions.   2.6 Microinfusions (chapters 5 and 6) Once a stable post-operative baseline was re-established following surgery animals were habituated to the infusion procedure with two mock infusions. The procedure for the mock infusions was the same as detailed below, with the exception that no drug or vehicle was delivered.  All infusions were performed over a 3-day cycle, starting with a baseline session. Subsequently animals received a test day where either drug or vehicle was infused. On the last day animals remained in their home cages and no testing was performed. During infusions animals were gently restrained, obturators were removed and 30-gauge injectors (extending 1mm beyond the guide) were inserted into the guides. A volume of 0.5µl/side was infused using a 24  duel-channel infusion pump (Harvard Apparatus, Holliston, USA) at a rate of 0.25µl minute.  Injectors were left in place for one additional minute to allow for diffusion. Injectors were then removed, obturators reinserted and the animals returned to their home cages for 10-minutes before behavioural testing commenced.   2.7 Osmotic pump implantation (chapters 7 and 8) Animals were implanted with model 2ML4 osmotic mini-pumps (Alzet, DURECT corporation, Cupertino, CA) delivering either 5mg/kg/day ropinirole hydrochloride (Tocris, R&D systems, MN) or 0.9% saline solution for 28 days. Doses were based on previous reports (Iida et al. 1999; Matsukawa et al. 2007; Millan et al. 2004; Rogers et al. 2000). The dose chosen is comparable with single daily dose of the prolonged release formulation of ropinirole used in human patients (Nashatizadeh et al. 2009).  The osmotic pumps were sterilely filled with solution for each rat a day prior to surgery and kept overnight in a sterile 50ml falcon tube containing 0.9% saline solution. Calculations for filling the pumps were based on the Alzet guide. Animals were anesthetised using 2% isoflurane delivered in O2 and monitored continuously throughout surgery. Animals were administered anafen and buprenorphine for systemic and local analgesia, respectively. During surgery a small incision was made between the rats’ shoulder blades, a ‘pocket’ formed under the skin on the animals back using sterile haemostats and the osmotic pump slide into this space. Animals were allowed to recover in their home cage for 1-day before testing resumed. During this time water was available ad libitum and animals were fed 20g rat chow per day. Post-surgical monitoring of the animals continued for 10 days. Procedures for osmotic mini-pump removal were similar to those followed for implantation.  25  2.8 Behavioural measures for the rSMT The following variables were analysed for each trial type: the percentage of trials on which animals pressed the collect lever, the average latency to respond on the collect lever, and the latency to respond in each aperture when the light inside was flashing. The number of trials completed per session was also analysed. The latency to choose the roll lever after each trial was not included in the formal analysis as this measure was skewed by the higher incidence of erroneous collect responses, resulting in a 10 s time penalty, on some trial types, and the time taken to consume sugar pellets on win trials.   2.9 Statistical analysis  All data were subjected to within-subjects repeated measures analysis of variance (ANOVAs), conducted using SPSS software (SPSS v16-23, Chicago, IL) and were applicable were subjected to an arcsine transformation to limit the impact of an artificial ceiling (i.e. 100%). During training, the collect lever choice and collect lever latency were analysed with session (five levels) and trial type (eight levels) as within-subjects factors. A stable baseline was defined as the lack of a significant effect of session or trial type × session interaction. To determine the impact of the number of lights illuminated, regardless of spatial position, data were pooled across 2-light trials ([1,1,0], [1,0,1], and [0,1,1]) and one-light trials ([1,0,0], [0,1,0], and [0,0,1]). ANOVA’s were then performed with session and lights illuminated (four levels, 0–3) as within-subjects factors. In order to determine if illuminated apertures temporally closer to a decision point altered animals’ behaviours, analyses were conducted to test if the last light setting to on increased animals erroneous lever choice (e.g. [1,1,0], [1,0,0], [0,1,0] vs [0,1,1], [0,0,1], [1,0,1]). 26  Follow up analyses were conducted only when there was a main effect of drug. The average latency to respond on the collect lever on each trial type was analysed using similarly structured ANOVA’s. In order to determine whether responding on the next hole was affected by the illumination of the previous hole, the average latency to respond in the middle hole if the first hole had set to on or off was calculated, regardless of trial type. Likewise, the average latency to respond in the last hole if the middle hole had set to on or off was determined. These data were then subjected to ANOVA with session, hole (two levels: middle and last) and previous hole state (two levels: on and off) as within-subjects factors. Trials completed per session were analysed by a simple ANOVA with session as the only within-subjects factor. The response to the different pharmacological challenges was analysed using similar ANOVA methods, but the session factor was replaced with a dose factor. The significance level for all effects was p ≤0.05.  Analyses for which p ≤ 0.1 were described as trends. All data are presented as mean ± standard error of the mean (SEM).    27    Figure 2.1 The standard five-hole operant chamber used for the rSMT.  (A) Side view of the chamber, with five-hole stimulus array on the left and food tray on the right and external pellet dispenser that delivers sugar pellets. (B) The five-hole stimulus array.   28   29  Figure 2.2 Schematic diagram showing the trial structure of the rSMT. All testing took place in 5-hole operant chambers, of which the three middle holes (holes 2-4) were used for this task. (a) Animals initiated each trial by responding on the "roll", lever. This lever retracted and the light inside hole 2 began to flash.  Once the rat responded at this aperture, the light inside set to on or off for the remainder of the trial and either a 20kHz (light on) or 12 kHZ (light off) tone sounded for 1 s, after which the light in hole 3 began to flash. (b) Again, a nosepoke response resulted in the light setting to on or off and the sounding of the 20kHZ/ 12kHZ tone, after which the light in hole 4 started to flash. (c) Once the rat responded in hole 4 and the light inside set to on or off, again accompanied by the relevant tone, both the "collect" and roll levers were presented.  The rat was then required to respond on one of the levers; the optimum choice was determined by the pattern of lights in holes 2-4. (d) On win trials, all three lights were set to on [1,1,1], and a response on the collect lever lead to delivery of 10 sugar pellets. (e) If any of the lights had set to off (i.e. a ‘loss’ trial), a response on the collect lever lead to a 10 second time-out period, during which reward could not be earned.  If the rat chose the roll lever on any trial type, then the collect lever retracted, the potential reward or time-out was cancelled, and a new trial began.  Hence, on win trials, the optimal strategy was to respond on the collect lever to obtain the scheduled reward, whereas on loss trials, the optimal strategy was to instead respond on the collect lever and start a new trial.  If the rat chose to collect, both the collect and roll levers retracted until the end of the reward delivery/time-out period, after which the roll lever was presented and the rat could initiate the next trial. (f) There were 8 possible trial types.  Figure is modified from (Winstanley et al. 2011) 30  Chapter 3: Experiment 1: A selective role for dopamine D4 receptors in modulating reward expectancy in a rodent slot machine task.    3.1 Introduction Cognitive accounts of gambling propose that gambling is sustained because of the erroneous or distorted beliefs about the independence of gambling outcomes, the intervention of luck, and the ability of personal skills to confer success when gambling (Ladouceur et al, 1988; Toneatto et al, 1997). One prominent hypothesis is that the experience of almost-winning—a so-called ‘near-miss'—can invigorate gambling activity, and may accelerate the development of DG in vulnerable individuals (Reid, 1986; Griffiths, 1991; Clark, 2010). Near-miss events can produce similar psychological and physiological changes as winning outcomes (Griffiths, 1991). Near-misses may therefore heighten reward expectancy due to their similarity to wins, making continued play more likely (Reid, 1986). In line with this theory, near-misses have been shown to increase the desire to continue gambling (Kassinove and Schare, 2001; Cote et al, 2003;) and to enhance neural activity within the mid-brain and the ventral striatum (Clark et al, 2009; Habib and Dixon, 2010). These observations suggest that near-misses convey a positive reward signal encoded by the dopaminergic circuits that support reward expectancy and reinforcement learning (Schultz et al, 1997;Schultz, 1998; Fiorillo et al, 2003). In support of this general hypothesis, drugs that alter dopaminergic activity have been shown to modify slot-machine play, a form of gambling in which near-misses are particularly salient. The psychostimulant amphetamine, which potentiates dopamine's (DA) actions, can 31  increase the motivation to play slot machines (Zack and Poulos, 2004), whereas the preferential D2 receptor antagonist, haloperidol, can enhance the rewarding properties of such behaviour (Zack and Poulos, 2007). Aberrant dopamine signaling is a critical component of drug addiction, and drives the increased incentive salience of drug-paired cues that galvanize drug seeking (Robinson and Berridge, 1993). The observation that slot machine play is often the most common gambling activity in pathological gamblers has lead to the suggestion that slot machine gambling may be particularly compulsive (Breen and Zimmerman, 2002; Choliz, 2010). Given that animal research has significantly advanced our understanding of goal-directed behaviour and addiction, an animal model of slot machine play may make a valuable contribution to gambling research (Potenza, 2009), and a preliminary reports indicates that rats are capable of learning such a task (Winstanley et al. 2011). Using the rSMT, we previously reported that rats are also susceptible to putative win signals in non-winning trials, akin to a near-miss effect (Winstanley et al. 2011).  In addition, the erroneous expectation of reward following such a loss can be increased by administration of the dopamine D2-like receptor agonist quinpirole (Winstanley et al. 2011).  Numerous studies suggest an important role for the D2 receptor family in determining vulnerability to dependency (Blum et al. 1996; Comings et al. 1996; Noble 2000).  Furthermore, this receptor class plays a pivotal role in the incentive salience theory of addiction, which proposes that environmental stimuli previously paired with drugs or rewards can develop considerable influence over behaviour (Flagel et al. 2010; Heinz 2004; Robinson and Berridge 1993).  Given that the near-miss effect could arguably reflect the misattribution of incentive salience to a seemingly reward-related stimulus, it is perhaps unsurprising that a D2-like receptor agonist increased the misinterpretation of near-misses as wins in our rodent model. 32   However, it is unclear which D2 receptor subtype was critically involved in enhancing the erroneous expectation of reward in the rSMT.  The majority of studies that target D2-like receptors often attribute their findings to the D2 receptor itself, potentially due to its relative abundance within the D2 family (Marsden 2006) and its localisation within reward-related neural structures such as the dorsal striatum and NAc (Jaber et al. 1996). However, the D2 receptor class also contains D3 and D4 receptors, both of which are affected by drugs such as quinpirole (Caine and Koob 1993; Chemel et al. 2006) , and may play an important role in addictive and impulsive behaviours.  D3 receptors are co-localised with D2 receptors in limbic areas known to be critical for the reinforcing properties of addictive drugs (Heidbreder et al. 2005) leading to speculation that D3 antagonism may be a promising treatment for addiction (Heidbreder and Newman 2010; Le Foll et al. 2005).  Indeed, D3 antagonists can attenuate cocaine- and nicotine-induced conditioned place preference in rats (Le Foll et al. 2005; Xi et al. 2004) and reduce drug seeking behaviours (Higley et al. 2011; Xi et al. 2004).  Given that the effects of D3 antagonists are most pronounced when drug self-administration depends on conditioned cues, it has been postulated that D3 receptors may play an important role in the attribution of incentive salience (Beninger and Banasikowski 2008; Khaled et al. 2010; Le Foll et al. 2005) and thus may contribute to the near-miss effect.    In contrast, D4 receptors are primarily located within frontal cortical regions (Van Craenenbroeck et al. 2010), and consequently represent a potential target for modulating higher-order cognitive processes (Van Tol et al. 1991).  D4 receptor polymorphisms are associated with a wide range of psychiatric disorders that have impulsivity or thought disturbances as a key component, such as schizophrenia, attention deficit hyperactivity disorder (ADHD), substance abuse and PG (Comings et al. 2001; Comings et al. 1999; Demiralp et al. 2007; Tarazi and 33  Baldessarini 1999).  However, clinical trials of selective D4 agents as neuroleptics have not been encouraging (Kramer et al. 1997; Tarazi and Baldessarini 1999), and animal studies investigating the behavioural effects of D4 receptor manipulations have yielded mixed results (Le Foll et al. 2009; Oak et al. 2000).  Still, evidence is emerging to suggest that D4 receptors play a critical role in attributing emotional salience to environmental stimuli and guiding response to these cues (Lauzon et al. 2009; Lauzon and Laviolette 2010; Yan et al. 2012).  The following pharmacological experiments using the rSMT were therefore performed to determine whether D2, D3 or D4 receptors are critically involved in the expression of the near-miss effect.  3.2 Additional methods Pharmacological challenges Once stable baseline behaviour had been established, rats were separated into two cohorts matched for task performance.  The effects of the following compounds were assessed in group 1:  the selective D2 receptor antagonist L-741,626 (0, 0.1, 0.3 and 1.0 mg/kg), the selective D3 receptor antagonist SB277011A (0, 0.5, 1.5 and 5 mg/kg) and the selective D4 receptor agonist PD169077 (0, 0.5, 1, 5, 5 and 10 mg/kg)  The effects of the following compounds were assessed in group 2:  the D2-like receptor agonist quinpirole (0, 0.0125, 0.0375 and 0.125 mg/kg) (Winstanley et al. 2011), the selective D3 receptor agonist PD128907 (0, 0.01, 0.03, 0.1 mg/kg) and the selective D4 receptor antagonist L-745,870 (0, 0.5, 1.0, 5.0 and 10.0 mg/kg).  All drugs were administered 10 minutes prior to testing, with the exception of L-745,870 (20 minutes) and SB27011A (30 minutes).  In the second series of drug challenges, L-741,626 (0.5 mg/kg), SB277011A (1 mg/kg), eticlopride hydrochloride (Sigma-Aldrich; 0.01 mg/kg), or L-745,870 (5mg/kg) were 34  administered prior to quinpirole (group; 0 & 0.0125 mg/kg).  L-741,626 and eticlopride were administered 10 minutes before the injection of quinpirole.   SB277011A was given 30 minutes prior to the second injection, whereas L-745,870 was administered 20 minutes prior.  Behavioural testing commenced 10 minutes after the injection of quinpirole.  L-741,626 and eticlopride were administered 10 minutes before the injection of quinpirole.  SB277011A was given 30 minutes prior to the second injection, whereas L-745,870 was administered 20 mins prior.  Behavioural testing commenced 10 mins after quinpirole was administered.   All drug doses were calculated as the salt and dissolved in 0.9% sterile saline, except for SB277011A, which was dissolved in a 10% solution of beta-cyclodextrin in sterile water, and L-741,626, which was dissolved in a 20% solution of ethylene glycol 400 and saline.  All drugs were prepared fresh daily and administered via the intraperitoneal route with the exception of PD128907 which was administered subcutaneously.  All drugs were given in a volume of 1 mg/ml, apart from the highest dose of SB277011A which was given in a volume of 2 mg/ml.  Quinpirole hydrochloride was purchased from Sigma-Aldrich Canada (Oakville, Canada).  L-741,626, L-745,870 trihydrochloride, PD128907, and PD168077 were purchased from Tocris Bioscience (Ellisville, MO). SB277011A was a gift from Dr Le-Foll.    3.3 Results Six animals were excluded from the analysis because they failed to meet performance criteria as defined previously (Winstanley et al. 2011): at least 50 trials completed per session and less than 50% error rate on clear loss trials ([0,0,0]).  The final number of animals included in the analysis was therefore 26, split evenly between the two groups.  For any drug condition, data were excluded from any animal that failed to complete at least 24 trials (3 of each type).  35   Baseline behaviour  In keeping with our previous findings, choice of the collect lever varied significantly across the different trial types (figure 3.1: trial type: F7,175 = 158.09, p<0.001) such that there was a positive relationship between the number of lights illuminated and preference for the collect lever (figure 3.1: lights illuminated: F3,75 = 862.31, p<0.001, 3 v 2: F1,25 = 359.17, p<0.001; 2 vs 1: F1,25 = 669.72, p<0.001; 1 vs 0: F1,25 = 100.89 p<0.001).  Although rats correctly chose the collect lever on almost 100% of win trials ([1,1,1]), and likewise switched their preference to the now optimal ‘roll’ lever on approximately 90% of clear loss trials ([0,0,0]), animals consistently preferred the collect lever on 2-light loss trials (68.54% ± 1.93 (SEM). Rats therefore responded to this class of loss trials as if these stimuli signaled that reward was available, analogous to a near-miss effect. There was no significant effect of trial type when comparing the different forms of 2-light or 1-light losses, suggesting that the exact spatial location of the lights is less important in guiding response outcomes as the number of lights set to on (trial type: 2-light trials: F2,50 =2.71, NS; 1 light trials: F2,50 = 1.74, NS).    Although collect response latency varied between trial types, there was no significant difference between win and near-miss trials (table 3.1: trial type: F7,175 = 22.9, p<0.0001; lights illuminated: 3 vs 2: F1,25 =0.14, NS).  In contrast, animals were significantly faster to make collect errors on near-loss or clear loss trials (lights illuminated: 2 vs 1: F1,25 = 33.29, p<0.0001; 1 vs 0: F1,25 = 35.9, p<0.001), indicating that such errors may result from disinhibited or impulsive responding, rather than from a lack of cognitive deliberation.  Animals were also quicker to respond at the next hole if the previous hole had set to on (table 3.2: previous hole state: F1,25 = 4.54, p=0.04), suggesting that illumination of an aperture acted as a positive 36  reinforcer and facilitated responding at the subsequent hole.  On average, animals completed 100.39 ± 2.07 trials per session (table 3.3). Effects of the D2-like receptor agonist quinpirole, the D3 receptor agonist PD129807, and the D4 receptor agonist PD168077. Replicating our previous findings using this task, quinpirole administration significantly increased erroneous collect responses at all doses (figure 3.2 a, b; dose x lights illuminated: F9,108 = 4.71, p<0.001; saline vs 0.0125 mg/kg: F3,36 = 3.24, p=0.03; saline vs 0.0375 mg/kg: F3,36 = 11.17, p<0.001 saline vs 0.125 mg/kg: F3,36 = 10.62, p<0.001). This was evident at all trial types, with the exception of wins (dose x trial type: F21,252 = 2.27, p= 0.002; dose (1,1,1): F3,36 = 0.33, NS).  The selective D3 agonist did not significantly affect choice behaviour on the rSMT (figure 3.3 a & b; dose x lights illuminated: F9,108 = 0.56, NS; dose x trial type: F21,252= 0.68, NS). The highest dose of PD 168077 significantly increased erroneous collect responses (figure 3.2 c, d; dose x trial type: F7,70 = 3.44, p=0.003; sal vs 10.0 mg/kg: F1,10 = 16.21, p=0.002).  This impairment was significant for 0-light (dose: F1,10 = 13.04, p=0.005) and 1-light trials (dose: F1,10 = 19.57, p<0.001), although there was also a trend towards impaired performance on near miss trials (dose: F1,10 = 4.40, p=0.06). There was no appreciable effect on preference for the collect lever on win trials (dose: F1,10 = 0.62, NS).   In addition to its effects on choice, all doses of quinpirole increased the latency to respond on the collect lever regardless of trial type or the number of lights illuminated (table 3.1, dose x trial type: F21,252 = 0.79, NS; dose x lights illuminated: F9,108 = 0.90, NS).  The increase in latency became more pronounced as the dose increased (sal vs 0.0125mg/kg: F1,12 = 8.64, p=0.01; sal vs 0.0375mg/kg: F1,12 = 14.74, p=0.002; sal vs 0.125mg/kg: F1,12 = 20.22, p=0.001). Similarly, quinpirole increased the latency to respond in apertures regardless of the state of the 37  previous hole (Table 3.2; dose: F3,36 = 4.4, p=0.01; dose x previous hole state: F3,36 = 0.49, NS).  Quinpirole also lead to a dose-dependent decrease in the number of trials completed (table 3.3; dose: F3,36 = 47.74, p<0.0001; sal vs 0.0125 mg/kg: F1,12 = 33.64, p<0.0001; sal vs 0.0375 mg/kg: F1,12 = 108.83, p<0.0001; sal vs 0.125mg/kg: F1,12 = 134.63, p<0.0001).  PD128907 did not affect the latency to choose the collect lever (table 3.1; dose x trial type: F21, 252 = 1.0, NS; dose x lights illuminated: F9, 108 = 0.89, NS).  There was, however, an effect of drug administration on the latency to respond at the subsequent hole based on the light status of the previous hole (table 3.2; dose x light status: F3, 36 = 2.80, p = 0.05). This effect was driven by an increase in latency to respond if the previous hole had set to off at the highest dose PD128907 (1.0 mg/kg: previous hole state: F1,12 = 6.02, p = 0.03). The highest dose of PD128907 also decreased the number of trials completed (table 3.3; dose: F 3,36 = 8.05, p<0.0001; sal vs 0.01 mg/kg: F 1,12 = 0.21, NS; sal vs 0.03 mg/kg: F 1,12 = 3.01, NS;  sal vs 0.1 mg/kg: F 1,12 = 14.81, p=0.002).  At the highest dose of the D4 agonist PD168077 (10mg/kg), two animals failed to complete at least 24 trials (3 of each trial type) and were thus excluded from any analysis involving this dose.  In addition to its effects on choice, PD168077 produced a general increase in the amount of time taken to select the collect lever, although this was only evident at the highest dose (table 3.1; dose x trial type: F28,280 = 1.05, NS; sal vs 10.0 mg/kg: F1,10 =14.22, p=0.004).  The two higher doses also decreased the number of trials completed (table 3.3, dose: F4,40 = 15.71, p<0.0001; sal vs 5.0mg/kg: F1,10 = 6.90, p=0.03; sal vs 10.0mg/kg F1,10 = F1,10 = 26.66, p<0.0001), potentially indicative of a general reduction in motor activity.  However, counter to such a conclusion, there was no effect of PD168077 on the time taken to respond in the apertures (table 3.2, dose: F4,40 = 2.42, NS; dose x previous hole status: F4,40 = 0.77, NS).  38   Effects of the D2 receptor antagonist L-741,626, the D3 receptor antagonist SB277011A and the D4 receptor antagonist L-745,870. Neither of the selective D2 or D3 antagonists significantly affected choice behaviour on the rSMT (figure 3.3a-f; L-741,626: dose x lights illuminated: F9,135 = 0.97, NS; dose x trial type: F21,252 = 0.95, NS; SB277011A: dose x lights illuminated: F9,108 = 0.47, NS; dose x trial type: F21,252= 0.85, NS).  In contrast, the D4 receptor antagonist significantly improved choice behaviour, decreasing erroneous collect responses overall (figure 3.2f; dose x trial type: F28,336 = 1.55, p=0.04).  There was an inverse relationship between the improvement in the animals’ performance and the strength of the administered dose (figure 3f; dose x trial type; sal vs 0.5 mg/kg: F7,84 = 2.52, p=0.02; sal vs 1.0 mg/kg: F7,84 = 2.36, p=0.03; sal vs 5.0 mg/kg: F7,84 = 1.88, p=0.08; sal vs 10.0 mg/kg: F7,84 =0.50, NS).  Unlike in previous analyses, grouping the trials by the number of lights illuminated diminished these effects (figure 3e; dose x lights illuminated: F3,36 = 1.57, NS).  However, grouping the trials by the spatial position of the lights indicates that the improvements in task performance seen at the lowest dose were primarily evident in trials where the last light was off ([1,1,0], [1,0,0] & [0,1,0]) rather than on ([0,0,1], [1,0,1] & [0,1,1]) (sal vs 0.5 mg/kg- last light off; dose: F1,15 = 6.21, p=0.03; -last light on: F1,15 =2.28, NS).  Hence, the D4 antagonist appeared to increase the salience of non-reward signals earlier in the array that were temporally further away from the choice point.  The D2 antagonist L-741,626 failed alter the time taken to press the collect lever (table 3.1; dose x trial type: F21,252 = 0.94, NS; dose x lights illuminated F9,108 = 0.71, NS). Likewise, L-741,626 administration did not change the latency to respond at the apertures based on the light 39  status of the previous hole (table 3.2; dose x previous hole state: F3,36 = 0.78, NS) or the number of trials completed (table 3.3; dose: F3,45 = 1.76, NS)    Similar to the D2 antagonist, the selective D3 antagonist SB27011A did not affect choice behaviour (figure 3.3e, f; dose x lights illuminated: F9,108 = 0.47, NS; dose x trial type: F21,252= 0.85, NS) or lever choice latency (table 3.1; dose x trial type: F7,84 =1.05, NS; dose x lights illuminated: F3,36 0.4, NS).  There was also no discernible affect on the latency to respond at the apertures, regardless of the status of the previous hole (table 3.2; dose x previous hole state: F3,36 = 0.75, NS), or on the number of trials completed (table 3.3; dose: F1,15 = 0.85, NS). Besides the improvements in lever choice performance, L-745,870 had no effect on collect response latency (table 3.1, dose x trial type: F28,336 = 1.43, NS; dose x lights illuminated F12,144 = 1.01, NS), the latency to respond at the array (table 3.2, dose: F4,44 = 0.87, NS; dose x previous hole state: F4,44 = 1.65, NS) or on the number of trials completed (table 3.3, dose: F4,48 = 0.84, NS).  Effect of pre-treatment with antagonists selective for different D2- receptor subtypes on the response to quinpirole. If any animal failed to complete at least 24 trials (3 of each type), their data were excluded. The final number of rats included in the analysis was 11, except any analysis in which just SB277011A and saline were used (n=12).  In agreement with our previous results, quinpirole significantly increased collect errors (figure 3.4a; dose x lights illuminated: saline – saline vs saline–quin: F3,30 = 10.95, p<0.0001). Both SB277011A and eticlopride failed to prevent the quinpirole-induced increase in choice of the collect lever, while administration of the D2 receptor antagonist L-741,626 exacerbated the 40  erroneous choice behaviour (figure 3.4b; dose: saline-quinpirole vs SB277011A–quinpirole: F1,10 = 0.001, NS; saline–quinpirole vs eticlopride–quinpirole: F1,10 = 0.19, NS; saline-quinpirole vs L741,626–quinpirole: F1,9 = 9.61, p=0.01).  In contrast, administration of the D4 receptor antagonist L-745,870 attenuated the deleterious effects of quinpirole on choice behaviour (figure 3.4b; dose: saline–quinpirole vs L-745,870–quinpirole: F1,10 = 6.33, p=0.03), although not to baseline levels. (figure 3.4b; dose: saline–saline vs L-745,870-quinpirole: F1,11 = 9.14, p=0.01). Examining this effect more closely, we see that blockade of D4 receptors returns performance following quinpirole to near baseline levels on win and near-miss trials (figure 3.4a; saline–saline vs L-745,870–quinpirole- 3 light trials: F1,11 = 1.00, NS; 2 light trials: F1,11 = 1.81, NS; 1 light trials: F1,11 = 12.66, p=0.004; 0 light trials: F1,11 = 5.15, p=0.04). Administration of quinpirole also significantly increased the time taken to choose the collect lever (table 3.1, saline–saline vs saline–quinpirole: F1,10= 27.15 p<0.001), and abolished the robust pattern observed at baseline for collect latencies to decrease with the number of lights illuminated (table 3.2, saline-quinpirole, lights illuminated: F3,30 = 1.95, NS).  None of the antagonists could block the general increase in collect lever response latency (all Fs > 6.22, p< 0.003), although L-745,870 did reinstate the tendency of rats to respond on the collect lever more quickly as the number of lights illuminated declined (saline – quinpirole vs L-745,870 – quinpirole: F3,30 =6.16, p=0.002).  Quinpirole also increased the latency to respond at the array, irrespective of the hole placement or light status of the previous aperture (table 3.2: saline–saline vs saline – quinpirole; dose: F1,11 =6.26, p=0.03; hole placement x previous light status: F1,11 = 1.45, NS), and this effect could not be blocked by any of the antagonists (saline-quinpirole vs [antagonist]-quinpirole: all Fs < 2.15, NS). As previously observed quinpirole administration 41  reduced the number of trials completed (table 3.3, dose: F6,60 = 27.56, p<0.0001). This effect could not be blocked via prior administration of any of the antagonists.   3.4 Discussion Here we show that the D4 receptor plays a critical role in guiding reward expectancy on the rSMT.  Consistent with our previous findings, administration of the D2-like receptor agonist quinpirole increased erroneous attempts to collect reward on non-winning trials (Winstanley et al. 2011).  Although quinpirole has a higher affinity for both D2 and D3 receptors over D4 receptors (Seeman and Van Tol 1993) its effects in the current study could not be blocked via prior administration of a selective D2 (L-741,626), D3 (SB277011A), or mixed D2/3 (eticlopride) antagonist, all of which were without effect when administered independently. Interestingly L-741,626 appeared to exhibit some synergistic effects when administered prior to quinpirole. This may be due to the relatively low dose administered, low doses of D2 antagonists have been suggested to bind to autoreceptors and actually potentiate the effects of phasic dopaminergic firing. However, as administration of L741,626 had no effects when administered in isolation this is uncertain. In contrast, the increase in collect errors induced by quinpirole could be partially attenuated by prior administration of the selective D4 receptor antagonist L-745,870.  Administered in isolation, this D4 antagonist produced an increase in animals’ choice of the optimal lever whereas the D4 agonist PD 168077 impaired discrimination between winning and losing trials. Hence, D4 receptors, rather than D2 or D3 receptors appear to play a critical role in mediating erroneous expectations of reward on the rSMT, and may therefore represent a compelling target for pharmacotherapies to reduce compulsive slot machine play. 42  Although L-745,870 is a highly selective D4 receptor antagonist, we only obtained significant improvement in rSMT performance following the lowest dose of L-745,870 tested (0.5 mg/kg).  This may suggest that, as doses increased, the drug began to bind to receptors other than the D4 receptor, and this acted to oppose the beneficial effects of D4 antagonism.  It is also important to note that L-745,870 exhibits weak partial agonist activity in vitro (Gazi et al. 1998; Gazi et al. 1999; Stewart et al. 2004) .  The fact that this drug had the opposite behavioural effect in the rSMT as compared to a direct D4 agonist, and also attenuated the effects of a direct D2-like agonist, indicates that the improvement in collect errors we have observed is unlikely to result from any partial agonist activity.  Additionally, L-745,870 has been suggested to exhibit partial affinity for 5-HT and NE receptors, however, as the majority of behavioural effects were evidenced at the lowest dose and the reciprocal nature of PD168077 and L-745,870 we feel confident the results are as a result of direct D4 receptor stimulation. Nevertheless, it will be important to verify in follow-up experiments that a similar behavioural profile is observed using a neutral D4 antagonist once we get access to such a compound.   In the current study, the disparity between the effects of D4 receptor -selective agents and those with high affinity for the D2 or D3 receptor was striking, yet these receptors all share a high degree of homology in their transmembrane domains, and all couple to the Gα/o family of G proteins which inhibit adenylyl cyclase activity (Beaulieu and Gainetdinov 2011).  One property that is unique to the D4 receptor, however, is its distribution.  Compared to the other D2-family receptor subtypes, D4 receptor expression is much lower in the brain, and is enriched within the cerebral cortex where its expression is primarily post-synaptic (Ariano et al. 1997; Khan et al. 1998; Le Foll et al. 2009; Primus et al. 1997; Rivera et al. 2008).  Immunohistochemistry data indicates that D4 receptors are most concentrated in the motor, sensory, parietal and cingulate 43  cortices, with moderate D4 immunoreactivity observed in the granular insular (GI), whereas levels of expression are undetectable in the infralimbic region and markedly lower in the prelimbic cortex (Rivera et al. 2008).  D4 receptors are also observed within the striosomal compartment of the striatum (caudate putamen), and are also especially dense in the hippocampus and amygdala (Ariano et al. 1997; Khan et al. 1998; Rivera et al. 2002; Rivera et al. 2008) .  There is some overlap between the areas in which D4 receptors are present and the brain regions that are activated by near-miss stimuli, namely at the level of the ACC and the insula (Clark et al. 2009), this latter area having been implicated in reactivity to drug stimuli (Forget et al. 2010; Pushparaj et al. 2012).  It can therefore be hypothesised that the mechanism by which D4-selective agents are altering performance of the rSMT most likely involves modulation of these brain areas, although this obviously remains to be determined empirically. Within the ACC and GI, the majority of D4 receptors are found in pyramidal-like cells within superficial layers II/III, and also within small cell bodies in layer IV that are probably GABAergic interneurons (Mrzljak et al. 1996).  Neurons in these layers receive thalamic and sensory inputs, and form local corticocortical networks (Jones 1985; McFarland and Haber 2002) .  These superficial layers of cortex also receive considerable dopaminergic input (Berger et al. 1976; Descarries et al. 1987).  However, D4 receptors are not concentrated within pyramidal neurons in layers V and VI, the output pathway by which frontocortical areas modulate striatal activity (Berendse et al. 1992).  In contrast, D1 and D2 receptors are commonly found within the deeper layers (Boyson et al. 1986; Vincent et al. 1993).  Collectively, these data indicate that dopaminergic modulation specifically via D4 receptors is involved in the local processing of information within cortical regions, including the evaluation of sensory and environmental information, rather than in the regulation of afferent cortical projections.   44  Electrophysiological recordings within the PFC provide some support for this suggestion.  D4 receptor activation within the PFC can control the modulatory effect of BLA input on the spontaneous activity of PFC pyramidal neurons, suppressing BLA-evoked inhibition (Floresco and Tse 2007), and appear to enhance the excitatory effect of dopamine on network activity by reducing feed-forward inhibition (Ceci et al. 1999; Yuen and Yan 2009).  A recent theory ventured by Lauzon and colleagues (2010) posits that activation of D4 receptors in the PFC can therefore amplify the emotional salience of environmental stimuli by boosting the facilitatory effects of inputs from both the VTA and BLA on pyramidal neuron firing (Lauzon and Laviolette 2010), inputs known to convey information regarding the affective properties of cues.  The idea that D4 receptor activity can regulate the attribution of salience to cues may explain a number of findings.  D4 receptor knockout mice exhibit reduced exploration of novel stimuli (Dulawa et al. 1999) and a D4 antagonist has been demonstrated to attenuate cue-induced reinstatement in a nicotine self-administration paradigm (Yan et al. 2012).  Hence, decreased neurotransmission via D4 receptors reduces the ability of environmental cues to invigorate behaviour i.e. diminishes their incentive salience.  In contrast, a D4 agonist has been shown to facilitate fear conditioning to normally sub-threshold cues (Lauzon et al. 2009), thereby enhancing the impact of emotionally-significant stimuli.   Applying this theory to the current data, administration of a D4 agonist could lead to an over-emphasis on putative win signals (lights illuminated) within a losing array, leading to the generation of a false expectation of reward and biasing the animal towards responding on the collect lever.  In contrast, the D4 antagonist may dampen the incentive salience of such cues, leading to a more realistic expectation that reward will not be delivered on non-win trials, and a reduction in erroneous collect responses.  The theory that D4 receptors could play a particular 45  role in regulating the behavioural significance of environmental stimuli may also explain why such dramatic effects of D4 agents are observed on the rSMT, yet these compounds have produced negligible behavioural effects in related tests of impulsivity and decision-making (Koffarnus et al. 2011; Milstein et al. 2010; St. Onge and Floresco 2009) .  In the latter, complex conditioned stimuli are not routinely used to signal the appropriate response, whereas in the rSMT the pattern of illuminated lights is the key predictor of optimal choice.  However, the exact locus of these results remains elusive, further studies will attempt to elucidate which brain regions are critical for mediating the effects of D4 agents. Subsequent investigations will examine the effects of micro-injections or deactivating key areas of interest, namely the insula and ACC.  In sum, the data presented here suggest a novel role for the dopamine D4 receptor in guiding reward expectancy in a slot machine task, potentially through D4-receptor mediated regulation of the incentive salience of reward-paired cues.  The D4 receptor antagonist L-745,870 could decrease the frequency of erroneous attempts to collect reward on non-win trials at baseline, and could also attenuate the dramatic increase in erroneous collect responses caused by the D2-family agonist quinpirole.  These effects were not observed using selective D2 or D3 antagonists, highlighting a selective role for D4 receptor-selective agents.  The voracity with which slot machines can engender pathological gambling, coupled with the lack of efficacious treatments, make this a timely and important observation.  As such, these findings have potential clinical implications for the treatment of compulsive slot machine play, not only in GD but also in PD patients treated with dopamine agonist therapy  46   Figure 3.1 Baseline performance of rSMT. On win trials, when all three lights had set to on ([1,1,1]), animals chose the collect lever 100% of the time (a, b).  As the number of lights illuminated decreased, so did the preference for the collect lever.  Animals consistently showed a strong preference for the collect lever on 2-light losses i.e. near-miss trials.  All data shown are the mean across five sessions ±SEM. 47    Figure 3.2 Effects of the D2-like agonist quinpirole, the D4 agonist PD168077 and the D4 antagonist L-745,870 on performance of the rSMT. Quinpirole administration increased the proportion of erroneous collect responses on all non-winning trials (a,b). This effect was particularly evident at the higher doses administered.  The highest dose of PD168077 similarly impaired performance (c,d).  In contrast L-745,870 administration led to a decrease in erroneous 48  collect responses (e,f). These improvements were most evident on trials when the last light set to off.  All doses are given in mg/ml and data are shown as mean ±SEM.     Figure 3.3 Effect of the D2 antagonist L-741,626, the D3 antagonist SB27701A and the D3 agonist PD129807 on performance on the rSMT. Neither the D3 agonist or antagonist had any effect on lever choice performance; similarly, a selective D2 antagonist did not produce any appreciable effects. 49    Fig 3.4 Effect of pre-treatment with antagonists selective for different D2- receptor subtypes on the response to quinpirole. Consistent with earlier administration quinpirole produced robust deficits in choice behaviour, increasing preference for the collect lever on all non-winning trials. These deficits could not be attenuated via pre-treatment with any of the antagonists except L-745,870 which significantly improved optimal lever choice (a,b).  All doses are given in mg/ml and data are shown as mean ±SEM.     Table 3.1 Latency to respond on the left lever by trial type at baseline and during pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose (mg/kg) 1,1,1 1,1,0 1,0,1 1,0,0 0,1,1 0,1,0 0,0,1 0,0,0 Baseline  0.66± 0.08 0.73 ± 0.06 0.63 ± 0.09 0.38 ±0.05 0.65 ± 0.08 0.37 ± 0.06 0.37 ± 0.07 0.13 ± 0.02 SB277011A 0 0.76 ± 0.29 0.67 ± 0.09 0.57 ± 0.23 0.49 ± 0.23 0.58 ± 0.23 0.28 ± 0.15 0.43 ± 0.22 0.19 ± 0.01  0.5 0.67 ± 0.22 0.84 ± 0.21 0.61 ± 0.16 0.38 ± 0.12 0.58 ± 0.21 0.42 ± 0.14 0.34 ± 0.09 0.27 ± 0.14  1.5 0.67 ± 0.15 0.86 ± 0.13 0.68 ± 0.18 0.45 ± 0.13 0.63 ± 0.19 0.32 ± 0.11 0.40 ± 0.14 0.13 ± 0.06 50  Drug Dose (mg/kg) 1,1,1 1,1,0 1,0,1 1,0,0 0,1,1 0,1,0 0,0,1 0,0,0  5.0 0.73 ± 0.24 0.93 ± 0.15 0.65 ± 0.13 0.38 ± 0.13 0.63 ± 0.18 0.28 ± 0.12 0.47 ± 0.21 0.20 ± 0.07 L-741,626 0 0.79 ± 0.21 0.85 ± 0.09 0.68 ± 0.22 0.24 ± 0.07 0.57 ± 0.14 0.32 ± 0.11 0.29 ± 0.11 0.22 ± 0.06  0.1 0.70 ± 0.17 0.83 ± 0.20 0.67 ± 0.21  0.34 ± 0.13 0.51 ± 0.15 0.37 ± 0.16 0.38 ± 0.19 0.11 ± 0.05  0.3 0.82 ± 0.19 0.71 ± 0.13 0.62 ± 0.18 0.40 ± 0.18 0.63 ± 0.22 0.43 ± 0.13 0.31 ± 0.18 0.18 ± 0.12  1.0 0.88 ± 0.26 0.93 ± 0.13 0.85 ± 0.28 0.60 ± 0.24  0.59 ± 0.24 0.53 ± 0.13 0.36 ± 0.18 0.29 ± 0.11 PD128907 0 0.73 ± 0.05 1.10 ± 0.07 0.90 ± 0.07 0.57 ± 0.06 0.86 ± 0.06  0.48 ± 0.04 0.49 ± 0.05 0.14 ± 0.01  0.01 0.74 ± 0.04 0.94 ± 0.05 0.81 ± 0.06 0.34 ± 0.02 0.75 ± 0.03 0.45 ± 0.02 0.48 ± 0.03 0.12 ± 0.01  0.03 0.86 ± 0.05 1.13 ± 0.08 1.00 ± 0.06 0.76 ± 0.05 0.93 ± 0.07 0.51 ± 0.03 0.56 ±0.04 0.29 ± 0.03  0.1 1.04 ± 0.05 1.07 ± 0.04 1.00 ± 0.05 0.66 ± 0.04 0.94 ± 0.05 0.70 ± 0.03 0.74 ± 0.03 0.25 ± 0.02 Quinpirole 0 1.34 ± 0.28 0.99 ± 0.26 1.13 ± 0.29 0.69 ± 0.17 0.99 ± 0.19 0.52 ± 0.11 0.40 ± 0.12 0.14 ± 0.06  0.0125 1.24 ± 0.27 1.76 ± 0.46 1.54 ± 0.28 1.08 ± 0.28 1.22 ± 0.22 0.92 ± 0.15 0.97 ± 0.22 0.56 ± 0.13  0.0375 1.65 ± 0.22 2.50 ± 0.41 1.75 ± 0.23 1.64 ± 0.25 1.78 ± 0.29 1.87 ± 0.46 1.69 ± 0.40 2.10 ± 0.77  0.125 3.84 ± 1.10  3.79 ± 0.99 3.99 ± 1.03 2.47 ± 0.65 3.30 ± 0.55 1.91 ± 0.31 3.38 ± 1.09 3.39 ± 1.14 PD168077 0 0.84 ± 0.19 0.61 ± 0.09 0.60 ± 0.20 0.20 ± 0.04 0.62 ± 0.14 0.19 ± 0.06 0.19 ± 0.08 0.03 ± 0.01  0.5 0.88 ± 0.21 0.64 ± 0.11 0.64 ± 0.17 0.26 ± 0.07 0.60 ± 0.16 0.17 ± 0.08 0.17 ± 0.08 0.24 ± 0.06  1.0 0.75 ± 0.14 0.69 ± 0.11 0.45 ± 0.13 0.16 ± 0.04 0.59 ± 0.16 0.21 ± 0.03 0.14 ± 0.04 0.06 ± 0.03  5.0 0.91 ± 0.17 0.71 ± 0.12 0.67 ± 0.18  0.21 ± 0.05 0.62 ± 0.19 0.20 ± 0.03 0.22 ± 0.06 0.10 ± 0.03  10.0 1.01 ± 0.23 0.94 ± 0.11 0.98 ± 0.17 0.55 ± 0.10 0.81 ± 0.16 0.64 ± 0.10 0.68 ± 0.13 0.32 ± 0.10 L-745,870 0 1.11 ± 0.05 1.16 ± 0.06 0.84 ± 0.04 0.73 ± 0.04 0.95 ± 0.05 0.58 ± 0.06 0.44 ± 0.05 0.29 ± 0.03  0.5 1.19 ± 0.06 1.02 ± 0.04 0.81 ± 0.03 0.34 ± 0.02 0.78 ± 0.03 0.46 ± 0.03 0.29 ± 0.02 0.16 ± 0.01  1.0 1.10 ± 0.05 1.04 ± 0.05 0.92 ± 0.06  0.37 ± 0.02 0.83 ± 0.03 0.63 ± 0.04 0.39 ± 0.04 0.16 ± 0.01 51  Drug Dose (mg/kg) 1,1,1 1,1,0 1,0,1 1,0,0 0,1,1 0,1,0 0,0,1 0,0,0  5.0 1.16 ± 0.05 0.85 ± 0.04 1.00 ± 0.05  0.50 ± 0.04 0.86 ± 0.03 0.60 ± 0.04 0.51 ± 0.04 0.18 ± 0.01  10.0 1.11 ± 0.05 1.38 ± 0.06 0.86 ± 0.05  0.65 ± 0.03 0.95 ± 0.04 0.73 ± 0.05 0.39 ± 0.03 0.20 ± 0.02 Saline - Saline 0 – 0  1.03 ± 0.13 1.05 ± 0.13 0.90 ± 0.14 0.35 ± 0.07 0.89 ± 0.17 0.42 ± 0.11 0.39 ± 0.14 0.20 ± 0.05 Saline - quinpirole 0 – 0.0125 2.02 ± 0.29 2.03 ± 0.29 2.59 ± 0.59 2.10 ± 0.50  1.89 ± 0.32 1.90 ± 0.36 1.70 ± 0.24 1.64 ± 0.56 SB277011A - quinpirole 1.0 – 0.0125 1.44 ± 0.24 1.93 ± 0.41 1.80 ± 0.24  1.69 ± 0.50 1.52 ± 0.24 1.99 ± 0.38  1.66 ± 0.39 2.60 ± 0.82 L-741,626 - quinpirole 0.5 – 0.0125 2.12 ± 0.43 2.26 ± 0.36 1.56 ± 0.25 1.75 ± 0.34 1.78 ± 0.36 2.42 ± 0.49 1.75 ± 0.36 2.09 ± 0.89 Eticlopride - quinpirole 0.01 – 0.0125 1.05 ± 0.13 1.71 ± 0.39 1.11 ± 0.22 1.50 ± 0.31 1.61 ± 0.68 1.99 ± 0.72 1.53 ± 0.45 2.13 ± 0.64 L-745,870 - quinpirole 0.5 – 0.0125 1.33 ± 0.20  2.08 ± 0.32 1.89 ± 0.15 1.50 ± 0.62 1.30 ± 0.23 0.85 ± 0.17 0.92 ± 0.17 0.86 ± 0.24  Table 3.2 Latency to respond at subsequent hole based on the status of the previous hole for baseline and pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose (mg/kg) Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Baseline  1.08 ± 0.10 1.77 ± 0.41 1.37 ± 0.17 1.58 ± 0.18 SB277011A 0 1.33 ± 0.22 1.62 ± 0.22 1.31 ± 0.19 1.34 ± 0.13  0.5 1.35 ± 0.25 1.53 ± 0.21 1.37 ± 0.22 1.47 ± 0.14  1.5 1.43 ± 0.22 1.34 ± 0.21 1.32 ± 0.19 1.46 ± 0.15  5.0 1.25 ± 0.19 1.56 ± 0.28 1.24 ± 0.18 1.62 ± 0.17 L-741,626 0 1.46 ± 0.27 1.33 ± 021 1.21 ± 0.22 1.57 ± 0.33  0.1 1.32 ± 0.26 1.32 ± 0.19 1.36 ± 0.19 1.38 ± 0.16  0.3 1.35 ± 0.24 1.42 ± 0.20 1.45 ± 0.25 1.57 ± 0.24  1.0 1.91 ± 0.53  1.78 ± 0.31 1.50 ± 0.22 2.06 ± 0.47 PD128907 0 1.04 ± 0.24 1.39 ± 0.26 1.17 ± 0.14 1.32 ± 0.15  0.01 1.33 ± 0.20 1.53 ± 0.36 1.26 ± 0.17 1.49 ± 0.16  0.03 1.14 ± 0.16 1.67 ± 0.37 1.26 ± 0.18 1.73 ± 0.28  0.1 1.21 ± 0.16 2.30 ± 0.67 1.72 ± 0.40 2.65 ± 0.51 Quinpirole 0 0.90 ± 0.15 1.43 ± 0.40 1.28 ± 0.25 1.36 ± 0.13 52  Drug Dose (mg/kg) Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off  0.0125 2.71 ± 1.25 3.40 ± 1.48 1.81 ± 0.34 2.91 ± 0.77  0.0375 12.91 ± 8.77 25.75 ± 16.20 29.76 ± 18.82 23.73 ± 19.58  0.125 9.65 ± 2.86 5.33 ± 1.15 6.37 ± 2.34 7.62 ± 2.76 PD 168077 0 0.98 ± 0.26 1.02 ± 0.15 0.99 ± 0.17 0.99 ± 0.10  0.5 1.09 ± 0.26 0.93 ± 0.14 1.00 ± 0.19  1.01 ± 0.15  1.0 1.19 ± 0.28 0.98 ± 0.19 0.99 ± 0.19 1.05 ± 0.15  5.0 1.05 ± 0.15 1.09 ± 0.17 1.03 ± 0.17 1.05 ± 0.13  10.0 1.26 ± 0.29 1.53 ± 0.52 1.13 ± 0.24 1.82 ± 0.58 L-745,870 0 1.19 ± 0.20 1.76 ± 0.30 0.93 ± 0.12 1.31 ± 0.28  0.5 0.88 ± 0.13 1.04 ± 0.11 0.92 ± 0.10 1.17 ± 0.18  1.0 1.08 ± 0.16 3.55 ± 2.36 0.95 ± 0.12 2.27 ± 1.31  5.0 1.02 ± 0.15 2.14 ± 0.85 0.88 ± 0.08 1.52 ± 0.43  10.0 0.82 ± 0.09 1.34 ±  0.25  1.09 ± 0.14 1.07 ± 0.14 Saline - Saline 0 – 0  1.03 ± 0.17 1.04 ± 0.19 1.00 ± 0.18 1.20 ± 0.20 Saline - quinpirole 0 – 0.0125 79.63 ± 49.61 15.78 ± 7.0 10.98 ± 7.13 4.94 ± 1.71 SB277011A - quinpirole 1.0 – 0.0125 2.89 ± 1.28 7.10 ± 2.46 5.39 ± 2.84 3.81 ± 0.77 L-741,626 - quinpirole 0.5 – 0.0125 2.20 ± 0.36 5.41 ± 1.76 7.00 ± 4.37 14.59 ± 5.75 Eticlopride - quinpirole 0.01 – 0.0125 3.20 ± 0.99 10.42 ± 5.46 3.22 ± 1.63 9.92 ± 3.85 L-745,870 - quinpirole 0.5 – 0.0125 5.21 ± 3.42 18.06 ± 6.12 1.64 ± 0.54 17.89 ± 11.09 53  Table 3.3 Trials completed at baseline and following pharmacological challenges. Data shown are mean ± SEM Drug Dose (mg/ml) Trials completed Baseline  100.39 ± 2.07 SB277011A 0 106.69 ± 7.73  1.0 105.69 ± 7.66  3.0 106.31 ± 7.50  10.0 104.54 ± 6.68 L-741,626 0 126.46 ± 6.18  0.1 124.69 ± 6.02  0.3 121.62 ± 7.34  1.0 117.69 ± 5.50 PD 128907 0 95.15 ± 5.33  0.01 96.62 ± 4.85  0.03 86.92 ± 5.10  0.1 78.92 ± 5.38 Quinpirole 0 94.63 ± 10.48  0.0125 57.08 ± 6.40  0.0375 33.54 ± 5.63  0.125 31.54 ± 3.02 PD 168077 0 124.46 ± 6.12  0.5 124.69 ± 6.02  1.0 121.62 ± 7.34  5.0 117.69 ± 5.50  10.0 94.91 ± 5.69 L-745,870 0 97.38 ± 5.38  0.5 99.69 ± 4.99  1.0 92.15 ± 6.90  5.0 93.08 ± 5.64  10.0 93.23 ± 6.26 Saline - Saline 0 – 0  88.15 ± 8.66 Saline - quinpirole 0 – 0.0125 31.15 ± 5.55 54  Drug Dose (mg/ml) Trials completed SB277011A - quinpirole 1.0 – 0.0125 30.23 ± 4.98 L-741,626 - quinpirole 0.5 – 0.0125 25.46 ± 4.00 Eticlopride - quinpirole 0.01 – 0.0125 30.31 ± 5.82 L-745,870 - quinpirole 0.5 – 0.0125 44.5 ± 6.60   Table 3.4 Relative binding sites and affinities for drugs. Data shown are Ki (nM) Drug (purported target) Receptor/affinity L741,626 (D2 antagonist) Rat, DA, D2-4.78. D3-31.55, D4-316.22. D1-794.33, D5-630.96 (Millan et al. 2000) 5-HT1A-1,000,  5-HT1B-1,000, 5-HT1D-1,000, 5-HT2A- 316.23, 5-HT2B-630.96, 5-HT2C-1,000 (Millan et al. 2000)  Adrenergic α1-251.19, adrenergic α2A-1,258.93 (Millan et al. 2000) Eticlopride  (D2-like antagonist) DA D2 – 0.04, D3 – 0.31 (Millan et al. 2004; Tang et al. 1994), D4 – 24.65 (Tang et al. 1994), D1 & D5 - >10,000 (Sunahara et al. 1991) 5-HT1 – 6,220.00, 5-HT2 – 830.00 (Hall et al. 1986) Adrenergic α1 – 112.00, α2 – 699.00,  β - >10,000 (Hall et al. 1986) Cholinergic, muscarinic - >10,000 (Hall et al. 1986) Histamine H1 - >10,000 (Hall et al. 1986) Quinpirole (D2-like agonist) DA D2-15.5, D3-19.3 (Tang et al. 1994), D4-33.8, D1 & D5 > 10,000 (Millan et al. 2002) 5-HT1A-1,698.24, 5-HT1B,D>10,000,  5-HT2A,B & C >10,000 5-HT3->10,000  (Millan et al. 2002) adrenergic α1A,B,D & α2A,C,D >10,000, adrenergic α2B-2,511.89 (Millan et al. 2002) Histamine H1-1,047.13 (Millan et al. 2002).  SB277011A (D3 antagonist) DA D3 – 10.71, D2 – 2,818.38 (Reavill et al. 2000)  5-HT1A – 5,011.87, 1B – 6,309.87, 1D – 1,621, 1E - >10,000, 1F-6,309.57, 5-HT2A - >10,000, 2B – 1,288.24, 2C - >10,000. 5-HT4 – 5,011.87, 5-HT6 - >10,000, 5-HT7 - >10,000 (Reavill et al. 2000) PD128,907 (D3 agonist) DA D2—389.00 (Sautel et al. 1995), D3-1.97 (Cussac et al. 2000; Sautel et al. 1995) L-745,870 (D4 antagonist) D2-1600, D3-6200 & D4-1.5 (Patel et al. 1997) 5-HT 1A <300 (Patel et al. 1997) Sigma-1 <300 (Patel et al. 1997) PD168077 (D4 agonist) DA D4-8.07, >400 fold over D2 and >300 over D3  (Glase et al. 1997) 55  Chapter 4: Experiment 2: Elucidating the role of D4 receptors in mediating attributions of salience to incentive stimuli on Pavlovian conditioned approach and conditioned reinforcement paradigms   4.1 Introduction  Drug addiction is a chronic relapsing disorder.  Estimates suggest that almost 90% of addicts will relapse following their first diagnosis (DeJong 1994). This tendency to relapse, even following long periods of abstinence, is characteristic of the disorder and an intractable problem for treatment (Koob and Volkow 2010).  Theories of drug addiction suggest that the reason relapse rates are so high is because drugs co-opt associative learning mechanisms through potentiated dopamine (DA) signaling, and as a result cues associated with drugs become imbued with incentive salience promoting continued drug seeking (Field and Cox 2008; Robinson and Berridge 1993; 2001; 2003).  Certainly, cues associated with drugs can reliably induce craving in addicts (see Carter and Tiffany 1999 for review). Cues are therefore thought to play a key role in the maintenance and formation of substance addiction. There is a growing recognition that the phenomenology of behavioural and substance addictions may overlap, which has culminated in gambling disorder (GD) being recently reclassified as an addictive disorder in the DSM 5 (American Psychiatric Association. and American Psychiatric Association. DSM-5 Task Force. 2013). However, it is currently less clear if cues play a similar role in the etiology of behavioural addictions.  In contrast to substance addiction, which is typified by altered dopaminergic signaling within the striatum (Volkow et al. 2004), there does not appear to be any such 56  equivalent prototypical dysfunction in the DA system within gamblers (Limbrick-Oldfield et al. 2013).  Animal models can play an important role in studying the neurobiological basis of psychiatric symptoms.  We have developed and validated the rSMT, and our previous data suggest that D4 receptors modulate animals’ responsivity to reward-related stimuli and contribute to erroneous expectations of reward in this paradigm (Cocker et al. 2014; Winstanley et al. 2011).  These data add to multiple lines of enquiry that indicate that D4 receptors are involved in appropriate encoding of associative emotional information (Cocker et al. 2014; Grace et al. 2005; Lauzon et al. 2012; Lauzon et al. 2009; Lauzon and Laviolette 2010). D4 receptors are part of the D2-like receptor family, and signaling through the D2-like class of receptors has been generally associated with salience attribution (Berridge and Robinson 1998). Yet a particular role for D4 receptors in this process has, to our knowledge, not been explored.  Relatively simple behavioural paradigms such as Pavlovian conditioned approach (autoshaping) and conditioned reinforcement (CRf) may be useful in revealing any role for D4 receptors in either Pavlovian or instrumental forms of incentive motivation. Ostensibly, both of these paradigms measure how reward-paired cues can influence behaviour, but differ in regards to brain areas and neurochemical regulation (Cardinal et al. 2002b). Furthermore, these tasks could be considered hierarchical, in that the property of the cues increases in behavioural significance, from attracting attention and eliciting approach (autoshaping), to becoming the goal itself (CRf) (Flagel et al. 2009).  During autoshaping, a classically conditioned stimulus (CS) reliably predicts delivery of an unconditioned stimulus (US), typically food. Over repeated CS-US pairings, animals begin to approach and interact with the CS, even though the US is not contingent on any such response.  57  Typically, animals vary in their behavioural response; some animals, termed ‘sign-trackers’ (ST), predominantly approach and interact with the CS in an appetitively-specific manner e.g. bite or chew a lever that predicts food, while other “goal-tracker (GT)” animals instead orient towards the delivery location of the US (Flagel et al. 2007). The expression of both goal- and sign tracking appears contingent on DA, as perturbations to dopaminergic signaling via administration of D2-like agonists or antagonists impair the expression of both ST and GT (Di Ciano et al. 2001; Fraser et al. 2016).  In contrast, interventions that lead to augmented dopaminergic activity, such as amphetamine sensitisation, chronic stress or exposure to uncertainty lead to an increase in the acquisition of a ST over a GT response (Robinson et al. 2015). Conditioned stimuli signaling non-contingent food delivery has been linked with increased dopamine release in the core, but not the shell sub-region of the nucleus accumbens (NAc) (Biesdorf et al. 2015). Moreover, the increased incentive salience assigned to the CS by ST has been linked with increased DA release within the NAc (Flagel et al. 2011). But, while sign tracking could be taken as evidence that reward paired cues are salient and attractive, it does not necessarily imply that they can influence goal-directed action.   In tests of CRf, a CS is first paired to the delivery of a US during training.  In the subsequent test phase that determines whether the CS has become a true conditioned reinforcer, the animal is required to perform a novel response, such as lever pressing, that is reinforced solely by the CS.  As with ST, CRf also appears to be primarily influenced by dopaminergic activity within the NAc, as infusion of amphetamine into this area potentiates animals responding for the CS, an effect that is remediated by prior blockade of D1 or D2 receptors (Taylor and Robbins 1984; Wolterink et al. 1993). Similarly, infusion of non-selective D1-like 58  and D2-like agonists into the NAc increase responding for the CS, an effect mimicked by a systemically administered D2-like agonist (Beninger and Ranaldi 1992; Wolterink et al. 1993).  Here, we investigate whether systemic administration of a selective agonist at the D4 receptor can affect cue reactivity on both an autoshaping and CRf paradigm. Elucidating the role of D4 receptors on these simple tasks could not only clarify whether these receptors are involved in the attribution of salience to environmental stimuli, but also refine hypotheses regarding the cognitive mechanism by which D4 receptors agents may theoretically improve behavioural addictions.   4.2 Additional methods Subjects  Animals (n=47 total, 27 female) were bred in house from animals obtained from Charles River (Charles River Laboratories, St. Constant, Canada) and the Rat Resource and Research Centre (RRRC, Columbia, MO) as part of a breeding program for transgenic rats that express cre recombinase (Cre) in neurons that contain choline acetyl transferase (ChAT; Long Evans-Tg(ChAT-Cre)5.1Deis, RRRC#00658) or tyrosine hydroxylase (Long Evans-Tg(TH-Cre)3.1Deis, RRRC#00659).  All the rats used in these experiments were transgene negative, and therefore theoretically indistinguishable from outbred Long Evans rats.  Rats were weaned at post-natal day 21 and housed in same sex groups of two-to-three animals per cage.  Animals were given ad libitum access to standard rat chow and water until males reached approximately 300g and females approximately 200g.  A week before behavioural testing, subjects were food restricted to 85% of their free feeding weight and maintained on 14g (male) or 9g (female) rat chow given daily.   59   Pharmacological challenges PD168077 (Tocris, Ellisville, MO, USA) was prepared fresh daily, calculated as the salt and dissolved in 0.9% sterile saline to a volume of 10mg/ml.  The dose used was based on a previous report (Cocker et al. 2014).  All injections were delivered via the intraperitoneal route.   Experiment 1: The effect of the dopamine D4 agonist PD168077 on conditioned approach. Animals (n=19 total, 11 female) were trained on an autoshaping paradigm wherein either the left or the right lever would extend into the chamber, accompanied by illumination of the cue light located above the respective lever, for five seconds, after which the lever would retract and the cue light would be turned off.  The lever designations were counterbalanced, such that for half the animals the left lever was the CS+ and the right the CS-.  During CS+ trials, a single sugar pellet would be delivered to the hopper, signaled by illumination of the tray light.  The tray light would remain illuminated until the animal made a nose poke response at the tray, at which point another trial could begin.  Presentation of the CS- lead to no programmed consequences, and the countdown to the next trial was triggered automatically following stimulus presentation.  Trials started on a variable interval (VI) 60s schedule and trial types varied pseudo-randomly such that animals experienced equal exposure to both CS+ and CS- trials.  Following 10 successive training sessions, animals were matched for goal- and sign-tracking behaviour and separated into two groups.  On the first pharmacological challenge, one group received 10mg/kg PD168077 and the other saline.  These designations were subsequently reversed on the second challenge day allowing for a within-subjects design.  All injections were administered 10 minutes prior to the start of the test session.  Drug challenge sessions were administered over a 3-day cycle, starting 60  with a baseline session followed by a drug or saline injection and lastly a non-testing day during which animals remained in their home cage.   Experiment 2: The effect of the dopamine D4 agonist on responding for CRf.  Animals (n=28 total, 16 female) were trained on a Pavlovian approach task for 10-successive days during which a cue light acted as a CS+, predicting the delivery of a single sugar pellet. A cue light located on the opposite side of the food hopper (CS-) was never associated with reward.  The location of the CS+ (left or right of the food hopper) was counter-balanced across the cohort.  Trials started on a VI60s schedule and trial types varied pseudo-randomly such that animals experienced equal exposure to both the CS+ and CS-.  After the 10 training sessions, animals were matched for the number of nose-poke entries made into the food hopper during presentation of the CS+, indicative of the extent to which the CS-US association had been acquired, and split into two equal groups.  One group was injected with 10mg/kg of PD168077 and the other saline 10 minutes prior to the start of the CRf test session.  During this test session, both levers were extended into the chamber.  Responding on each lever resulted in a 5s illumination of the cue light located above it (i.e. the CS+ or CS-) on a fixed ratio-1 (FR) schedule.  Importantly, no reward was delivered for either operant response; the only ‘reward’ received for responding was the presentation of visual stimuli that either were, or were not, previously associated with reward delivery.   Data analysis All analyses were conducted using SPSS (v23.0).  The variables of interest in the autoshaping paradigm were the number of lever contacts and hopper entries during both the CS+ and the CS-. 61  The ratio of these responses was also analysed (CS+ responses divided by the total number of lever responses (CS+ responses + CS- responses)) in order to control for the general tendency to approach the stimuli/hopper independent of any association with the CS+ and reward delivery.  Lastly, the latency to nosepoke at the food hopper following presentation of the CS+ and CS- was also measured.  Animals were labelled as either goal- or sign-trackers if they demonstrated at least a sixfold greater propensity to approach either the food hopper over the CS+ or vice versa, similar to (Robinson et al. 2015).  Analysis were conducted with all animals, but based on an a priori assumption of differences within the groups, separate analyses were also conducted including only animals categorized as ST and GT.  Baseline data was conducted as a repeated measures ANOVA with session (10 levels) and stimuli (2 levels) as within subjects’ factors and sex (2 levels) as a between subjects factor. Analyses comparing performance between ST and GT were conducted using similarly structured ANOVA’s but included phenotype (2-levels) as a between subjects’ factor. Similar to previous reports (Fraser et al. 2016), responses to drug challenges were analysed via repeated measures ANOVA with phenotype (2-levels, goal tracker or sign tracker) and drug (two levels, saline vs. PD168077) as within subject factors and sex as a between subjects’ factor.  The key dependent variables for conditioned reinforcement training were the number of nose-poke responses made at the food tray in response to the CS+ and the CS-, the ratio of these responses as well as the latency to make those responses.  For the CRf test, the main variables analysed were the number of responses on the lever that lead to CS+ presentation, the number of responses on the lever that delivered the CS-, the ratio of these responses, the number of nosepokes at the food delivery magazine during presentation of the CS+ and the number of nosepokes at the food delivery magazine following the CS-. Baseline data were analysed via a 62  repeated measures ANOVA with session (10-levels), stimuli (2 levels, CS+ or CS-) and sex as a between subjects’ factor.  Data from the single test day were analysed using independent samples t-tests.   4.3 Results  Experiment 1: Pavlovian conditioned approach  Of the 19 animals trained on the autoshaping task, 6 (2 females) were classified as ST’s and 3 (2 females) met the criteria for GT representing 31.58% and 15.79% of the cohort respectively.  Baseline behaviour All animals made significantly more responses on the lever that predicted reward delivery in comparison to the CS- lever (stimuli: F1,17 = 7.41, p=0.01). There was no effect of sex on the number of lever contacts animals made (sex: F1,17 = 0.64, NS). When comparing animals who were classified as either ST or GT, although both groups still differentiated between the CS+ and the CS- (figure 4.1; stimuli: F1,7 = 7.30, p=0.03), ST’s made significantly more lever contacts (stimuli X group: F1,7 = 8.64, p=0.02) and only ST animals showed a continued increase in the number of lever contacts throughout training (figure 4.1; ST; F1,5 = 17.06, p=0.009; session X stimuli; F8,40 = 8.35, p<0.0001; GT; stimuli: F1,2 = 129.31, p<0.0001; session X stimuli: F8,16 = 1.90, NS). In contrast to the lever effects, the number of nose poke responses animals made during CS+ and CS- presentation only differed at a trend level (stimuli: F1,17 = 3.33, p=0.09).  There were, again, no sex differences in animals nosepoke responses (sex: F1,17 = 0.05, NS).  However, examining just ST and GT, there was a significant interaction based on group, indicative of the different response profiles of ST and GT (stimuli X group: F1,7 = 20.4, p=0.003).  Interestingly, when the number of nose poke entries for either GT or ST groups were examined, 63  only the ST’s showed a significant effect based upon the stimulus presented (ST; stimuli: F1,5 = 34.0, p=0.002; GT; stimuli: F1,2 = 4.39, NS).  The effect observed in the ST group appears due to an increase in head entries during the CS-, i.e. during a CS+ animal classified as ST’s are more likely to approach the lever, but during a CS- alternative behaviours are expressed.  The lack of differentiation observed in the GT group is likely due to the combination of low numbers of animals expressing a GT phenotype, and the high variability within the group.  All animals were quicker to nosepoke into the food hopper in response to the CS+ presentation, independent of behavioural phenotype or sex (stimuli; F1,17 = 49.69, p<0.0001, p=0.023; sex: F1,17 = 1.10, NS stimuli X group: F1,5 = 1.32, NS), indicating all animals learnt the contingencies in play.   The effects of the D4 agonist PD168077 on behavioural responding during a conditioned approach paradigm.  There were no effects of PD168077 on the number of lever responses animals made during the CS+ or CS- (drug: F1,7 = 0.12, NS; drug  X sex: F1,17 = 0.09, NS). These null effects were the same regardless of animals’ response phenotype. (figure 4.1; drug: F1,5 = 0.71, NS).  Similarly, there were no alterations in the number of nosepokes animals made into the food hopper during CS+ or CS- presentation (drug: F1,17 = 0.25, NS;, NS; drug X sex: F117= 0.19, NS drug X group: F1,5 = 3.64) or animals latency to respond (drug: F1,17 = 0.08, NS; drug X sex: F1,17 = 0.61, NS; drug X group: F1,5 = 0.07)  Experiment 2: Conditioned reinforcement  Three animals were excluded from the analysis as they failed to differentiate between CS+ and the CS- during the Pavlovian training phase i.e. they had a CS+ ratio <0.55.  A further 64  five animals were removed from the analysis as they showed severely limited behavioural output during the CRf test day (total lever press responses <30). The total number of animals included was therefore 20 (10 females; 6 saline, 4 PD168077 and 10 males 5 saline and 5 PD168077). Baseline behaviour Animals demonstrated robustly elevated nosepoke responding at the food magazine tray during CS+ presentation in comparison to CS-, indicating that animals learnt the contingencies of the task (figure 4.2; stimuli: F1,22 = 62.81, p<0.0001).  There was no effect of sex on animals’ ability to differentiate between CS+ and CS- (sex: F1,22 = 1.69, NS). There were also no sex differences in the CS+/CS- response ratio or the latency to approach or retrieve reward (sex: F1,22 <0.91, NS).  The effects of the D4 agonist PD168077 on behavioural responding during a CRf paradigm: There were no differences in baseline behaviour based on sex, therefore the two groups were pooled for all drug analysis.  Administration of the D4 agonist produced no alterations in animals CS+/CS- ratio (figure 4.2a; drug: T11 = -0.22, NS) i.e. it did not make animals any more or less likely to respond on the lever that delivered the CS+ in comparison to the CS-. However, in comparison to saline treated animals, PD168077 did lead to a general reduction in the number of lever presses; including on the CS+ lever at a trend level (figure 4.2b; drug: T11 = 2.01, p=0.07), on the CS- lever (figure 4.3c; drug: T11 = 3.51, p=0.005) and consequently on the total number of lever responses (figure 4.3d; drug: T11 = 3.05, p=0.01).  Rather than modulation of the salience of reward-paired cues, any drug-induced decreases in responding for the cues were therefore likely due to a decrease in motor output.  When animals did make a response, however, there were no alterations in latency, or the number of nosepokes at the food magazine during, or after presentation of the CS+ (figure 4.3e-f; drug: T11 all T’s<1.5, NS).  65   4.4 Discussion Here, we show for the first time that pharmacological modulation of the D4 receptor does not affect cue-driven responding on either an autoshaping or CRf paradigm. Simple behavioural tasks, such as these, arguably allow for a measure of Pavlovian and instrumental incentive motivation, and it has been theorized that increased salience attributed to drug associated cues is essential in the formation and maintenance of addiction (Field and Cox 2008; Robinson and Berridge 2008). For instance, animals with a ST phenotype show greater locomotor sensitisation to cocaine administration and increased cocaine intake, leading to the suggestion that ST may represent a behavioural marker for addiction proneness (Flagel et al. 2008; Tunstall and Kearns 2015). Pharmacological agents broadly acting at D2-like receptors modulate ST (Fraser et al. 2016; Lopez et al. 2015; Robinson et al. 2015). Therefore, the lack of effect of the highly selective D4 agonist PD168077 is significant, given the suggestion that D4 receptors may play a crucial role in controlling attributions of salience to gambling-related stimuli (see Cocker and Winstanley 2015 for discussion). Furthermore, consistent with former investigations, there do not appear to be any sex differences in either the acquisition or expression of motivated behaviors to conditioned stimuli. (Anderson and Spear 2011; Bertz et al. 2016).  The failure of a D4 agonist to alter behaviour here is in some regards surprising, in that a canon of evidence implicates D4 receptor dysfunction in multiple disease states that are characterised by inappropriate or maladaptive attributions of salience.  For instance, polymorphisms of the D4 receptor that lead to reduced transcription and expression are associated with both substance and behavioral addictions, attention-deficit-hyperactivity-disorder (ADHD) and schizophrenia (Comings et al. 2001; Comings et al. 1999; Demiralp et al. 2007; Tarazi and Baldessarini 1999). 66  On the other hand, despite decades of research, a definitive role for the D4 receptor in cognitive processes has yet to identified (see Oak et al. 2000 for discussion).   Caution should be taken not to overstate these results, as our autoshaping sample included very few animals that displayed a GT phenotype.  Previous reports have indicated that animals either develop ST, GT or vacillate between these two behaviours with roughly equal probabilities (Flagel et al. 2009).  However, the presentation of these behavioural phenotypes varies widely as a function of both intra- and inter-vendor variability (Fitzpatrick et al. 2013).  Intriguingly, visual inspection of the graph suggests that PD168077 may have decreased head entries into the food tray and increased lever contacts in animals displaying a GT phenotype, although the statistical analyses were too underpowered to detect any reliable change.  Whether the D4 agonist could selectively facilitate a switch to sign-tracking in animals exhibiting a strong GT phenotype may warrant additional investigation. However, reanalyzing the data using less stringent criteria for delineating GT and ST did not reveal any underlying drug effects obfuscated by the small groups.  Relatedly, the use of a single dose of the D4 agonist is somewhat problematic for conclusively eliminating any role for D4 receptors on these tasks. However, the dose was specifically chosen as it had previously been shown to alter behavioral responses to incentive stimuli without producing profound motor deficits (Cocker et al. 2014). As animals did begin to demonstrate motor impairments here, we can be relatively confident the dose was, at least partially, efficacious, yet no behavioural alterations in cue-driven behavior were observed. Significantly higher doses would likely not only induce motor deficits, but also perhaps result in non-selective agonist effects; although PD168077 is highly selective for D4 receptors, it does show affinity for D1- and other D2-like receptors at higher concentrations (Glase et al. 1997). Conversely, systemic administration of lower doses has repeatedly been shown to be without 67  effect on a variety of other behavioural paradigms that require animals to respond to incentive stimuli, and thus we can be reasonably confident that the dose chosen was appropriate, whilst remaining relatively selective (Cocker et al. 2014; St Onge and Floresco 2009; Yan et al. 2012).  Ultimately however, before a role for D4 receptors on either of these tasks is categorically eliminated, the effects of various doses of both agonists and antagonists administered during acquisition may be worth examining.  Certainly the acquisition stage appears to depend on areas containing D4 receptors, such as frontal cortical areas, whereas expression of these behaviours is predominantly mediated by striatal regions (Chudasama and Robbins 2003; Taylor and Robbins 1984; 1986).  Indeed, dopamine D4 receptors are unique amongst the D2-like receptors in that they are not principally expressed in the striatum, but rather located within areas that critically contribute to decision making, including prefrontal cortical regions and the amygdala (Defagot and Antonelli 1997; Mrzljak et al. 1996; Wedzony et al. 2000). Previous reports have intimated that D4 receptors within prefrontal regions may be acting to stabilize representations of reward (Cocker et al. 2014). D4 receptors are well placed to modulate local activity within these regions, in that they are located on both pyramidal neurons within superficial layers II/III and GABAergic interneurons (Mrzljak et al. 1996) but not within efferent layers V and VI (Boyson et al. 1986; Vincent et al. 1993) the output pathway by which frontocortical areas modulate striatal activity (Berendse et al. 1992). Perhaps, therefore, the failure of PD168077 to effect performance on either autoshaping or CRf should not be too surprising as, beyond initial acquisition, neither task appears to be frontally mediated.   Instead, both tasks remain sensitive to manipulations of dopaminergic activity within the NAc regions (Biesdorf et al. 2015; Chudasama and Robbins 2003; Day et al. 2006; Hall et al. 68  2001; Taylor and Robbins 1984; 1986). The NAc receives extensive and overlapping coticolimbic afferents from the hippocampus, PFC, amygdala, association cortex and thalamus, amongst others, and thus is uniquely positioned to integrate reward-related information and mediate incentive stimulus-response learning (Kelley and Domesick 1982; Kelley and Stinus 1984). Yet, the upstream inputs that might be important for driving behaviour during the performance (rather than the acquisition) of tasks such as autoshaping or CRf remain elusive (Everitt and Wolf 2002).  Responsivity to incentive stimuli on these tasks has been argued to be due to activation of the mesocortical afferents to the NAc augmenting the salience of the CS, and also mesolimbic projections to the striatum increasing psychomotor activation in response to reward-related stimuli (Everitt and Robbins 2005). Given the D4 receptor’s distribution within both prefrontal cortical and amygdalar regions (Defagot and Antonelli 1997; Mrzljak et al. 1996; Wedzony et al. 2000), and the failure of systemically administered D4 ligands to alter behaviour here, it would appear that D4-mediated dopaminergic activity within prefrontal or amygdala regions does not appear to contribute to this process.   Dopamine plays a vital role in signaling the predictive quality of environmental stimuli (Schultz 1998) and dopaminergic signaling has consistently been shown to underlie behaviour on both autoshaping and CRf.  Systemic administration of agents that potentiate the actions of dopamine such as amphetamine, apomorphine and cocaine analogues enhance responding for CS+ during CRf (Robbins et al. 1983).  Mixed dopamine D2-like agonists such as pipradol, quinpirole and bromocriptine also increase CRf responding (Fletcher and Higgins 1997; Robbins et al. 1983) and D2-like antagonists such as pimozide and ɑ-flupenthixol block associative learning about a CS+ during CRf  (Beninger and Phillips 1981; Fletcher and Higgins 1997; Sutton et al. 2001). In addition to the effects at the D2-like receptors, D1 receptor agonism has 69  been shown to impair acquisition of a CRf response.  Thus, broadly speaking, activation of D1-like receptors could be considered to impair, whereas D2-like activation enhances, the acquisition of CRf response (Beninger and Ranaldi 1992).  During autoshaping, systemic pharmacological studies have similarly revealed a critical role for D2-like receptors, but, in contrast to CRf, both agonists and antagonists impair the acquisition of a ST response (Danna and Elmer 2010; Fraser et al. 2016; Lopez et al. 2015). Generally, ST is considered to represent a more dopamine dependent behavior, as it involves an otherwise neutral stimuli acquiring incentive properties (Morrison et al. 2015).  These data indicate that signaling through D2-like receptors is crucial for attributions of salience to CS+ during both CRf and autoshaping.  The D2-like receptor family includes D2, D3 & D4 receptors.  We have shown here that D4 receptors do not appear to contribute to the expression of either autoshaping or CRf.  Presumably therefore, one of the other sub-types may play a more prominent role.  In this regard D3 receptors appear ideally placed, given that these receptors are densely localized both pre and post-synaptically within the NAc (Bouthenet et al. 1991). D3 receptors have been suggested to play a key role in the manifestation of human gambling (see Boileau et al. 2015 for review) and have been repeatedly implicated in cue-driven reinstatement of drug seeking in animal models (Beninger and Banasikowski 2008; Le Foll et al. 2005).  However, with regards to autoshaping, mixed results have been observed with the partially selective D3 agonist SB-277011A and the more selective antagonist 7-OH-DPAT did not alter performance (Fraser et al. 2016; Sutton et al. 2001), intimating that D3 receptors cannot exclusively account for responsivity to CS+ on these simpler behavioural tasks.  The limited effects of D3 or D4 receptor modulation are interesting given that selective targeting of these receptors mediates alterations in salience attribution to reward-related stimuli 70  during complex operant tasks (Barrus and Winstanley 2016; Cocker et al. 2014). Indeed, this study was, in part, motivated by recent findings on the rSMT that suggested D4 receptors are critically involved in mediating attributions of salience to reward-related stimuli during gambling-like decision making (Cocker et al, 2016; Cocker et al, 2014). Consequently, we have speculated that targeting these receptors could represent a potential pharmacotherapy for GD (Cocker et al, 2015). However, the failure of D4 agonism to alter behaviour on either autoshaping or CRf, tasks designed to capture aspects of incentive motivation, could be considered incongruent with our previous findings. During complex behavioural tasks, such as the rSMT, animals have to integrate reward-related information from multiple sources to guide behaviour. Additionally, animals retain a representation of the outcome of the previous trial that exerts considerable influence over subsequent choices (Cocker et al. 2016; Silveira et al. 2016) and behaviour on such tasks, remains goal-directed, even after extensive training (Winstanley et al. 2011). D4 receptor activity has been suggested to contribute to maintaining homeostatic PFC activity levels (Gu and Yan 2004; Yuen et al. 2010), theoretically stabilizing representations of reward (Cocker et al. 2014; Cocker and Winstanley 2015). Consequently, D4 ligands may contribute to tasks wherein there is a higher cognitive load and responses to incentive stimuli is more purposive, but these receptors do not appear to be engaged during simple behavioural tasks.  As such, the efficacy of the rSMT to model addictive disorders, which are putatively governed by the transition from goal-directed to habitual control over drug seeking, is questionable (Everitt and Robbins 2005).  However, the converse argument should perhaps also be considered, in that whilst simple tasks such as autoshaping and CRf may offer insight into an animals’ tendency to approach a stimulus associated with reward, it is far from clear to what degree these tasks offer meaningful insight into the phenomenology of addictive disorders.  Although the etiology of 71  differing addictions are likely to be varied, both impulsivity and dysfunctional decision making are perhaps the most ubiquitous behavioural dysfunctions exhibited by human addicts (Bechara 2005).  Individuals with either substance or behavioural addictions consistently demonstrated impaired cost-benefit decision making on laboratory tasks designed to probe ‘real-world’ decision making, such as the Iowa Gambling Task (IGT) (Bechara et al. 2001; Cavedini et al. 2002; Power et al. 2012; Roca et al. 2008).  A recent study showed that whilst D3 receptor modulation did not affect decision making on a rodent version of the Iowa Gambling Task (rGT) (Di Ciano et al. 2015), it did alter performance on a modified version of the rGT, wherein riskier choices were associated with audiovisual cues (Barrus and Winstanley 2016).  These data would indicate that the addition of cues augmented the salience of risky options, consistent with a role for D3 receptors in cue driven drug seeking (Beninger and Banasikowski 2008; Le Foll et al. 2005), but incongruent with the limited role suggested for D3 receptors in the autoshaping data (Fraser et al. 2016; Sutton et al. 2001).  Furthermore, we have preliminary evidence that suggests animals’ instrumental motivation for cues on a CRf paradigm does not correlate with performance on either the regular or cued version of the rGT (Tremblay, Ferland, Hounjet and Winstanley unpublished observations).  As such, sensitivity to reward-associated cues on simple behavioural tasks may not predict perturbations in cost-benefit decision making.  Impulsivity is a non-unitary construct (Evenden 1999) and is likely a precipitating vulnerability in the manifestation of both substance and behavioural addictions (Leeman and Potenza 2012; Potenza 2007). Yet there is little evidence suggesting that cue reactivity and impulsivity are meaningfully related. Animals bred for high locomotor reactivity in a novel environment show an increased propensity to self-administer cocaine and increased cocaine intake (Piazza et al. 1989b).  Animals with this high responsive phenotype typically develop a 72  ST response during autoshaping, but interestingly show lower levels of impulsivity on a delay discounting procedure in comparison to low reactive animals (Flagel et al. 2010). Relatedly, a recent study has demonstrated that animals with high levels of motor impulsivity, as measured by premature responding on the 5-choice serial reaction time task (5-CSRTT), did not show elevated responding for the CS+ during CRf.  Indeed, animals who showed the lowest levels of impulsivity were actually the most willing to work for the CS+ (Zeeb et al. 2016).  It would therefore appear that responding to (ST) and for (CRf) cues does not track impulsivity, yet may track reactivity to novelty.  However, elevated impulsivity predicts compulsive drug taking, indicative of a more addiction-like behavioural response, whereas novelty reactivity only predicts more rapid acquisition of cocaine self-administration (Belin et al. 2008; Flagel et al. 2010). Thus increased cue reactivity on CRf and autoshaping, a putative biomarker for addiction vulnerability, does not appear to meaningful relate to measures of impulsivity or decision making, canonical measures of dysfunction in addiction. Ultimately, these data provide novel insight into the role of D4 receptors in controlling responsivity to reward-related stimuli, in that acute agonism does not increase sensitivity to food-associated cues, despite modulating performance of the rSMT.  The disparity between these results and previous datasets may suggest that placing too much emphasis on modelling addiction vulnerability with very simple cue-driven behaviours may result in both false positives and negatives with respect to treatment development, in that the pharmacological regulation of ST and CRf does not reflect that governing more cognitively challenging tasks with greater face validity to problematic behaviour seen in GD.  Unfortunately, given the lack of effective treatments for behavioural or chemical addictions beyond drug substitution, true validation of animal models of addiction vulnerability remains elusive.  The diversity within behavioural tasks 73  may therefore be invaluable, particularly given the heterogeneity within both substance and behavioural addictions, and the recent suggestion that different behavioural phenotypes (Pavlovian approach, procedural learning, deliberative decision making) may provide dissociable yet equally important routes into the addicted state (Redish et al. 2016). Elucidating the basis of individual aspects of behavioural perturbations is therefore likely to be more achievable, and beneficial from a treatment standpoint, then assuming a single pathogenesis controlling an entire disease state.   Figure 4.1 Summary of goal- and sign-tracking responses during training and following PD168077 administration (1a) Animals designated as sign trackers (n=6) showed a far greater propensity to contact the lever associated with the CS+, rather than approach the food delivery tray. This behavioural phenotype became progressively more pronounced throughout the training sessions. (1b) In contrast animals designated as goal trackers (n=3) were far more likely to make a nose-poke response in the food delivery tray, this phenotype was established early and 74  remained relatively stable. (1a,b) Administration of the D4 agonist PD168077 did not affect animals response to either the lever or the food tray.    Figure 4.2 Head entry responses during training sessions during CRf training. As training progressed animals learned the contingencies of the task and made significantly more head entries into the food tray during presentation of the CS+ in comparison to the CS-.   75   76  Figure 4.3 Summary of behavioural responses on single CRf test day following administration nof either saline or PD168077. (3a) Administration of the D4 agonist did not appear to alter the salience of the CS+ as the ratio of responses on the lever that lead to the CS+ was no different to responses on the CS-. However, PD168077 did lead to a general decline in the total number of nose poke responses both of the (3b) CS+ lever, (3b) the CS- lever and (3d) consequently therefore total lever responses. (3e,f) There were no significant differences in the animals’ nosepoke responses either during or after presentations of the CS+.                  77  Chapter 5: Experiment 3: Activation of dopamine D4 receptors within the anterior cingulate cortex enhance the erroneous expectation of reward on a rat slot machine task.    5.1 Introduction Gambling is an enjoyable and innocuous pastime for many, but for some it can develop into a maladaptive compulsion akin to drug or alcohol abuse (see Potenza 2006; 2008 for review). Cognitive theories of gambling suggest that cognitive biases and distorted beliefs regarding decision-making under risk or uncertainty contribute to the transition from recreational to problem gambling (Clark 2010; Ladouceur et al. 2001; Sylvain et al. 1997).  “Near-misses” are one such bias that has garnered considerable attention. Near-misses can occur throughout the panoply of gambling, but are particularly prevalent in slot-machines, where they may contribute to the disproportionate levels of harms associated with slot-machine engagement (e.g. Breen and Zimmerman 2002; see Murch and Clark 2015 for review). Near-misses reliably galvanize game play (Clark et al. 2009; Cote et al. 2003; Kassinove and Schare 2001), theoretically because they foster expectations of an imminent win (Reid 1986). As treatment options for gambling are currently limited (Grant and Kim 2006; Ladouceur et al. 2001), a better understanding of the neurobiological mechanisms underpinning the near-miss bias may stimulate the development of novel therapeutics.  Using the rSMT, we have recently demonstrated that rats, like humans, are susceptible to the reinforcing effects of putative winning signals that occur within the context of an objective 78  loss trials (Winstanley et al. 2011). We have argued that this constitutes a translational analogue of the human near-miss effect. Our previous findings indicate that the expectation of rewarding outcomes on the rSMT is critically modulated by the dopamine D2-like receptor family, particularly the D4 receptor (Cocker et al. 2014; Winstanley et al. 2011).  The selective D4 antagonist L-745,870 improved, whereas the selective D4 agonist PD168077 impaired animals’ ability to differentiate winning from losing outcomes. These results were obtained using systemic administration of these ligands, such that the neural locus of these effects is as yet unclear. D4 receptors are predominantly located within frontocortical regions, including areas that have been implicated in reward-based decision making and slot-machine play such as the anterior cingulate cortex (ACC) (Khan et al. 1998; Liu et al. 2011; Mrzljak et al. 1996).   The exact role of the ACC in decision making is contentious, yet performance errors and cognitive conflict have been shown to reliably activate ACC (Brown and Braver 2005; Hyman et al. 2013), and also in response to stimuli signaling a necessity to change a response pattern (Bush et al. 2002). Thus the ACC appears to be broadly engaged in signaling error likelihood and consequently guiding decision making (Kennerley et al. 2006).  The ACC is therefore an interesting area to investigate in relation to disordered gambling, a condition fundamentally characterised by maladaptive reward-based decision making (Cavedini et al. 2002; Lawrence et al. 2009).  ACC activity increases in healthy human subjects experiencing near-misses during slot machine play (Clark et al. 2009), and reduced activity was seen in the same region in response to gambling cues in individuals with GD (Potenza 2008).  These findings collectively suggest that ACC dysfunction may contribute to pathological gamblers’ aberrant responses to reward-related stimuli, including near-misses (Brand et al. 2005; Habib and Dixon 2010). The ACC in rodents can be considered broadly homologous to those in humans and primates based 79  upon location, cytoarchitecture, laminar pattern, receptor binding and local connectivity (Vogt et al. 2013), making rodent models a valuable assay for assessing the function of this region in regards to GD. The ACC receives dense dopaminergic projections, and D4 receptors are highly expressed within this region (Lewis 1992; Westlund et al. 1990). The goals of this study were therefore to determine whether selective inactivation of the ACC, and targeted administration of a D4 agonist, would augment erroneous behavioural responses to reward related stimuli on the rSMT.   5.2 Additional methods  Surgery Animals were deeply anaesthetized using 2% isoflurane in O2 and implanted with bilateral 22-gauge stainless steel guide cannula (Plastics One; Roanoke, VA) aimed at the ACC (incisor bar set to -3.3mm (flat skull) anteroposterior +2.0mm, mediolateral ± 0.7mm, dorsoventral -1.2mm from dura) using standard stereotaxic techniques.  Cannula were fixed to the skull with three stainless steel screws and dental acrylic.  29-gauge obturators (flush with guides) were inserted and covered with a dust cap to protect the infusion site.  Analgesic agents were given as standard via subcutaneous injection (5 mg/kg anafen; 2.5mg/kg bupivacaine). Animals were allowed to recover for at least one week following surgery before testing resumed.   Microinfusions The ACC was inactivated using a mixture of the GABAA agonist muscimol and the GABAB agonist baclofen (0.125µg of each compound in 0.5µl) administered bilaterally at a rate of 0.25µl/minute.  Infusions were structured such that on the first infusion day half the animals 80  received either saline or baclofen-muscimol; these administrations were reversed on the second infusion day, allowing for a within-subjects design.  Animals were tested drug free for a week to allow a stable baseline to re-establish and to minimise any potential carry over effects of the baclofen/musimol infusions.  Animals were subsequently infused with the D4 agonist PD168077 (0, 0.5, 1.0 & 5.0µg/side) administered according to a digram balanced Latin square design. In summary, each rat received a total of 6 infusions, initially baclofen/muscimol and saline followed by a washout period before saline and the three doses of PD168077.   Pharmacological challenges All drugs were prepared fresh daily.  Doses of PD168077 (Tocris, Ellisville, MO) were calculated as the salt and dissolved in 0.9% sterile saline.  Baclofen and muscimol (Sigma Aldrich, Oakville, Canada) were prepared separately at 0.5µg/µl in saline and mixed together in equal volumes to create a 0.25µg/µl solution.  5.3 Results Four animals were excluded from the analysis due to failure to meet established performance criteria, defined as at least 50 trials completed per session and at least 50% accuracy on clear loss trials ([0,0,0]) (Winstanley et al. 2011).  Five rats experienced post-surgical complications and were euthanized prior to completion of the experiment. Cannulae tips were located within the ACC in 11 out of 14 rats (figure 5.1), and only data from these animals was included in subsequent behavioural analyses.  Data from the 3 excluded animals was analysed separately to 81  act as an anatomical control region; of the three excluded animals, 2 had placements that were too rostral and one was unilateral.    Baseline behaviour  Similar to previous reports, collect responses varied significantly across different trial types (figure 5.2a: trial type: F7,70 = 62.57, p<0.0001). The likelihood of erroneous responses on the collect lever increased with the number of illuminated apertures in the array (figure 5.2b: lights illuminated: F3,30 = 195.64, p<0.0001, 3 vs 2: F1,10 = 294.71, p<0.0001, 2 vs 1: F1,10 = 220.67, p<0.0001, 1 vs 0: F1,10 = 28.58, p<0.0001). There was no difference between the different types of 2-light or 1-light trials, suggesting that the total number of lights illuminated facilitated erroneous responding, rather than the exact spatial position of the lights (trial type: 2-light trials F2,20 = 2.23, NS, 1-light: F2,20 = 2.76, NS). The latency to respond on the collect lever also varied significantly between different trial types (table 5.1; trial type: F7,70 = 9.29, p<0.0001), and was predominantly driven by faster responding on trials in which few lights were illuminated within the array. There was no difference between response speed on wins and near-miss trials, but erroneous collect responses became progressively quicker for both one-light and zero-light trials (lights illuminated: 3 vs 2, F1,10 = 0.33, NS: 2 vs 1, F1,10 = 6.85, p=0.026, 1 vs 0: F1,10 = 32.56, p<0.0001). These findings indicate that responding on the collect lever following 1- and 0-light trials may have resulted from disinhibited or impulsive responding.  Animals’ responses to subsequent holes were quicker if the preceding hole had set to on (table 5.2; light status: F1,10 = 5.46, p=0.042), suggesting that illuminated holes served to invigorate responding in keeping with the hypothesis that these cue lights are assigned incentive salience.  82  The likelihood that animals would respond on the collect lever on subsequent trials was also affected by the status of the preceding trial (trial type: F7,70 = 17.35, p<0.0001). Examining this effect by the number of lights illuminated, animals were more likely to collect on the trial following a win, and less likely to collect if the preceding trial was a near-miss, whereas both 1- and 0-light trials had similar effects on subsequent choice (Figure 5.5a: lights illuminated: F3,30 = 21.35, p<0.0001; 3 vs 2; F1,10 = 33.73, p<0.0001; 2 vs 1: F1,10 = 6.68, p=0.03; 1 vs 0: F1,10 = 1.60, NS). The status of the preceding trial also affected the speed at which animals began a new trial (trial type: F7,70 = 17.35, p<0.0001).  Animals were quicker to initiate following a win and slower to initiate following 0-light trials, but there was no difference in the time taken to begin a new trial between 2- and 1-light trials, in contrast to the effect on lever choice (figure 5.5b: lights illuminated: F3,30 = 27.02, p<0.0001; 3 vs 2: F1,10 = 24.82, p=0.001; 2 vs 1: F1,10 = 0.89, NS; 2 vs 0: F1,10 = 13.12, p=0.005; 1 vs 0: F1,10 = 18.11 = 0.002). Animals completed an average of 80.95 ± 2.79 trials per session (table 5.3). There was no correlation between animals’ weight and either erroneous lever choice (r2 = 0.02, NS) or the number of trials completed (r2 =0.02, NS).   Effects of ACC inactivation Temporary inactivation of the ACC resulted in a significant increase in erroneous collect responses (figure 5.3a: drug X trial type: F7,70 = 2.31, p=0.036; figure 5.3b: drug X lights illuminated: F3,30 = 4.12, p=0.014), driven by differences on 1- and 0-light trials (3-light trials: F1,10 = 0.67, NS, 2-light trials: F1,10 = 0.02, NS: 1-light trials: F1,10 = 8.27, p=0.02; 0-light trials: F1,10 = 7.567, p=0.02).  Thus ACC inactivations impaired animals’ performance only on trials in which few winning signals were present within the array, when animals would normally choose optimally. These effects were not present in the control group (drug X trial type: F7,14 = 2.03, NS; 83  drug X lights illuminated: F3,6 = 0.68, NS) and were not evidently contingent on the dorsal – ventral aspect of the cannula placement (r2 (9) = 0.003, NS).  ACC inactivations did not affect the time taken to choose the collect lever on any trial type (table 5.1; dose X TT: 7,70 = 0.87, NS). However, the tendency seen at baseline for animals to respond more quickly on the collect lever on trials wherein fewer lights were illuminated was reduced to a trend (lights illuminated: F3,30 = 2.59, p=0.07).  ACC inactivations ameliorated the tendency to respond quicker at the array if the preceding aperture was illuminated (table 5.2; light status: saline; F1,10 = 14.63, p=0.003, BacMus; F1,10 = 2.55, NS), suggesting that inactivations may be blunting animals’ sensitivity to reward-related signals within the array. Inactivating the ACC did not alter the impact previous trial outcomes had on the probability of collect lever choice (drug X trial type: F7,70 = 0.69, NS; figure 5.5c: drug X lights illuminated: F3,30 = 1.21, NS), nor did inactivations affect the latency to begin a new trial following different trial types (drug X trial type: F7,70 = 1.61, NS; figure 5.5d: drug X lights illuminated: F3,30 = 0.02, NS). Lastly, there were no significant effects on the number of trials completed (table 5.3; dose: F1,10 = 2.56, NS).    Effects of PD168077 Similar to inactivation of the ACC, local administration of a D4 agonist produced an overall impairment in animals’ optimal lever choice that reached statistical significance at the highest dose (figure 5.4a, b: dose: F3,30 = 3.15, p=0.039; dose- saline vs 0.5mg/kg: F1,10 = 2.16, NS; saline vs 1.0 mg/kg: F1,10 = 0.84, NS; saline vs 5.0mg/kg: F1,10 = 5.33, p=0.044).  Unlike ACC inactivations, this pattern of behaviour did not appear directly related to effects on multiple trial types or even the total number of lights illuminated (dose X trial type: F21,210 = 099, NS; figure 5.4b: dose X lights illuminated F9,90 = 1.61, NS) but rather the spatial positioning of the lights. 84  Animals were more likely to press the collect lever on non-winning trials if the last light set to off (dose: F3,30 =3.39, p=0.03; last light on: F1,10 = 0.29, NS; last light off: F1,10 = 5.12, p=0.047).  This effect was also predominantly driven by those trials where the animal initially experienced two consecutive winning signals, followed by a single negative predictor at the end of the array: the archetypal near-miss (trial type [1,1,0]: dose: F1,10 = 10.21, p=0.01; all other trial types: dose1,10 NS, p> 0.08).  Thus the D4 agonist appeared to diminish animals’ ability to modify their choice in response to a negative event temporally proximal to the decision point if such an event had been preceded by positive predictors of reward. Similar to ACC inactivations these perturbations were not evident in the control group (dose: F3,6 = 1.71, NS; dose X trial type: F21,42 = 0.89, NS) and could not be attributed to cannula location along the dorsal – ventral axis (saline vs 0.5ug: r2 (9) = 0.01, NS; saline vs. 1.0ug: r2 (9) = 0.16, NS; saline vs. 5.0ug: r2 (9) = 0.0001, NS). PD168077 administration did not affect latency to choose the collect lever (table 5.1: dose X trial type: F21,210 = 0.81, NS). Similarly, the tendency of animals to respond quicker at the subsequent hole if the preceding hole was illuminated was still evident following intra-ACC administration of the D4 agonist (table 5.2: light status: F1,10 = 18.20, p=0.002; dose X light status: F3,30 = 1.49, NS). Consistent with the effects of ACC inactivations, administration of PD168077 did not lead to changes in subsequent collect lever choice (figure 5.5e: dose X trial type: F21,210 = 0.88, NS: dose X lights illuminated: F9,90 = 0.68, NS) or latency to begin a new trial (figure 5.5f: dose X trial type: F21,210 = 1.11, NS; Fig 6e: dose X lights illuminated: F9,90 = 0.44, NS) dependent on the preceding trial type. There were also no significant effects of PD168077 on the number of trials animals competed (table 5.3: dose: F3,30 = 0.56, NS).  85  5.4 Discussion Here we show that the ACC guides behavioural responses to reward-related stimuli in a rodent model of slot machine gambling. Inactivation of this region decreased animals’ ability to differentiate winning from losing outcomes, increasing erroneous attempts to collect reward on both 1- and 0-light trials. Critically, our data also indicate a novel role for D4 receptors within the ACC in gating responses to reward-salient stimuli. In a manner that is qualitatively similar to the inactivation findings, infusion of the D4 agonist PD168077 also increased erroneous expectations of reward on non-winning trials, but in contrast these effects were predominantly evident when the first two lights in the array set to on but the final light set to off. This specific trial type is most analogous to the classic near-miss instances in human gambling, where the expectation of reward is initially generated (e.g. two cherries on the first two reels) but then frustrated by the final event in a sequence.  Collectively, these data further implicate the ACC in mediating choice in complex discrimination tasks, and highlight a specific role for local D4-mediated signaling in the temporal and/or spatial summation of reward prediction signals.  The present study also included some novel analyses to examine how the outcome on the preceding trial influences subsequent behaviour on this task. The finding that winning outcomes increase the likelihood of making future errors could be explained by an increase in Pavlovian approach behaviour to the collect lever as a result of the temporal proximity of previous reward.  Conversely, the fact that near-misses had the opposite effect, decreasing the likelihood of erroneous collect lever choice beyond that observed after the other two classes of non-winning trials, may indicate that rats find the outcome of these trial types particularly aversive, as also seen in subjective ratings taken in human participants (Clark et al. 2009), leading to more effective inhibition of subsequent collect lever choice.  The effects of the previous trial on the 86  latency to begin a new trial are also broadly in keeping with recent human work suggesting that wins increase the speed of play to a greater degree than near-misses or loss trials (Shao et al. 2013). Interestingly, both inactivations and PD168077 appeared to affect animals’ performance acutely, insofar as administration of either substance affected animals’ likelihood of making erroneous collection responses on the current trial, but did not affect animals’ performance on subsequent trials as a function of the preceding trial. One interpretation of these data is that, although both manipulations affect animals’ ability to integrate salient information to optimally guide behaviour within each trial, neither affected the more diffuse impact that previous trial outcomes had on global levels of motivation.  Neither of these intra-ACC manipulations significantly altered the number of trials animals completed or the time taken to respond on the collect lever, further suggesting that the effects on choice did not reflect a motivational or locomotor impairment.  Collectively, these data indicate that the mechanism underpinning animals’ ability to correctly determine the outcome signaled by reward-related stimuli can be dissociated from that regulating the impact such outcomes have on task engagement.   A limitation of the study that should be acknowledged is that the repeated use of ANOVA’s to characterise the deficits seen following PD168077 administration may have increased the risk of a type I error. Follow-up analyses were only performed in response to a significant main effect, and given the large number of trial types, the use of post-hoc tests would have raised the criterion needed for statistical significance by more than an order of magnitude. The rSMT requires extensive training and such a robust statistical criterion would likely only capture global impairments in behaviour, as oppose to more subtle alterations in response to acute manipulations. Nevertheless, the lack of statistical rigor should be acknowledged and experiments aiming to replicate this finding would be beneficial.  87  The rSMT punishes animals for erroneous responses with a 10-second time-out punishment. The duration of the time-out was optimised using pilot studies and was designed to be sufficiently aversive to the animals without prompting them to disengage with the task. There is ample evidence to suggest that rats not only find time-out punishments aversive, but are also highly attuned to different temporal punishments associated with various rates of reinforcement (Richardson and Baron 2008; Zeeb et al. 2009). Additionally, the loss of opportunity to earn reward is commonly used in animal tasks and has been argued to represent a reasonable facsimile of comparable laboratory based tasks in humans (Cocker and Winstanley 2015). Thus is unlikely that the observed deficits here following either manipulation could have arisen as a result of an increase in the negative valence attributed to time-out punishments, as tasks that are dependent on the subjective appraisal of time are unaffected by either ACC lesions or non-selective D2 agonists (Floresco et al. 2008).  There is a demarcation of function within the frontal cortex along the rostral – dorsal axis (see Grabenhorst and Rolls 2011 for review), and as such, these results could have arisen via diffusion of the infusion agents into adjacent regions such as the prelimbic cortex. However, pharmacological inactivations encompass a relatively small brain area (~1mm) (Floresco et al. 2006a; Marquis et al. 2007), and thus infusions into the ACC are unlikely to have significantly spread to other regions. Moreover, there was no correlation between any of the observed effects and cannula placement.  Given the multi-faceted nature of impulsivity (Evenden 1999), and the prominent relationship between high impulsivity and GD (Potenza 2007), it may be worth considering whether poor choice on the rSMT reflects an aspect of impulsive decision-making.  Perhaps notably, ACC inactivations increased erroneous collection responses predominantly on 1- and 0-light trials, responses that at baseline are made more rapidly and may therefore be due at 88  least in part to disinhibited or impulsive responding (Cocker et al. 2014; Winstanley et al. 2011).  ACC dysfunction in humans has been hypothesized to underlie adolescent impulsivity (Eshel et al. 2007) and there have been reports that ACC lesions increase motor impulsivity on rodent tests of visuospatial attention but see (Hosking et al. 2014a; Muir et al. 1996), (but see (Chudasama et al. 2004)). However, neither ACC inactivations nor D4 agonism speeded choice of the collect lever, implying that an increase in motor impulsivity is unlikely to account for the observed deficits.  Furthermore, recent data suggest that high trait impulsivity may facilitate slot machine play by enhancing the neural response to wins rather than exacerbating the near-miss (Shao et al. 2013).  Combined with the null effects of ACC lesions on delay-discounting tasks (Cardinal et al. 2001), there is little evidence to link the ACC’s contribution to choice on the rSMT with the construct of impulsivity at this time.   Perhaps the most parsimonious conclusion is that during the rSMT, the ACC acts to constrain maladaptive reward-seeking responses by integrating information signaled by multiple sources into a prediction regarding reward availability on each trial.  Contemporary theories suggest a more nuanced role for the ACC in monitoring performance and guiding behaviour, beyond only action-selection or conflict monitoring (Holroyd and McClure 2015; Rushworth et al. 2007). The ACC appears to integrate multiple aspects of environmental information, including environmental volatility, past reward history, error likelihood, & current performance (Bush et al. 2002; Hyman et al. 2013; Kennerley and Wallis 2009; Kennerley et al. 2006; Magno et al. 2006). This diverse range of functions suggests that the ACC is integrally (albeit broadly) involved in maintaining the long-term value of predicted outcomes and informing optimal goal-orientated decisions.  For example, the ACC appears critical in prompting individuals to 89  disengage from a current strategy and alter behaviour in response to changing task demands (Kennerley et al. 2006; Shenhav et al. 2014).   In the rSMT, pressing the collect lever is the only response, out of the many made on-task, that ever results in reward delivery.  Hence, presentation of the collect lever likely arouses substantial instrumental and Pavlovian incentive salience, pre-disposing animals to approach and respond.  Thus optimal performance of the task requires the subject to prioritise the integration of information provided by the stimulus light array in order to constrain the appetitive valence signaled by the collect lever.  On 1- and 0-light trials, the aperture array and lever yield conflicting information about reward availability, decreasing the incentive motivation to respond on the collect lever and facilitating selection of the optimal roll lever.  Inactivation of the ACC increases erroneous collect responses on such trials, leading us to infer that the signal provided by the ACC contributes to over-riding the pre-potent urge to respond on the collect lever on such trial types.  On 2-light trials, the multiple putative winning signals within the array may act synergistically with presentation of the collect lever to promote the representation of available reward.  The failure of ACC inactivations to increase the already high levels of collect errors is unlikely to have arisen as a result of ceiling-effects, as we and others have previously demonstrated that behavioural response rates of ~80% can reliably be increased through pharmacological manipulation (Cocker et al. 2012; Cocker et al. 2014; St Onge and Floresco 2009; Winstanley et al. 2011).  Furthermore, these data suggest that the output signal generated by the ACC is concordant with that of other areas involved in the decision-making process, such as the ventral striatum or basolateral amygdala (Shao et al. 2013), the net impact of which is to promote responding on the collect lever on the majority of 2-light trials.   90  However, the observation that a D4 agonist administered intra-ACC can amplify errors on a subtype of 2-light trials indicates that the ACC signal can be altered to enhance erroneous reward expectations, but does so in a manner that is distinct form the effects seen following inactivations.  The ACC receives heavy dopaminergic innervation (Gaspar et al. 1989) and computational models have previously highlighted a role for dopamine in facilitating stimulus-outcome associations within this area (Brown and Braver 2005).  The effects of intra-ACC D4 agonist administration only reaches significance on archetypal near-miss trials in which the first two stimulus events are concordant with a winning outcome ([1,1,0]).  It would therefore appear that D4 receptor-mediated transmission is somehow specifically involved in emphasising the temporal or spatial proximity of reward-predictive stimuli.  Dopamine neurons within the mid-brain fire to conditioned stimuli that predict reward (Fiorillo et al. 2003).  In the rSMT, the degree to which illumination of a given aperture accurately predicts that reward will be delivered fundamentally depends on the status of both the previous and subsequent stimulus lights in the array.  Consequently, temporal summation and integration of stimulus events is critical in informing lever choice.  As the stimuli are delivered sequentially, it is conceivable that two lit holes in direct succession at the start of the array would be particularly effective at potentiating the firing of mesolimbic dopaminergic neurons, leading to increased dopamine release in the ACC.  Activation of local D4 receptors appears to further magnify this prediction of a positive outcome, and could thereby mask the relative dip in dopamine signaling when the final hole sets to off. Relatedly, the trend in increased erroneous responding on clear loss trials ([0,0,0]), is likely a result of low baseline levels of errors coupled with an augmented incentive salience to the collect lever following D4 agonism. Such a theoretical explanation for PD168077’s effects fits with data demonstrating that local infusion of a D4 agonist into the medial PFC impairs set-91  shifting using visual cues, whereas a D4 antagonist improves performance (Floresco et al. 2006b). Additionally, recent studies have shown that prefrontal D4 receptor activation augments the salience of and behavioural impact of otherwise-subthreshold stimuli (Lauzon et al. 2012; Lauzon et al. 2009).  The local distribution of D4 receptors indicates that this receptor subtype likely modulates intra-regional processing, but does not directly influence output.  Although D4 receptors are relatively concentrated in the ACC, they are located on pyramidal neurons within superficial layers II/III as well as GABAergic interneurons (Mrzljak et al. 1996), but not within the efferent layers V and VI (Boyson et al. 1986; Vincent et al. 1993).  Electrophysiological recordings from PFC slices in vitro indicate that a D4 receptor agonist generally inhibits AMPA-receptor-mediated glutamatergic signalling (Yuen et al. 2010), but the net inhibition would be much milder, and more selective, than would result from GABA agonist administration.  The divergent behavioural effects of pharmacological inactivation versus D4 agonist administration highlights the differential impact on decision-making between silencing any signal from the ACC as compared to subtly modulating it via one receptor subtype.   It is worth noting that some studies have suggested lower levels of D4 receptor expression in the human brain relative to the rat (Matsumoto et al. 1996), potentially limiting our ability to make inferences about the importance of D4 receptors in GD. However, D4 receptor polymorphisms in humans have been robustly correlated with both behavioural and chemical addictions (Comings et al. 1999), which would indicate that these receptors are involved in different forms of maladaptive reward-related behaviour. Additionally, polymorphisms that are alleged to confer vulnerability to behavioural addictions are associated with reduced 92  transcription and expression (Asghari et al. 1995; Asghari et al. 1994) such that protein expression may be a relatively poor indicator of importance in behavioural disorders.  These data partially replicate our previous findings that systemic administration of a D4 agonist impairs performance on this task (Cocker et al. 2014). However, the specific impairments seen on the archetypal near-miss trials imply that whilst D4 receptors within the ACC are importantly engaged with this task, other D4 containing brain regions are also likely involved in mediating behaviour. The rostral ACC forms part of a wider network organised around the VMPFC (Barbas and Pandya 1989; Ongur and Price 2000), and hypoactivity in this region may contribute to poor decision-making under uncertainty (Clark et al. 2009; Potenza 2008; Shao et al. 2013).  The PFC appears underactive in problem gamblers (Balodis et al. 2012b; Potenza 2008; Reuter et al. 2005) and this population’s performance on laboratory-based decision-making tasks resembles that of patients with lesions to the VMPFC (Brand et al. 2005; Lawrence et al. 2009).  Pathological gamblers also display augmented attributions of salience to game-related stimuli (Miedl et al. 2014; Moodie and Finnigan 2005).  Our results support the suggestion that under-activation of frontal areas such as the ACC may contribute to this maladaptive focus on reward-related cues in gambling games, potentially due in part to hyper-activation of local D4 receptors.  These findings further indicate that D4 receptors represent a potentially viable target for the pharmacological treatment of GD.      93   Figure 5.1 Histological analysis of cannula. Location of all acceptable ACC infusion sites. Coordinates are relative to bregma, plates modified from Paxinos and Wilson (1998)    Figure 5.2 Baseline rSMT performance (a,b) Animals showed optimal responding on win trials (1,1,1), choosing to collect the available reward nearly 100% of the time.  Similarly, animals showed a marked preference for the optimal response, now the roll lever, when no lights were 94  illuminated (0,0,0), only responding on the collect lever approximately 15% of the time.  Erroneous collect responses increased to 41.56% when 1 light was illuminated.  However, when two-lights were illuminated in the array, animals responded erroneously on the collect lever at a far greater then chance level (81.65% ± 3.95 (SEM)), indicating that rats, like humans, treat such stimuli as more indicative of a win than a loss, and are hence susceptible to the near-miss effect.  All data shown are the mean across five sessions ± SEM.    Figure 5.3 Effects of ACC inactivation on rSMT performance. (a, b) Infusion of baclofen-muscimol into the ACC resulted in impairments in animals’ ability to differentiate winning from non-winning outcomes on 1- and 0-light trials i.e. when few putative winning signals were presented within the array. Data are shown as mean ±SEM. Where conducted the results of follow-up analyses are denoted by * (p<0.05) and # (p<0.1).   95    Figure 5.4 Effect of activation of local D4 receptors within the ACC via infusion of the D4 agonist PD168077 on rSMT performance.  (a, b) The lower and mid-dose of PD168077 did not affect performance on the rSMT, yet the highest dose of PD168077 increased erroneous choice of the collect lever on non-winning trials. In contrast to inactivations these deficits were principally evident on trials when putative winning signals were presented in close spatial proximity in the absence of any preceding negative stimuli i.e. 1,1,0 trials. These trials could be considered analogous with ‘classic’ near-miss trials in human slot machines. All doses are given in µg/µl and data are shown as mean ±SEM. Where conducted the results of follow-up analyses are denoted by * (p<0.05) and # (p<0.1).  96    97  Figure 5.5 Effect of preceding trial type on subsequent task performance. a) Animals show the highest proportion of collect responses on trials that follow a winning outcome and lower levels of collection responses subsequent to 2-light trials.  b) Animals are also faster to begin a new trial following a winning outcome. In contrast, animals display reduced motivation to commence a new trial when there are no putative winning signals presented on the previous trial. c, d) ACC inactivations did not produce any significant differences in either the proportion of collect responses or the latency to begin a new trial as a function of prior exposure to different trial types. e, f) Similar to inactivations, intra-ACC administration of the D4 agonist PD168077 did not produce any changes in animals’ behaviour in regards to collect lever choice or latency to begin a new trial following different trial types.   Table 5.1 Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Baseline  0.61 ± 0.05 0.81 ± 0.11 0.65 ± 0.91 0.61 ± 0.08 0.46 ± 0.09 0.57 ± 0.12 0.55 ± 0.11 0.17 ± 0.04 BacMus 0 1.15 ± 0.38 0.99 ± 0.15 0.84 ± 0.18 0.80 ± 0.23 0.83 ± 0.14 0.74 ± 0.20 0.65 ± 0.10 0.31 ± 0.09  0.125 µg 0.82 ± 0.19 0.96 ± 0.23 0.81 ± 0.17 0.72 ± 0.20 0.87 ± 0.25 1.21 ± 0.65 0.68 ± 0.13 0.66 ± 0.24 PD168077 0 0.94 ± 0.18 0.93 ± 0.18 0.87 ± 0.14 0.84 ± 0.12 0.49 ± 0.12 0.56 ± 0.17 0.61 ± 0.14 0.20 ± 0.09  0.5 µg 1.65 ± 0.77 1.01 ± 0.14 0.76 ± 0.11 0.79 ± 0.12 0.46 ± 0.10 0.53 ± 0.10 0.50 ± 0.10 0.42 ± 0.13  1 µg 1.01 ± 0.17 0.77 ± 0.09 0.70 ± 0.12 0.68 ± 0.11 0.52 ± 0.18 0.25 ± 0.06 0.53 ± 0.27 0.18 ± 0.10  5 µg 0.95 ± 0.18 1.02 ± 0.14 0.87 ± 0.14 0.79 ± 0.14 0.70 ± 0.17 0.53 ± 0.09 0.66 ± 0.32 0.21 ± 0.06  98  Table 5.2 Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Baseline  0.80 ± 0.12 2.51 ± 0.87 0.98 ± 0.17 1.55 ± 0.16 BacMus 0 0.63 ± 0.06 1.33 ± 0.22 0.83 ± 0.10 1.25 ± 0.21  0.125 µg 1.65 ± 0.59 3.60 ± 1.26 1.67 ± 0.29 3.77 ± 1.74 PD168077 0 0.55 ± 0.07 0.83 ± 0.13 0.73 ± 0.07 1.03 ± 0.12  0.5 µg 0.61 ± 0.10 0.78 ± 0.09 0.71 ± 0.11 1.18 ± 0.17  1 µg 0.75 ± 0.10 1.06 ± 0.18 0.93 ± 0.08 2.01 ± 0.66  5 µg 0.63 ± 0.11 1.58 ± 0.44 0.73 ± 0.07 1.29 ± 0.17  Table 5.3 Trials completed at baseline and following pharmacological challenges. Data shown are mean ± SEM Drug Dose  Trials completed Baseline  80.95 ± 2.79 BacMus 0 80.64 ± 7.97  0.125 µg 67.27 ± 6.33 PD168077 0 93.0 ± 7.53  0.5 µg 96.09 ± 7.62  1 µg 91.0 ± 6.27  5 µg 90.09 ± 8.04      99  Chapter 6: Experiment 4: The agranular and granular insula differentially contribute to gambling-like behaviour on a rat slot machine task: effects of inactivation and local infusion of a dopamine D4 agonist on reward expectancy.    6.1 Introduction Gambling is an enjoyable and innocuous past-time for many, but for some individuals it can become a maladaptive compulsion akin to drug or alcohol addiction (Potenza 2006). Within the panoply of gambling games, slot machines represent a particularly virulent form of gambling (Breen and Zimmerman 2002; see Murch and Clark 2015 for review). The deleterious outcomes associated with slot machines are purportedly due to the potency with which these forms of gambling foster cognitive distortions or biases, such as the ‘near-miss’ effect (Clark 2010; Sylvain et al. 1997). Near-misses generate the expectation of imminent reward, consequently galvanizing game play (Cote et al. 2003; Kassinove and Schare 2001).   Using a rodent slot machine task (rSMT), we have previously demonstrated that rats, like humans, are sensitive to win-related stimuli presented within non-winning trials, a phenomenon we have argued is translationally analogous to the near-miss effect (Cocker and Winstanley 2015).  Reward expectancy on the rSMT is sensitive to pharmacological manipulation of the dopamine D2-like receptor, particularly the D4 receptor (Cocker et al. 2014; Winstanley et al. 2011).  However, the neural loci mediating these effects is currently unclear.  To address this issue, we have chosen to focus our efforts on brain regions in which D4 receptor expression is 100  relatively high, and that have been implicated in slot machine play by human imaging studies (Clark et al. 2009; Worhunsky et al. 2014).  The insular cortex is an obvious candidate region in this regard: D4 expression is comparatively robust, and this area has been heavily implicated in mediating addiction-related decision-making across a number of domains in addition to slot machine play (Droutman et al. 2015; Naqvi and Bechara 2009; Rivera et al. 2008). A role for the insula in addictive disorders was first noted following the observation that damage encompassing the insula can lead to immediate and maintained smoking cessation in patients who smoked prior to their injury (Abdolahi et al. 2015; Naqvi et al. 2007; Noel et al. 2013).  Relatedly, damage to the insula abrogates gambling-related cognitions (Clark et al. 2014) and insula activity has been positively correlated with the reinforcing effects of near-miss trials in healthy controls (Clark et al. 2009).  Collectively these data indicate that increased insula activity may confer vulnerability to the development of gambling disorder (GD).  Conversely, subjects with GD show decreased insula activation during anticipation of both rewards and losses (Balodis et al. 2012b).  Similarly, increased insula activity has been observed in healthy individuals during anticipation of losing money (Samanez-Larkin et al. 2008), and lastly damage to the insula has been reported to impair performance on a gambling task (Clark et al. 2008).  The conflicting results as to whether hypo- or hyper- activity of the insula confers vulnerability to GD may arise, in part, due to the relative ubiquity of insula activation.  The insula monitors multiple aspects of behavior, such that changes in activity may reflect epiphenomenon as a result of alterations in autonomic arousal, rather than a definitive role for information-processing within the insula while on task (see Sterzer and Kleinschmidt 2010 for review).  Animal studies may therefore be useful in delineating the role of the insular cortex in behavioural addictions, and potentially inform the development of novel therapeutics. 101  The insular cortex in the rat can be divided into the agranular insula (AI) and granular insula (GI); areas that can be considered broadly homologous to human anterior and posterior insula in regards to connectivity and cytoarchitectonics (Allen et al. 1991; Singer et al. 2009).  The AI is the predominant output domain of the insula, sending projections to amygdalo-frontostriatal circuitry purportedly crucial to addictive behaviors (Allen et al. 1991; Reynolds and Zahm 2005). In contrast, the GI predominantly receives thalamic inputs, and projects mainly to the AI (Shi and Cassell 1998). The relative connectivity of the two regions would suggest that the AI is more likely involved in behavioral addictions (Cocker and Winstanley 2015). Indeed, recent evidence suggests that manipulations of the AI, but not the GI, alter performance on rodent ‘gambling’ tasks; although the directionality of these effects is unclear as AI lesions and pharmacological inactivations have been reported to increase choice of smaller rewards delivered with greater certainty (Ishii et al. 2012; Pushparaj et al. 2015b), whereas chemogentic silencing increases risky choice (Mizoguchi et al. 2015).  However, D4 receptor expression is considerably higher in the GI, in comparison to the AI (Rivera et al. 2008). Given the importance we, and others, have attributed to D4 receptors in mediating attributions of salience to environmental stimuli, the GI may be integral for mediating aspects of gambling-related decision making.  Moreover, both regions have been implicated in controlling drug seeking, as inactivations of either the AI or GI disrupt drug self-administration (Cosme et al. 2015; Di Pietro et al. 2008; Forget et al. 2010; Kutlu et al. 2013; Pushparaj et al. 2015a).  Nevertheless, the relative roles of either area in relation to gambling warrant further investigation.  The goals of this study were therefore to determine whether selective inactivation of the AI or GI, and targeted administration of a D4 agonist would affect performance on the rSMT.  102  6.2 Additional methods  Surgery Surgical and microinfusion procedures were identical to those published previously (Hosking et al. 2014a).  In sum, animals were deeply anaesthetized using 2% isoflurane in O2 and implanted with bilateral 22-gauge stainless steel guide cannula (Plastics One; Roanoke, VA) aimed at either the GI (anteroposterior (AP) + 0.4mm, (ML) mediolateral ± 4.8mm, dorsoventral (DV) - 5.0mm from dura) or AI (AP + 2.8mm, ML ± 4.5, DV -5.0) using standard stereotaxic techniques (incisor bar set to -3.3mm (flat skull cannula at 10o laterally)).  Cannula were fixed to the skull with four stainless steel screws and dental acrylic.  29-gauge obturators (flush with guides) were inserted and covered with a dust cap to protect the infusion site.  Analgesic agents were given as standard via subcutaneous injection (5 mg/kg anafen; 2.5mg/kg bupivacaine). Animals were allowed to recover for at least one week following surgery before testing resumed.  Microinfusions The GI and AI was inactivated using a cocktail of the GABAA agonist muscimol and the GABAB agonist baclofen (0.125µg of each compound in 0.5µl) administered bilaterally at a rate of 0.25µl/minute. On the first infusion day half the animals received baclofen-muscimol and the other half saline according to a balanced design. These designations were subsequently reversed on the second infusion day allowing for a within-subjects design. Following at least a week and reestablishment of statistically stable baseline performance both groups of animals were infused with the D4 agonist PD168077 (0, 0.5, 1.0 & 5.0µg/side) administered according to a digram-balanced Latin square design.  Pharmacological challenges 103  All drugs were prepared fresh daily.  Doses of PD168077 (Tocris, Ellisville, MO) were calculated as the salt and dissolved in 0.9% sterile saline.  Baclofen and muscimol (Sigma Aldrich, Oakville, Canada) were prepared separately at 0.5µg/µl in saline and mixed together in equal volumes to create a 0.25µg/µl solution.   6.3 Results Five animals were excluded from the baseline analysis due to failure to meet established performance criteria, defined as at least 50 trials completed with >50% accuracy on clear loss trials ([0,0,0]), thus the total number of animals included in the baseline analysis was 39. Five animals experienced post-surgical complications and were excluded, and of the remaining animals, a further six were excluded for inaccurate cannula placement (4 GI, 2 AI) yielding a total of 28 animals for analysis of insula inactivations (13 GI, 15 AI). The location of the injector tips for the animals included in the study is depicted in figure 6.1 (panel a: AI; panel b: GI). During re-establishment of baseline following inactivations, a further 8 animals developed health issues and were excluded, therefore 20 animals received PD168077 (10 GI and 10 AI).  Baseline behaviour.  Prior to surgery animals were matched for baseline performance and separated into two groups. There were no statistical differences between the AI and GI group on any behavioural variable and consequently the data for both groups was pooled (surgery group: all F’s <1.2, NS). Consistent with previous reports, choice of the collect lever varied significantly across different trial types (figure 6.2a: trial type: F7,259 = 241.76, p<0.0001). The likelihood of erroneous attempts to collect reward increased with the number of illuminated apertures within the array (figure 6.2b: lights illuminated: F3,111 = 761.47, p<0.0001; 3- vs 2-lights: F1,37 = 117.63, 104  p<0.0001; 2- vs 1-light: F1,37 = 644.71, p<0.0001; 1- vs 0-lights: F1,37 = 249.23, p<0.0001). There was no difference in responding between the different types of 2- or 1-light trials, suggesting the total number of illuminated apertures facilitated erroneous responding rather than the spatial position of the lights (2-light trials: F2,74 = 0.25, NS; 1-light trials: F2,74 = 0.63, NS).  The latency to respond on the collect lever also varied significantly dependent on the trial type (table 6.1a, b: trial type: F7,259 = 27.70, p<0.0001). Erroneous responses were generally slower following near-miss trials in comparison to all other trial types, including wins, indicating these sorts of trials may have fostered more cognitive deliberation within the animals.  Conversely, erroneous responding on 1- and 0-light trials was significantly quicker, indicating that errors appear to arise result of disinhibited or impulsive responding (lights illuminated: F3,111 = 44.40, p<0.0001; 3- vs 2-lights: F1,37 = 3.26, p=0.08; 2- vs 1-light: F1,37 = 22.96, p<0.0001; 1- vs 0-lights: F1,37 = 79.94, p<0.0001). Animals were quicker to respond at the subsequent aperture if the preceding hole had set to on, indicating win-related stimuli facilitated engagement with the task and hastened responding (table 6.2a, b: light status: F1,37 = 8.68, p=0.006). The average number of trials competed per session was 93.22 ± 2.4 (table 6.3a: AI 94.32 ± 3.33, table 6.3b: GI 91.95 ± 0.43).  Agranular insula inactivation Inactivation of the AI via administration of baclofen-muscimol produced robust impairments in animals’ ability to refrain from responding on the collect lever (figure 6.3a, b; dose: F1,14 = 9.36, p=0.008). Interestingly, this effect was not directly related to either the particular trial type or number of lights illuminated (dose X trial type: F7,98 = 1.08, NS). Further analysis revealed that animals showed increased erroneous responding on both 2- and 1-light trials, but not wins or clear losses (dose: 3-lights: F1,14 = 1.17, NS; 2-lights: F1,14 = 7.72, p=0.02; 1-light: F1,14 = 9.01, p=0.01; 0-lights: F1,14 = 0.01, NS). Agranular inactivation also tended to alter the latency for 105  animals to respond on the collect lever (table 6.1a: dose: F1,14 = 4.13, p =0.06). Investigating this further reveals that, following saline administration, animals exhibited a similar pattern to that seen at baseline, in that the latency to press the collect lever on near-miss trials was significantly longer then on other kinds of loss trials (saline, lights illuminated: F3,42 = 23.99, p<0.0001).  However, this pattern was weakened to trend-level significance after inactivation of the AI (BacMus, lights illuminated F3,42 = 2.41, p = 0.08).  Furthermore, animals were no longer quicker to respond at the subsequent hole if the preceding one had been illuminated during AI inactivation (table 6.2a: dose X light status: F1,14 = 11.00 p=0.005; saline light status: F1,14 = 11.76, p=0.004; BacMus light status: F1,14 = 1.70, NS).  AI inactivation may have therefore made animals less sensitive to reward-related stimuli.  AI inactivations also resulted in a significant reduction in the number of trials completed (table 6.3a: dose: F1,14 = 11.76, p=0.005).  This effect may result, at least in part, from the increased time spent in penalty time-outs following elevated erroneous collect responses.  Granular insula inactivation In contrast to AI inactivations, inactivation of the GI produced only minimal effects on task performance. Inactivation did not alter the animals’ propensity to select the collect lever on different trial types (figure 6.3c,d: dose: F1,12 = 0.3, NS, dose X trial type: F7,84 = 0.94, NS), the latency to respond on the collect lever in comparison to saline administration (table 6.1b: dose: F1,12 = 0.43, NS; dose X lights illuminated: F3,36 = 0.59, NS) or the latency to respond at the next aperture based on the status of the preceding one (table 6.2b: dose F1,12 = 1.15, NS; dose X light status: F1,12 = 1.17, NS). GI inactivation did, however, result in a reduction in the number of trials completed, although only at a trend level (table 6.3b: F1,14 = 4.11, p=0.06).  Intra-cortical PD168077 administration 106  Administration of the D4 agonist directly into the AI produced no significant changes in choice of the collect lever (figure 6.4a, b; dose: F3,27 = 1.75, NS; dose X trial type: F21,189 = 0., NS).  Similarly, the drug did not alter the latency to respond on the collect lever (table 6.1a: dose: F3,27 = 1.45, NS dose X trial type: F21,189 F = 0.85), the latency to respond at the subsequent hole based on the status of the preceding hole (table 6.2a: dose: F3,27 = 0.78, NS; dose X light status: F3,27 = 0.91, NS) or the number of trials completed (table 6.3a: dose F3,22 = 1.16, NS).  In contrast to the null effects seen following inactivations, intra-GI administration of PD1686077 produced significant alterations in animals’ behaviour, leading to a decrease in erroneous choice of the collect lever on non-winning trials (figure 6.4c, d dose: F3,27 = 2.91, p=0.05; dose X trial type: F21,189 = 1.91, p=0.01). This main effect was predominantly driven by a strong trend at the highest dose (dose- saline vs 0.5µg: F1,9 = 1.25, NS; saline vs 1.0µg: F1,9 = 0.03, NS; saline vs 5.0 µg: F1,9 = 4.56, p = 0.06). Interestingly, these improvements in performance were not contingent on the number of lights illuminated, but rather the illumination status of the last light in the array, such that intra-GI D4 agonism improved performance when the last light in the array was set to off (saline vs. 5.0 µg: dose X lights illuminated: F3,27 = 0.92, NS; last light status: F1,9 = 7.23, p = 0.03; last light on: F1,9 = 1.21, NS; -last light off: F1,9 = 5.11, p = 0.05). Intra-GI administration of PD168077 failed to alter animals’ latency to respond either on the collect lever (table 6.1b: dose: F3,32 = 0.35, NS; dose X trial type: F21,189 = 0.85, NS) or at the subsequent hole based on the status of the preceding aperture (table 6.2b: dose: F3,27 = 0.34, NS; dose X lights status: F3,27 = 0.84, NS). However, animals completed significantly more trials following intra-GI PD 168077, yet this increase became less pronounced as the dosage increased (table 6.3b: dose F3,22 =3.89, p = 0.02; saline vs 0.5µg: F1,9 = 4.63, p = 0.06; saline vs 1.0 µg: F1,9 = 2.56, NS; saline vs 5.0µg: F1,9 = 0.16, NS).  107   6.4 Discussion Here, we demonstrate differential roles for the granular and agranular insula in controlling reward expectancy in a model of gambling-like behavior.  Temporary inactivation of the AI, but not the GI, impaired the integration of a sequence of reward-related stimuli into an accurate prediction of whether reward was available, resulting in erroneous reward-seeking behavior.  These errors were only evident on trials in which both positive and negative stimuli were present in the sequence, such as on 2- and 1-light trials, but not on clear loss trials in which all lights set to off. In contrast, local infusion of the D4 receptor agonist PD168077 into the GI, but not AI, improved animals’ ability to differentiate winning form non-winning outcomes, but only when the last light in the array remained off.  Collectively, these findings provide novel insight into differing roles for the insula sub-regions; the AI region appears to be engaged in integrating reward related stimuli in order to guide decision making, whereas D4 receptors within the GI appear to play a specific role in boosting the salience of negative reward predictors closest to a decision point.  Intra-GI administration of the D4 agonist also increased trials completed, potentially suggestive of an invigoration of task engagement. The insula has been implicated in a broad range of functions, and as such there may be alternative explanations for the observed results, beyond the integration of reward-related information.  Firstly, the AI contains the primary gustatory cortex (Chen et al. 2011).  As such, impairments in animals’ performance may have arisen as a result of a decrease in the appetitive taste of the food reward.  The drop in the number of trials completed, in addition to the blunted behavioral responses to reward-associated stimuli (insofar as animals were no longer quicker to respond at the subsequent aperture if the preceding light had set to on following AI inactivation) 108  may be consistent with this interpretation.  However, incentive motivation (willingness to work for reward) can be dissociated from hedonic valuation (how much the animal likes the reward), in that a change in the latter does not necessarily impact the former (Balleine and Dickinson 1998; Berridge and Robinson 1998). Damage to the GI does not affect motivation to obtain food under a fixed ratio schedule (Forget et al. 2010) and AI lesions promote choice of the option associated with the most frequent delivery of reward in rodent ‘gambling’ tasks (Ishii et al. 2012; Pushparaj et al. 2015b).  Neurons within the gustatory cortex also continue to fire in response to food in satiated animals (Yaxley et al. 1988).  Even with a quiescent AI, animals should arguably still find the taste of the food appetitive.  Furthermore, if rats found the sugar pellets less appealing after AI inactivation, we should see a reduction in attempts to collect reward, rather than an increase.  Although we cannot rule out the fact that AI inactivations may have altered the hedonic appraisal of the sugar pellets’ taste, this explanation cannot easily explain all the behavioral changes we observe on the rSMT. Secondly, humans who suffer damage to their insular cortex display decision-making impairments that could be explained by a failure to update a strategy in response to changing contingencies (Clark et al. 2008; Weller et al. 2009). Therefore, the effects of AI inactivation on choice on the rSMT could reflect increased perseverative responding, in that animals are not attending to the status of the apertures, but rather perseverating on the collect lever due to its associations with reward delivery.  However, animals still demonstrate differential responding on the collect lever on different trial types i.e. fewer erroneous responses on 0-, vs 1-, vs 2-light trials. Likewise, insular cortex inactivations do not affect animals’ choice behavior on a risk discounting task, a task that requires animals to update strategies as contingencies change (St 109  Onge and Floresco 2010).  As such, increased perseveration is unlikely to underlie the effects of AI inactivations observed here. Contemporary theories of insula function specify a posterior to anterior organization that allows repeated integration of multiple stimulus inputs in order to generate a holistic impression of the body and its surroundings on a moment-to-moment basis (Craig 2002; 2009). The posterior GI is responsible for interoceptive awareness, whereas the more anterior AI integrates information from multiple regions to form a cohesive emotional percept (Craig 2009). Thus the most parsimonious explanation for the deficits seen following AI inactivation may be that the AI is integrating both externally-triggered (environmental) and internally-generated (from the limbic system and other brain regions) affective and interoceptive information in order to influence ongoing behavior.  Broadly, our results here are compatible with a role for the AI in integrating information from multiple regions, insofar as AI inactivations produced deficits on 2- and 1- light trials, i.e. when both winning and losing signals were presented within the array. In contrast, on trials where there is less ambiguity, in that either all or none of the apertures are illuminated, there are no deficits following AI inactivation. Thus, the AI appears to integrate reward related information in order to generate reward expectancy and guide animals’ behavior when environmental stimuli is ambiguous.  Such a supposition may also aid in elucidating the apparently contradictory findings in regards to whether hypo- or hyper-activity of the insula confers vulnerability toward behavioral perturbations. Insula activity appears to be critical for responding to cues predictive of reward (Kusumoto-Yoshida et al. 2015) and updating the incentive value of rewards in response to environmental changes (Parkes and Balleine 2013; Parkes et al. 2015). Theoretically, therefore, insula dysfunction may render animals less able to 110  parse subjectively favorable options. For instance, increases in choice of the risky option have been observed following insula activation on a task in which rats choose such options infrequently at baseline (Mizoguchi et al. 2015).  Conversely, during behavioural tasks that require animals to choose between multiple options, or respond to shifting contingencies, disruption of the insula increases the choice of smaller, safer options (Forget et al. 2010; Ishii et al. 2012). Superficially, these contrasting results appear as though insula inactivation is mediating both increases and decreases in risky choice. However, another explanation may be that these results arise due to an inability to integrate and utilize task-related stimuli. The role of the AI in controlling risk-reward decision-making is likely nuanced such that the AI may not be encoding risk per se, but rather allowing the animal to integrate multiple diffuse stimuli in order to generate the most subjectively appropriate response.  Perturbations to the insula leading to alterations in intrinsic activity likely impair animals’ ability to integrate this information optimally.  The AI is the predominant output domain of the insula, sending projections to key components of the amygdalo-frontostriatal circuitry purportedly crucial to addictive behaviors (Reynolds and Zahm 2005). Indeed, connectivity between the insula and the basolateral amygdala is essential in encoding and updating incentive value (Parkes and Balleine 2013). Additionally, glutamatergic projections from the AI to the nucleus accumbens appear to facilitate the development of aversion resistant alcohol consumption (Seif et al. 2013).  Beyond these direct connections, a modulatory role for the insula has also been proposed.  The prefrontal cortex (PFC) and amygdala both project to the striatum (Sesack and Grace 2010), an area essential for the expression of motivationally driven behaviors (Cador et al. 1989; Robbins et al. 1989). Putatively, the insula modulates the activity of both the PFC and amygdala in response to 111  homeostatic perturbations  (Naqvi and Bechara 2009; Noel et al. 2013; Reynolds and Zahm 2005).  Interestingly, exposure to addictive substances such as alcohol or cocaine appears to alter AI – accumbens connectivity in both animals and humans (McHugh et al. 2013; Mizoguchi et al. 2015; Seif et al. 2013).  These alterations may lead to changes in subjective incentive values, and consequently alter decision-making strategies following drug exposure.  Ultimately, these data suggest that not only are fronto-striatal circuits involving the AI integral for cost-benefit decision making, but that these circuits are readily altered by drug exposure, which may explain some of the disparate findings in regards to hypo- and hyper-activation of the insula in addictive disorders.  The results from the AI here are relatively similar to those we have previously obtained when investigating the role of the anterior cingulate cortex (ACC).  Activation of the insula is typically accompanied by concomitant activation of the ACC, strongly suggesting complimentary or overlapping roles in multiple task domains (Craig 2009). Our data provide further evidence that these two regions may play similar roles in behavioral addictions, but there are key differences in the observed deficits.  We previously reported that the ACC acts to constrain maladaptive reward-seeking responses by integrating information signaled by multiple sources into a prediction regarding reward availability on each trial (Cocker et al. 2016).  After ACC inactivation, animals exhibited an increase in erroneous collection responses on 1- and 0-light trial types, leading us to conclude that the ACC acts to constrain the Pavlovian approach behavior that is facilitated by the strong appetitive valence associated with the collect lever (Cocker et al. 2016).  The deficits observed following AI inactivation were on 2- and 1- light trials, trials in which winning and losing signals are presented within the array such that the animal must integrate conflicting information when deciding which lever to press.  Both the AI 112  and ACC appear to contribute to task performance in a qualitatively similar way, in that both are important when conflicting information is presented within a trial.  The difference however, is that the ACC appears to constrain motoric responses to stimuli with high appetitive value, whereas the AI appears to be more broadly engaged in guiding responses in the presence of conflicting environmental stimuli.  These differences may arise as a result of the two regions’ distinct integrative functions.  Although both integrate information from multiple areas in order to guide responses, the ACC has been proposed as a motor interface, consistent with our observation that animals fail to inhibit maladaptive approach behavior.  In contrast, the AI generates an integrated “feeling state” in response to sensory information, which could explain why deficits are present on trials when both winning and losing stimuli are presented in parallel (Craig 2009; Sterzer and Kleinschmidt 2010). In contrast to the pronounced effects seen following AI inactivation, infusion of baclofen and muscimol into the GI produced no statistically significant effects on task performance.  The lack of effects observed following GI inactivation, may result from the relative connectivity of the two areas.  Similar to the AI, the GI also appears to function primarily as an integrative region, receiving extensive thalamic connections but projecting mainly to the AI (Shi and Cassell 1998; Singer et al. 2009).  Given that AI inactivation results in choice deficits, whereas GI inactivation does not, it appears that any projections from the GI to the AI are contributing only minimally to this kind of reward-related behavior.  The GI may be critical in responding to deleterious perturbations from homeostatic norms (Wager et al. 2003).  The role of the GI in addictive disorders has generally been ascribed to signaling, and consequently prompting avoidance of, aversive states such as craving that precipitate behaviors such as drug seeking (Naqvi and Bechara 2009; Naqvi et al. 2007; Noel et al. 2013). For instance, patients who 113  smoked prior to suffering damage to their insula report immediate cessation of cigarette smoking, suggesting the craving involved in maintaining addiction was alleviated (Abdolahi et al. 2015; Naqvi et al. 2007; Noel et al. 2013).  Similarly, animal studies have demonstrated that damage to the GI leads to a cessation of nicotine self-administration (Forget et al. 2010).  In contrast to drug self-administration paradigms, the rSMT is unlikely to generate strongly aversive interoceptive states.  Whilst the failure to receive an expected reward may be frustrating, it is unlikely to be as aversive as drug withdrawal following prolonged access.  However, gamblers have been demonstrated to display craving (Potenza 2008), which may indicate that the GI is involved in the maintenance of problem gambling, but perhaps not its development nor in recreational gambling.  After all, while these rats certainly engage in regular gambling-like behaviour, it is far from obvious that these rats are “addicted” to playing the rSMT.  A model wherein animals display an addition-like phenotype would be useful in delineating any differential role for both the GI and AI in the formation and maintenance of problem gambling, and current studies are aiming to address such questions.  Lastly, the results of this investigation continue to indicate that D4 receptors may be important in signaling the salience of environmental cues. The fact that the observed effects were restricted to the GI is not unexpected; D4 receptor protein is relatively abundant within the GI, but sparse in the AI region (Rivera et al. 2008).  The insula is heavily innervated by dopaminergic projections arising from the ventral tegmental area (Gaspar et al. 1989).  These neurons have been canonically linked to the generation of prediction errors, insofar as these neurons fire in response to primary rewards, but if these rewards are reliably predicted by a stimulus, then the firing of these neurons shifts to the onset of such a conditioned stimulus (Schultz and Romo 1990).  We have argued that the potency of slot machines may be 114  underpinned by aspects of this process, in that the successive presentation of reward related stimuli generates an expectation of reward (Cocker and Winstanley 2015; Winstanley et al. 2011).  Indeed, our previous investigations using this task indicate that illuminated apertures function as incentive stimuli, and that responses to those stimuli can be modulated via systemic administration of D2-like ligands, particularly compounds selective for the D4 receptor (Cocker et al. 2014; Winstanley et al. 2011).  Such data are consistent with a growing body of evidence suggesting D4 receptors modulate salience attributions to environmental stimuli (Cocker and Winstanley 2015; Lauzon and Laviolette 2010; Yan et al. 2012).  However, in contrast to our previous findings, administration of a D4 agonist here improved, rather than impaired, animals’ performance.   In seeking to explain this observation, it is important to note that a subset of dopaminergic neurons fire in response to aversive outcomes, or stimuli that predict an aversive outcome (Schultz et al. 1997; Zweifel et al. 2011). Therefore, the differences observed here may be due to the proposed role of the GI in mediating unpleasant or aversive somatic states (Paulus and Stein 2006).  In other words, whilst systemic administration of a D4 agonist augments the salience of reward, the same compound administered into the GI appears to selectively increase the salience of a potential punishment.  The improvements seen following PD168077 administration are on trials wherein the last light remains off, and are therefore on trials in which a negative stimulus is presented most proximally in time to the decision point.  An increase in salience of a negative stimulus would have the greatest impact on such trials, as there would be little time for such a signal to decay or be overshadowed by subsequent presentation of a positive stimulus.  Certainly D4 receptors within prefrontal regions have been shown to selectively augment the salience of stimuli predicting a foot shock (Lauzon et al. 2012; Lauzon et al. 2009).  115  In summary, these data support the hypothesis that the insular cortices make important yet dissociable contributions to reward-based decision making, and that these processes can influence gambling-like behavior.  The AI appears to be specifically involved in integrating task related information in order to guide decision making when outcomes are ambiguous, potentially through its role in generating a cohesive emotional response.  In contrast, the GI does not appear to be critically involved in this process.  However, activation of D4 receptors within the GI can bias choice, potentially through selectively augmenting the salience of environmental cues that are predictive of aversive outcomes.  These two areas may be differentially engaged based on the temporal sequelae of addiction, with the AI having a greater influence during the initial acquisition stage, whereas the GI may be more involved in the maintenance of addiction. An alternative interpretation may be that dysfunction within each area may represent a different vulnerability.  Gamblers are a heterogeneous group and there is unlikely to be a single etiology underlying the development of GD.  The current data intimates that gamblers who experience difficulties in reward based decision making, an established endophenotype, may exhibit AI dysfunction.  In contrast, the GI may contribute to problem gambling in individuals who experience losses as particularly aversive, an explanation that may offer insight into the augmented loss-chasing seen in gamblers (Lesieur 1979). Thus, different areas of the insula may contribute to different aspects of GD.  A greater appreciation of the neuronal circuitry underlying different facets of GD will hopefully lead to better understanding of the nature of this addiction disorder, and how to best treat different patient subgroups.  Figure 6.1a  116   Figure 6.1b  117  Figure 6.1 Histological analysis of cannula. Location of all acceptable agranular insula (a) and granular insula (b) infusion sites. Coordinates are relative to bregma, plates modified from Paxinos and Wilson (1998)       Figure 6.2 Baseline rSMT performance (a,b) Animals showed optimal responding on win trials (1,1,1), choosing to collect the available reward nearly 100% of the time.  Similarly, animals showed a marked preference for the optimal response, now the roll lever, when no lights were illuminated (0,0,0), only responding on the collect lever approximately 10% of the time.  Erroneous collect responses increased to 42.45% when 1 light was illuminated.  However, when two-lights were illuminated in the array, animals responded erroneously on the collect lever at a far greater then chance level (80.87% ± 3.29 (SEM)), indicating that rats, like humans, treat such stimuli as more indicative of a win than a loss, and are hence susceptible to the near-miss effect.  All data shown are the mean across five sessions ± SEM.  118     Figure 6.3 Effects of insula inactivation on rSMT performance. (a, b) Infusion of baclofen-muscimol into the agranular cortex resulted in impairment in animals’ performance to differentiate winning from non-winning outcomes on both 2- and 1-lights were illuminated i.e. on trial types when both winning and losing stimuli were present within the array. (c, d) In contrast temporary inactivation of the granular cortex had no effect on animals’ task performance. All data are shown as mean ±SEM  119     Figure 6.4 Effect of activation of local D4 receptors within the insular cortex via infusion of the D4 agonist PD168077 on rSMT performance.  (a, b) Infusion of the D4 agonist into the agranular cortex had no effect on task performance. (c, d) However, activation of D4 receptors within the granular insula lead to improvements in animals’ performance, particularly on trials wherein the light in the last aperture remained off i.e. the aperture that was temporally and spatially proximal 120  to the decision point suggesting D4 receptor activation may be augmenting the salience of a potential punishment. All doses are given in µg/µl and data are shown as mean ±SEM  Table 6.1A Agranular insula; Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Baseline  0.55 ± 0.06 0.78 ± 0.10 0.55 ± 0.06 0.82 ± 0.38 0.52 ± 0.06 0.46 ± 0.09 0.34 ± 0.06 0.22 ± 0.06 BacMus 0 0.71 ± 0.16 0.94 ± 0.15 0.63 ± 0.10 0.60 ± 0.12 0.87 ± 0.24 0.65 ± 0.19 0.40 ± 0.05 0.26 ± 0.07  0.125 µg 0.71 ± 0.18 0.72 ± 0.13 1.05 ± 0.33 0.38 ± 0.08 0.60 ± 0.09 1.54 ± 0.96 0.83 ± 0.26 2.66 ± 2.49 PD168077 0 0.71 ± 0.08 0.71 ± 0.07 0.57 ± 0.09 0.43 ± 0.18 0.58 ± 0.09 0.53 ± 0.13 0.36 ± 0.09 0.10 ± 0.06  0.5 µg 0.56 ± 0.07 0.67 ± 0.1 0.60 ± 0.10 0.46 ± 0.10 0.63 ± 0.10 0.57 ± 0.15 0.36 ± 0.11 0.24 ± 0.09  1 µg 0.72 ± 0.12 0.73 ± 0.09 0.73 ± 0.15 0.59 ± 0.13 0.60 ± 0.06 1.58 ± 1.10 0.50 ± 0.16 0.32 ± 0.11  5 µg 0.58 ± 0.07 0.60 ± 0.09 0.55 ± 0.10 0.52 ± 0.24 0.54 ± 0.11 0.31 ± 0.07 0.33 ± 0.11 0.12 ± 0.05  Table 6.1B Granular insula; Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Baseline  0.37 ± 0.03 0.66 ± 0.08 0.54 ± 0.08 0.33 ± 0.05 0.46 ± 0.06 0.52 ± 0.12 0.36 ± 0.07 0.12 ± 0.03 BacMus 0 0.58 ± 0.07 0.83 ± 0.14 0.58 ± 0.07 0.50 ± 0.12 0.58 ± 0.08 0.40 ± 0.06 0.25 ± 0.05 0.14 ± 0.04  0.125 µg 0.60 ± 0.06 0.90 ± 0.19 0.73 ± 0.11 0.76 ± 0.19 0.69 ± 0.18 0.82 ± 0.23 0.38 ± 0.08 0.45 ± 0.16 PD168077 0 0.59 ± 0.16 0.77 ± 0.17 0.56 ± 0.21 0.44 ± 0.13 0.49 ± 0.12 0.43 ± 0.11 0.26 ± 0.07 0.07 ± 0.02  0.5 µg 0.65 ± 0.28 0.57 ± 0.09 0.52 ± 0.15 0.36 ± 0.10 0.58 ± 0.14 0.33 ± 0.05 0.22 ± 0.07 0.13 ± 0.04  5 µg 0.58 ± 0.10 1.90 ± 1.11 0.44 ± 0.07 0.41 ± 0.15 0.55 ± 0.10 0.24 ± 0.07 0.31 ± 0.10 0.05 ± 0.02  121   Table 6.2A Agranular insula; Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Baseline  0.93 ± 0.09 1.26 ± 0.17 1.11 ± 0.12 1.32 ± 0.16 BacMus 0 1.13 ± 0.18 1.44 ± 0.26 1.07 ± 0.12 1.63 ± 0.26  0.125 µg 1.27 ± 0.26 3.04 ± 0.89 3.71 ± 2.28 4.70 ± 2.51 PD168077 0 1.07 ± 0.20 1.54 ± 0.47 0.94 ± 0.20 4.06 ± 3.07  0.5 µg 0.83 ± 0.14 2.24 ± 0.39 0.96 ± 0.08 1.09 ± 0.21  1 µg 0.96 ± 0.21 1.65 ± 0.27 2.42 ± 1.30 1.36 ± 0.26  5 µg 0.71 ± 0.12 1.27 ± 0.28 0.86 ± 0.07 1.26 ± 0.22  Table 6.2B Granular insula; Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges. Data presented in time (s) shown as mean ± SEM Drug Dose Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Baseline  1.07 ± 0.13 1.60 ± 0.29 1.23 ± 0.12 1.67 ± 0.16 BacMus 0 0.84 ± 0.12 2.06 ± 0.35 1.25 ± 0.13 2.03 ± 0.33  0.125 µg 1.02 ± 0.08 1.61 ± 0.34 1.15 ± 0.08 1.29 ± 0.14 PD168077 0 1.11 ± 0.21 1.40 ± 0.51 0.99 ± 0.14 1.15 ± 0.16  0.5 µg 0.97 ± 0.19 1.12 ± 0.19 0.94 ± 0.21 1.23 ± 0.26  1 µg 0.82 ± 0.12 1.51 ± 0.22 0.89 ± 0.16 1.05 ± 0.17  5 µg 0.84 ± 0.17 1.45 ± 0.29 1.05 ± 0.21 1.39 ± 0.29  122  Table 6.3A Agranular insula; Trials completed at baseline and following pharmacological challenges. Data shown are mean ± SEM Drug Dose  Trials completed Baseline  94.32 ± 3.33 BacMus 0 85.19 ± 5.12  0.125 µg 59.88 ± 7.74 PD168077 0 85.7 ± 10.94  0.5 µg 78.0 ± 8.36   1 µg 81.7 ± 8.85  5 µg 91.3 ± 8.92  Table 6.3B Granular insula; Trials completed at baseline and following pharmacological challenges. Data shown are mean ± SEM Drug Dose  Trials completed  Baseline  91.95 ± 0.43  BacMus 0 83.14 ± 5.12   0.125 µg 73.0 ± 7.02  PD168077 0 87.9 ± 5.31   0.5 µg 103.3 ± 5.69   1 µg 77.9 ± 10.68   5 µg 85.6 ± 7.51     123  Chapter 7: Experiment 5: The dopamine D2/3 agonist ropinirole invigorates performance and induces compulsive-like gambling behaviour on a rodent slot machine task.    7.1 Introduction  Gambling disorder is a growing public health concern. Access to gambling opportunities are growing steadily, yet the etiology of GD remains unknown and treatment options lack efficacy (Grant et al. 2012; Limbrick-Oldfield et al. 2013). GD was recently reclassified as an addiction disorder in DSM5, in part due to the similarity in phenomenology between GD and drug addiction (Potenza 2006). In contrast to chemical dependency, a prototypical neurobiological phenotype, such as low D2/3 receptor expression, has not been identified for GD (Limbrick-Oldfield et al. 2013; Volkow et al. 2004; Volkow et al. 2007). Given the centrality of dopamine to substance use disorder, and its undisputed role in motivational processes, this neurotransmitter system has understandably been the focus of much research into the neurobiological basis of GD. Nevertheless, a definitive understanding of the role played by dopaminergic dysfunction in the formation or maintenance of GD has remained elusive (Potenza 2013), potentially due to the highly heterogeneous nature of both GD and the structural characteristics of different gambling scenarios (see Cocker and Winstanley, 2015 for discussion). However, one form of GD that can be causally linked to aberrant dopaminergic function is the iatrogenic gambling seen following dopamine replacement therapy. This particularly devastating form of pathological gambling has been predominantly described in Parkinson’s 124  disease (PD), but has also been observed in patients with restless leg syndrome (RLS), fibromyalgia and prolactinoma (Clark and Dagher 2014; Voon et al. 2007a; Weintraub and Potenza 2006).   Although there are instances of GD and impulse control disorders (ICDs) emerging following L-DOPA treatment, the majority of such cases develop in response to treatment with agonists that predominantly activate the D2/3 receptor sub-types, such as ropinirole and pramipexole (Dodd et al. 2005). Animal models have previously shown that both acute and chronic treatment with dopamine agonists with high affinity for the D2-like family promote a compulsive-like behavioural phenotype (Eagle et al. 2014; Szechtman et al. 1998), raising the possibility that dysfunction within the dopamine system could precipitate a compulsive form of gambling.  In support of this hypothesis, recent animal work has confirmed that repeated injections of pramipexole increase choice of uncertain outcomes in a probability-discounting task (Rokosik and Napier 2012). However, patients who develop GD as a result of dopamine agonist therapies appear to prefer simple, repetitive games, such as slot machines, in which there is little (if any) explicit computation of utility (Rossi et al. 2010).  Within the panoply of gambling games, slots and EGMs have been suggested as a particularly virulent form of gambling, precisely because these machines are engineered to promote compulsive, or unthinking, play (see Murch and Clark 2015 for discussion). Thus, given the qualitatively similar style of gambling engagement, there may be particularly relevant parallels between the development of iatrogenic GD following dopamine agonist therapy and compulsive engagement with EGMs. It would therefore be useful to determine the impact of chronic D2/3 agonist treatment in a rodent model of this type of gambling. 125  We have developed and validated a rodent analogue of a simple slot machine, wherein rats share key behavioural features with human gamblers.  Specifically, reward-salient cues are able to evoke the expectation of imminent reward. We have previously argued this phenomenon is translationally analogous to the so-called ‘near-miss effect’. Acute administration of the D2-like agonist quinpirole potentiated erroneous expectations of reward (Cocker et al. 2014; Winstanley et al. 2011). Thus our previous investigations intimate that acutely augmented dopaminergic signaling through D2-like receptors may increase animals’ responsivity to reward-related cues, in a manner potentially comparable with human gamblers (Habib and Dixon 2010).  Here, we intended to determine if chronic administration of the D2-like agonist ropinirole, a drug that is commonly associated with the formation of ICD’s in PD, alters animals’ performance on the rSMT. Moreover, we explored any ropinirole-induced changes in receptor expression and intracellular signalling pathways ex vivo in order to gain insight into the possible mechanism underlying behavioural change, and hence potentially inform our understanding of how dopamine may contribute toward the aberrant decision making observed in problematic gambling.  Such data may also be highly impactful for clinical use of D2/3 agonists in PD treatment; although a previous history of ICDs or earlier age of PD onset have been indicated as risk factors for the development of iatrogenic ICD’s (Voon et al. 2007b), little is known about the aetiology of these psychiatric side-effects or how to prevent them. As such, many neurologists have dropped this class of drugs from their already-limited toolbox of effective treatments for PD, resulting in a heavy reliance on L-DOPA with all the associated motor side-effects (Calabresi et al. 2010).   126  7.2 Additional methods Extinction and reinstatement The purpose of the extinction/reinstatement test was to determine if chronic ropinirole administration would prevent task performance from declining when reward was no longer available, as might be expected if behaviour had become habitual rather than truly goal-directed (Balleine and Dickinson 1998). Animals from both ropinirole and saline treated groups were matched for their task performance and split into groups. Groups of ropinirole and saline treated animals performed the rSMT in extinction during which a collect response following a win trial no longer led to a delivery of sugar pellets. Following 6 extinction sessions the animals were reinstated on the slot machine program for a final 4 sessions.  Ex-Vivo analysis Immediately following the last day of ropinirole administration half of the animals from each group were sacrificed. The remaining animals continued with behavioural testing for a further 4-weeks in order to elucidate any long-term effects of ropinirole administration. All animals were sacrificed by live decapitation immediately following the last behavioural test session. Tissue samples from the medial PFC, OFC, dorsal striatum and NAc were harvested and flash frozen. Tissue from the dorsal striatum and NAc was analysed via Western blotting to determine protein levels for; dopamine D1 and D2 receptors, dopamine and cyclic adenosine (cAMP) regulated phosphoprotein with molecular weight 32 Kda (DARPP), phosphorylated DARPP at serine 34 monophosphate (pDARPP34) & threonine-75 (pDARPP75), cAMP response element binding protein (CREB), pCREB, glycogen synthase kinase-3beta (GSK3β) and β-tubulin.  Quantitative Polymerase Chain Reaction (qPCR) was used to determine RNA levels of D1, D2, and 5-HT2A receptors in the mPFC and OFC.  127  Western blotting Frozen tissue samples were defrosted on ice in 50µL fresh lysis buffer: RIPA buffer (50 mM Tris, 150 mM NaCl, 10% SDS, 1% IGEPAL, 0.5% Sarkosyl, pH 8.0, 4°C) enriched with protease (complete protease inhibitor cocktail, Roche Diagnostics, Laval, QC, Canada) and phosphatase inhibitors (Halt phosphatase inhibitor cocktail, Thermo Scientific, Rockford, IL, USA). Samples were homogenised by sonification and centrifuged for 15 minutes (15,700 g, 4°C). Protein levels in the supernatant were determined with a NanoDrop 2000 spectrophotometer (Thermo Scientific) and 75 μg protein per sample were loaded onto 10% Tris polyacrylamide gel for electrophoresis separation. Samples were subsequently transferred to a polyvinyl difluoride membrane, washed briefly in phosphate-buffered saline (PBS), and blocked for one hour at room temperature (Odyssey blocking buffer, LI-COR Biosciences, Lincoln, NE). Blocked membranes were incubated overnight at 4°C with primary antibodies 1/1000 CREB , 1/500 pDARPP32 (Thr34), 1/500 pDARPP32 (ser75), 1/1000 GSK3β, 1/500 pGSK3β (Cell Signaling Technology, Danvers, MA); 1/1000 pCREB (Ser133), 1/200 D2 receptor, (Millipore, Billerica, MA); 1/500 DARPP32 (BD Biosciences, Mississauga, ON, Canada); 1/200 D1 receptor (Santa Cruz Biotechnology, Dallas, TX,) in blocking buffer containing 0.1 % Tween-20. Membranes were thoroughly washed in PBST (4x each for 10 minutes) and incubated for 2 hours at room temperature with the appropriate IRDye® secondary antibodies (1/10000 goat anti-mouse, 1/10000 goat anti-rabbit, LI-COR Biosciences) in blocking buffer containing 0.1 % Tween-20; goat anti-rabbit incubation solution also comprised 0.005% SDS. Membranes were thoroughly washed with PBST (4X each for 10 minutes) and then with PBS (4X each for 10 minutes). Visualisation and quantification of protein levels was undertaken with the Odyssey 128  Imaging System (LI-COR Biosciences), and protein content was normalised to levels of β-tubulin (0.2 µg/mL; anti-β-tubulin antibody, Millipore). qPCR  Frozen tissue samples were homogenised in TRIzol (Invitrogen, CA). Genomic DNA was removed from the sample using the DNA-free kit (Ambion, TX) and cDNA synthesis was performed using the SuperScript vilo kit (Invitrogen, CA). Transcripts for genes of interest were quantified using real-time qPCR (SYBR GreenER; Invitrogen, CA) on a StepOnePlus 96-well thermocycler (Applied Biosystems, CA). All primers were custom synthesised by the UBC campus NAPS unit (Michael Smith Laboratories, UBC, Canada; for sequences see table 7.1) and validated for linearity and specificity. All PCR data were normalised to levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which did not vary across group, based on the following formula ∆Ct=Ct (gene of interest)-Ct (GAPDH). Adjusted expression levels for ropinirole treated animals were calculated relative to the saline treated animals as follows ∆∆Ct=∆Ct (ropinirole group) -∆Ct (saline group). Consistent with previous reports expression levels relative to controls were then calculated using the following expression 2-∆∆Ct.   Data analyses:  During ropinirole administration data was grouped into session bins. Animals completed 14 sessions on the rSMT initially following mini-pump implantation. These data were parsed into bin 1 (sessions 1-5), bin 2 (sessions 6-10) and bin 3 (sessions 11-14) in order to meaningfully analyse main effects of session over 14 days. Extinction sessions were analysed separately (sessions 15-21) as were reinstatement sessions (sessions 22-24). All analyses remained the same as above with the exception that drug treatment (two levels; saline vs. ropinirole) was used as a between subjects factor. The animals used to assess the effects of a washout period were run for 129  a further 20 sessions, broken down into four 5-sessions bins. Western blot and PCR analysis was conducted with repeated measures ANOVA with drug (2 levels) and time point (2 levels – sacrificed immediately following cessation of ropinirole (early) or following a 4-week washout (late)). The significance level for all effects was p ≤0.05.  Analyses for which p ≤0.1 were described as trends. All data are presented as mean ± standard error of the mean (SEM).    7.3 Results  Twelve of the animals were excluded from the analysis due to failure to meet established performance criteria of at least 50 trials completed per session and at least 50% accuracy on clear loss trials ([0,0,0]) (Cocker et al. 2016; Cocker et al. 2014; Winstanley et al. 2011). The total number of animals’ included in the analysis was therefore twenty-four.  Baseline behaviour Prior to osmotic mini-pump implantation animals were matched for baseline performance into two groups. There were no statistical differences between animals in the ropinirole or saline group on any behavioural variable, and consequently the data for both groups is pooled here (group: all F’s <0.63, NS).  Consistent with all previous reports using this task, responses on the collect lever varied significantly across different trial types (figure 7.1a; trial type: F7,154 = 117.48, p<0.0001), with erroneous responses becoming increasingly likely as the number of illuminated apertures within the array increased (figure 7.1b; lights illuminated: F3,66 = 355.99, p<0.0001; 3 vs 2: F1,22 = 148.93, p<0.0001; 2 vs 1: F1,22 = 153.3, p <0.0001; 1 vs 0: F1,22 = 54.92, p<0.0001). The latency to respond on the collect lever also varied between trial types and was contingent on the number of lights illuminated within the array (table 7.1a, b; trial type: F7,154 = 10.98, p<0.0001; lights illuminated: F3,66 = 12.94, p<0.0001). Interestingly, there was no 130  difference between winning or near-miss (two-light) trials, but similar to previous reports responses became progressively quicker when 1- and 0-lights were illuminated, (3 vs 2: F1,22 = 0.51, NS; 2 vs 1: F1,22 = 22.54, p<0.0001; 1 vs 0: F1,22 = 17.32, p<0.0001), indicating that erroneous responses on these trial types are likely as a result of disinhibited or impulsive responding, similar to a recent report (Cocker et al. 2016). Animals’ latency to nose poke into subsequent apertures was significantly reduced if the light in the preceding hole had set to on, indicating that signals concordant with a winning outcome serve to invigorate future responding, potentially indicative of the attribution of incentive salience to the illuminated lights (table 7.3a, b; light status: F1,22 = 19.13, p<0.0001; see discussion in Cocker et al. 2014, 2016). Animals completed an average of 83.12 ± 3.71 trials per session (saline 84.32 ± 3.36; ropinirole 81.92 ± 4.05).   Ropinirole Chronic administration of the D2-like agonist initially produced robust impairments in animals’ ability to differentiate winning from losing trials in comparison to saline treated controls (figure. 7.3a, b; bin 1; trial type X drug: F7,154 = 3.21, p=0.003; figure. 7.3c, d: bin 2: trial type X drug F7,154 = 3.07, p=0.005). These deficits in performance are comparable with those observed following acute administration of a D2-like agonist (Cocker et al. 2014; Winstanley et al. 2011). However, over time, animals receiving ropinirole actually began to show improvements in performance on the rSMT, such that by the third trial bin they were almost less susceptible to erroneous collection responses on losing trials then control animals, although these results fell just short of statistical significance (figure 7.3e, f; bin 3; trial type X drug: F7,154 = 2.03, p=0.06).  The initial deficits observed may be best attributed to impaired performance on 2- and 1-light trials, although parsing the error rate by lights illuminated only reached trend-level significance 131  (bin 1; lights illuminated X drug: F3,66 = 2.44, p=0.072).  In contrast, the impairments observed in the second bin of sessions were not dependent on the number of lights illuminated (bin 2; lights illuminated X drug: F3,66 = 0.26, NS), but rather on the status of the last light within the array, such that errors in bin 2 were more likely on trials in which the last light set to on (bin 2; last light X drug: F1,22 = 7.58, p=0.01). Thus chronic ropinirole administration appeared to alter sensitivity to reward-related stimuli presented within close temporal proximity to a decision point. Four weeks after drug delivery had ceased, there was no difference in performance between ropinirole- and saline-treated animals (figure 7.4a; time point X drug: F7,49 = 0.76, NS).  With regards to the speed at which collect responses were made, ropinirole treated animals were slower to respond on the collect lever during the first bin of drug treatment, but this effect abated in the subsequent bins (table 7.2a, b; bin 1: time point X drug: F7,154 = 2.94, p=0.006; bin 2: F7,154 = 0.37, NS; bin 3: F7,154 = 1.15, NS). The tendency observed at baseline for animals to respond quicker in the subsequent hole if the light in the preceding hole had set to on was still present in both ropinirole and saline treated animals in the first two bins (table 7.3a, b; bin 1; light status: F1,22 = 7.77, p = 0.011; light status X drug: F1,21 = 0.09, NS; bin 2: light status: F1,22 = 7.87, p = 0.01; light status X drug: F1,22 = 2.66, NS). However, in the last bin, the characteristic “slowing” of the next response when a light in the stimulus array set to off was less apparent, though still present, in ropinirole-treated rats (table 7.3a, b; bin 3; light status: F1,22 = 14.57, p=0.001; light status X drug: F1,22 = 5.04, p=0.035; ropinirole treated; light status: F1,11 = 10.30, p=0.008; saline treated; light status: F1,11 = 9.77, p=0.01). After 4 weeks of washout, any differences in the latency to respond at an aperture based on the preceding one were no longer evident (figure 7.4a: light status X drug: F1,6 = 1.80, NS).  132  In contrast to the bidirectional effects ropinirole produced on animals’ erroneous attempts to collect reward on non-winning trials, chronic ropinirole produced a clear, robust and sustained increase in the number of trials animals completed.  This effect started to manifest after the first 5 sessions, and became more pronounced throughout treatment (figure. 7.3; bin 1: drug: F1,22 = 0.06, NS; bin 2: drug: F1,22 = 6.50, p=0.02; bin 3; drug: F1,22 = 17.03, p<0.0001). During extinction, both groups significantly reduced the number of trials completed, although animals receiving ropinirole did not extinguish responding as rapidly or as much as the saline-treated animals (figure 7.3: extinction; drug: F1,22 = 28.47, p<0.0001; session X drug: F5,110 = 4.37, p=0.001; ropinirole treated; session: F5,55 = 17.74, p=0.0001; saline treated; session: F5,55 = 27.84, p<0.0001).  Furthermore, the increase in the number of trials completed in the ropinirole group was preserved throughout reinstatement (figure 7.3: reinstatement; drug: F1,22 = 17.65, p<0.0001) and remained elevated following the cessation of ropinirole treatment, albeit only at the level of a strong trend (drug: F1,7 = 5.09, p=0.06). This effect began to ameliorate, such that by the last 5 sessions there were no significant differences between the two groups (drug: washout sessions#1-5: F1,7 = 11.25, p=0.01; #6-10: F1,7 = 6.44, p=0.04; #11-15: F1,7 = 6.65, p=0.04; #16-20: F1,7 = 2.21, NS).  Western Blot Given that ropinirole’s primary pharmacological mechanism of action is as an agonist at dopamine D2/3 receptors, one obvious hypothesis would be that chronic ropinirole brings about behavioural change via alterations in dopaminergic signaling. In the dorsal striatum, there was a slight increase in D2 receptor protein following ropinirole administration, but only at the level of a very weak trend (figure. 7.5a; drug: F1,3 = 5.47, p=0.1).  This effect appeared to be driven by an increase in D2 receptor expression in the group sacrificed while ropinirole was still on board, that 133  was not present in ropinirole-treated rats post-washout animals, but all follow up analyses were not significant (all F’s<1.72, NS). In contrast, the effects on D1 receptor protein were more pronounced: chronic ropinirole decreased expression of the D1 receptor protein, an effect that subsided following four weeks’ washout (figure. 7.5b; drug X time point: F1,3 = 12.72, p=0.04; early ropinirole vs. early saline: F1,5 = 16.84, p=0.009; late ropinirole vs late saline: F1,3 = 0.03, NS).  D2-like receptors are metabotropic receptors typically coupled to inhibitory G-proteins (Pierce et al. 2002).  Activation of D2 receptors decreases cAMP production, inhibiting PKA, leading to a decrease in phosphorylation of DARPP at serine 34, and a subsequent decrease in ERK signaling (Beaulieu et al. 2007).  Ropinirole treatment lead to a significant reduction in DARPP (figure 7.5c; drug: F1,3 10.05, p=0.05), although this was only significant at the later time point (early ropinirole vs early saline; F1,3 = 1.72, NS; late ropinirole vs. late saline: F1,3 = 10.50, p=0.05). There was also a trend toward decreased total ERK and DARPP75 in all ropinirole treated animals (figure 7.5d; ERK: F1,3 = 5.82, p=0.095; figure 7.5e; drug- DARPP: F1,3 = 6.2, p=0.09), but no changes in pDARPP34, CREB or pCREB (figure 7.5f-h; drug- pDARPP34: F1,3 = 1.14, NS, -CREB: F1,3 = 1.28, NS -pCREB: F1,3 = 0.37, NS). There is therefore only marginal evidence that ropinirole caused lasting change through this signaling pathway, at least in the dorsal striatum. D2 receptors can also signal through an alternative G-protein/cAMP independent cascade via a β-arrestin mediated – AKT/GSK3β cascade during hyperdopaminergic conditions (Beaulieu et al. 2007; Li et al. 2012). Chronic ropinirole administration did not lead to any alterations in total GSK3β (figure. 7.5i; drug: F1,3 = 2.35, NS). However, levels of pGSK3β were vastly decreased, an effect that was not ameliorated following washout (figure 7.5j; drug: F1,3 = 134  19.55, p=0.02; drug x time point: F1,3 = 1.45, NS; early ropinirole vs. early saline: F1,5 = 33.83, p=0.002; late ropinirole vs. late saline: F1,3 = 19.34, p=0.02). GSK3β is constitutively active, such that phosphorylation renders it inactive.  Hence, it would appear that ropinirole resulted in significantly greater activation of this alternative signaling pathway in the dorsal striatum.  In contrast to the multiple changes in both the canonical and alternative dopaminergic signaling cascades observed in the dorsal striatum, the only significant difference in protein expression that we detected in the NAc was a ropinirole-induced decrease in pCREB in animals who were sacrificed immediately following the cessation of treatment (figure 7.6h; group: F1,3 =10.79, p=0.046; drug x time point: F1,3 = 0.01, NS; early saline vs. early ropinirole: F1,5 = 7.96, p=0.037; late saline vs. late ropinirole: F1,3 = 5.33, NS).    PCR In contrast to the observed changes in dopamine related signaling in the striatum, there were no alterations in receptor mRNA for the D1, D2, or 5-HT2A receptors between saline and ropinirole treated animals at either time point in either the OFC or PFC (table 7.4: all F’s <1.03).   7.4 Discussion Here, we demonstrate that chronic administration of the D2/3 agonist ropinirole produces invigorated engagement on the rSMT, as indicated by a marked increase in the number of trials completed, and a reduction in the degree to which stimulus valence modulated the speed of subsequent responses at the array.  This potentially compulsive style of play was accompanied by a dramatic and long-lasting reduction in the inactive (phosphorylated) form of GSK3β in the dorsal striatum, potentially indicative of increased D2-mediated activation of the β-arrestin-AKT 135  intra-cellular signaling cascade in this area.  Chronic ropinirole treatment was also associated with decreased expression of D1 receptors in the dorsal striatum and reduced activation of CREB within the NAc.  These behavioural changes superficially resemble the increased desire to gamble observed in iatrogenic GD, and the molecular correlates observed may provide novel insight into the mechanism by which dopamine agonists effect such changes, and therefore through which these psychiatric side-effects could be reversed. From a clinical perspective, one important caveat with respect to the interpretation of the current data set is that we have not used a rodent model of PD.  The loss of dopamine producing neurons in conditions such as PD has been suggested to result in a compensatory increase in the sensitivity of receptors- so called ‘denervation supersensitivity’ (Lee et al. 1978)- that may influence the response to dopamine agonist treatments.  However, ICD’s have been reported following treatment with dopamine agonists in conditions that are not associated with any perturbations of the dopamine system, such as fibomyaglia, RLS and prolactinoma (Clark and Dagher 2014).  Furthermore, dorsolateral striatal dopamine depletion (a model of early stage PD) did not alter the ability of chronic pramipexole to increase risky choice in rats (Rokosik and Napier 2012).  Hence, it is likely that the ICDs exhibited by PD patients following dopamine agonist therapy can largely be attributed to the actions of the drugs themselves, rather than due to the interaction of these drugs and the pathophysiology of PD.  Nevertheless, future studies utilising an animal modes of PD would be useful in definitely addressing this point.  With regards to the behavioural processes through which dopamine agonist treatment may invigorate rSMT performance, one potential explanation is that augmenting the activity of D2 receptors may have altered the appetitive valence of the sugar reward.  Chronic quinpirole administration has been shown to enhance the reward-facilitating effects of amphetamine with 136  respect to responding for intracranial self-stimulation (Schmidt et al. 2013).  However, although an increase in appetitive motivation might explain the increase in the number of trials completed, we would also expect decreased reward collection latencies and perhaps more rapid responses to illuminated apertures in the stimulus array, neither of which were observed.  Another possibility is that the observed effects arose as a result of increased motor impulsivity, or a general trend toward increased motoric output.  The psychostimulant amphetamine, which potentiates dopamine has been repeatedly demonstrated to increase impulsive action on operant tasks (Cole and Robbins 1987; Zeeb et al. 2009).  Although it is not intuitively obvious as to what aspect of rSMT performance may best be considered “impulsive”, amphetamine leads to increased erroneous attempts to collect reward on 1- and 0-light trials.  These errors are made more rapidly at baseline than collect errors on near-miss trials, and may therefore represent impulsive-like responses.  However, although ropinirole did increase such mistakes in the first week of administration, this effect was only transient, whereas the increase in trials performed persisted for the duration of drug treatment.  Indeed, if anything, ropinirole-treated rats made fewer such errors towards the end of testing.  It is therefore hard to attribute the behavioural pattern caused by chronic ropinirole on the rSMT to a deficit in any specific aspect of impulse control. The behavioural endophenotype exhibited following chronic ropinirole may instead be more related to compulsivity, rather than impulsivity.  The relationship between these two cognitive-behavioural processes is complex.  Traditionally, these multifaceted constructs have been viewed as diametrically opposed, with individuals exhibiting a preponderance of one at the expense of the other, yet more contemporary theories now suggest that the relationship between the two is dynamic and can shift over time (see Fineberg et al. 2010 for discussion).  Thus, it may be reasonable to suggest that both impulsivity and compulsivity represent differing vulnerabilities 137  towards the development of GD.  There is ample evidence to indicate that impulsivity is a precursor for poor decision making, such that high trait impulsivity, or the diagnosis of a concurrent ICD, can enhance vulnerability towards both substance and behavioural addictions (Barrus et al. 2015; Jentsch and Taylor 1999; Michalczuk et al. 2011; Winstanley 2007).  In contrast, the archetypal pathology of aberrant compulsivity, obsessive compulsive disorder (OCD) is rarely co-morbid with GD (Fontenelle et al. 2005).  However, gamblers do score higher on self-report measures of compulsivity (Blaszczynski 1999) and many of the cognitive distortions such as an adherence to ‘lucky’ rituals, that have been suggested as central to the development of GD (Ladouceur et al. 1988; Toneatto et al. 1997), could be considered compulsive in nature, in that they are repetitive stereotyped behaviours performed according to certain rules (Fineberg et al. 2010).  Here, the large increase in the number of trials completed may be indicative of a compulsive like behavioural response.  Importantly, this behavioural change cannot be explained through a general increase in the speed of responding.  Hence, non-task related behaviour, such as grooming or exploration of the operant box, must necessarily have been reduced for the number of trials to have increased so dramatically, potentially indicative of greater task ‘focus’.  Although there was no overall difference in response latency between ropinirole and saline treated animals, the characteristic spike in the time taken to respond at the next aperture in the sequence when the light in the preceding hole remained off was far less pronounced in ropinirole treated animals.  Broadly speaking, goal directed actions are governed by an organisms’ appraisal of the action-outcome associations in effect, in addition to an appraisal of the outcomes’ value, whereas habitual responding is largely driven by simple stimulus-response (S-R) relationships (Balleine and O'Doherty 2010).  As such, the fact the behaviour of ropinirole-138  treated animals was less affected by the illumination status of the previous hole may indicate an increase in S-R control of behaviour.  Genuinely goal directed behaviour should be significantly affected by outcome devaluation, such that manipulations such as satiety or extinction have less impact on habit-based behaviour (Balleine and Dickinson 1998).  Although all rats decreased the number of trials completed when win trials were no longer included in the rSMT, this extinction of performance through non-reward was significantly slower in ropinirole-treated rats, and performance never declined to the level of saline-treated controls.  Thus, although ropinirole-treated animals remained broadly goal directed, the drug seemed to elicit more habit-like and compulsive task engagement.  Our supposition that an increase in compulsivity may underlie animal’s increased game play is potentially bolstered by the fact that most of the significant protein changes detected ex vivo were observed in the dorsal, rather than the ventral striatum (with the exception of a single increase in pCREB within the NAc).  The transition in control of behaviour from ventral to dorsal striatum has been hypothesised to underlie the development of compulsive drug seeking (Everitt and Robbins 2005).  The most robust change we observed was a decrease in phosphorylation, and therefore inhibition, of GSK3β (Beaulieu et al. 2009; Beaulieu et al. 2005).  The critical downstream effects of GSK3β activation that may result in behavioural change are as yet unknown, but GSK3β does play a fundamental role in a variety of functions, including receptor trafficking and cellular plasticity (see Li and Gao 2011 for discussion).  Interestingly, the increase in pGSK3β did not abate following cessation of ropinirole treatment, indicative of long lasting alterations in this pathway.  However, the increase in compulsive engagement with the rSMT – the rise in the number of trials completed – did return to baseline levels within 4 weeks.  This discrepancy between the time-course of the behavioural and molecular changes may 139  suggest that disinhibition of GSK3β is unlikely to cause compulsive rSMT engagement.  However, normalization of the number of trials exhibited by the ropinirole treated group to the saline-treated group was slow, only resolving in the last few sessions.  Whether the pervasive increase in active GSK3β would have eventually abated in line with the behavioural effect is unclear.  Regardless, the most pragmatic way to address issues of causation would be to determine whether the increase in trials resulting from ropinirole administration can be blocked by GSK3β inhibitors.  Predominant activation of the βarrestin/AKT/GSK3β pathway, rather than PKA-dependent intracellular signaling cascades, through ligand-binding at the D2-receptor is thought to only occur during periods of increased dopamine release (Beaulieu et al. 2009).  As such, it may not be surprising to find evidence of activation of this pathway following chronic administration of a dopamine agonist.  Certainly, chronic administration of the D2-like agonist pramiprexole increase the tonic activity of dopamine producing neurons (Chernoloz et al. 2009).  Excess dopamine transmission in idiopathic gambling is less clear, but some reports have suggested that GD patients exhibit augmented dopamine release during gambling episodes (Boileau et al. 2014; Linnet et al. 2011), and this may override the dips in phasic dopamine following unsuccessful outcomes (Schultz et al. 1997), making losses less aversive and promoting continued game play.  However, the effects of steady-state slow release ropinirole in the current study does not support the hypothesis that such drug treatments reduce the impact of loss. Interestingly, there was also a reduction in D1 receptor protein in the dorsal striatum.  As D1 receptors typically signal through Gs receptors, which have broadly opposing effects to Gi receptors, it raises the possibility that the small decreases in total ERK and DARPP and were due 140  to a decrease in D1 receptor signalling.  The reason for the reduction in D1 is unclear, but chronic administration of a D2-like agonist has previously been shown to result in desensitization of autoreceptors (Chernoloz et al. 2009).  G-protein coupled receptors are internalised as a homeostatic mechanism to modulate G-protein mediated signaling in response to agonist stimulation.  Therefore, the decrease in post-synaptic D1 receptors may be a compensatory effect as a result of increased tonic dopamine levels following a loss of autoreceptor regulation (Beaulieu and Gainetdinov 2011).  However, it is unlikely that the decrease in D1 receptors can account for the observed behavioural effects as previous investigations acutely targeting the D1 system did not affect rSMT performance (Winstanley et al. 2011).  In contrast qPCR did not reveal any alterations in either D1 or D2 receptor mRNA in prefrontal regions, which coupled with the western blot data, indicate that gross changes in receptor density contributed little to the behavioural effects seen here. Previous reports have shown that chronic ropinirole augments the activity of the serotonergic system (Chernoloz et al. 2009), but we did not see any changes in 5-HT2A receptor mRNA, a target chosen due to its relative abundance in the PFC and role in decision making (Winstanley et al. 2004b).  However, as we observed large scale changes in downstream signaling molecules associated with dopamine receptor activation, without a change in gross receptor density, we cannot preclude a role for 5-HT here, particularly given the multitude of centrally-located 5-HT receptors.  Although we have not investigated modulation of rSMT performance by selective serotonergic ligands, acute amphetamine administration, which potentiates the actions of all monoamines, had a far less pronounced effect on the rSMT then quinpirole, arguably indicating a more prominent role for dopamine over 5-HT on this task (Winstanley et al. 2011).  141  Ultimately, these data provide novel insight into a potential role for aberrant dopaminergic signaling in the development of compulsive gambling. Chronic activation of D2-like receptors resulted in a compulsive style of engagement on the rSMT putatively as a result of increased activation of the β-arrestin mediated AKT – GSK3β signaling cascade in the dorsal striatum. These results have implications for the treatment of both idiopathic and iatrogenic gambling. Excitingly, lithium chloride is a potent inhibitor of GSK3β and has been extensively used in human subjects for the treatment of affective disorders (Beaulieu et al. 2005; Burgess et al. 2001). Moreover, lithium has previously been explored as a treatment for GD, with some success (Hollander et al. 2005; Pallanti et al. 2002).  Our results indicate that gamblers who principally exhibit compulsive gambling engagement, such as slot machine players, may potentially respond favourably to lithium as a pharmacotherapy.  More generally, these results continue to hint at heterogeneity within GD (Cocker and Winstanley 2015; Limbrick-Oldfield et al. 2013) and highlight the need for endophenotypic animal models that may aid in elucidating differing vulnerabilities.           142        Figure 7.1 Baseline rSMT performance (a,b) Animals showed optimal responding on win trials (1,1,1), choosing to collect the available reward nearly 100% of the time.  Similarly, animals showed a marked preference for the optimal response, now the roll lever, when no lights were illuminated (0,0,0), only responding on the collect lever approximately 15% of the time.  Erroneous collect responses increased to 39.98% when 1 light was illuminated.  However, when two-lights were illuminated in the array, animals responded erroneously on the collect lever at a far greater then chance level (72.38% ± 4.76 (SEM)), indicating that rats, like humans, treat such stimuli as more indicative of a win than a loss, and are hence susceptible to the near-miss effect.  There were no differences between the saline and ropinirole groups. All data shown are the mean across five sessions ± SEM.  143   144  Figure 7.2 Effect of chronic ropinirole on reward expectancy. (a,b) during the first trial bin the D2/3 agonist lead to an impairment in animals performance on the rSMT, driven by an increase in erroneous collection responses on 2- and 1-light trials. (c,d) During the second trial bin ropinirole treated animals continued to show impairments in performance, although the deficits here were observed on trials wherein the last light in the array remained on. (e,f) By the final time bin, in contrast to the first two, ropinirole administration improved animals’ ability to differentiate winning from non-winning outcomes on the rSMT, albeit only at the level of a strong trend. In a similar manner to bin 2 these improvements were seemingly dependant on the status of the last aperture with improvements in performance on trials when the last light set to off. All data shown are the mean across four or five sessions ± SEM.    Figure 7.3 Effect of chronic ropinirole on the number of trials completed. In contrast to the bidirectional effects on reward expectancy, ropinirole administration produced a robust and 145  sustained increase in the number of trials completed. This increase in trials, which could be argued to represent a compulsive style of engagement presented after the first trial bin and continued throughout ropinirole administration including throughout extinction and reinstatement. All data shown as the mean ± SEM.    Figure 7.4 Long term behavioural effects of chronic ropinirole administration. (a) The significant improvement seen during ropinirole administration abated such that there were no significant differences on collect lever choice between either saline or ropinirole treated animals during washout. (b) The robust increase in the number of trials completed during ropinirole was preserved through the majority of the washout period but began to return to baseline and was not significantly elevated in comparison to control animals after 20 sessions. All data shown are the mean across twenty sessions ± SEM.  146   147  Figure 7.5 Summary of changes in protein level in the dorsal striatum in animals treated with chronic ropinirole in comparison to saline. The largest change in the dorsal striatum was a decrease in pGSK that persisted in the washout group, indicating a potent activation of the alternative D2 signaling cascade underlying the observed behavioural effects. Data are expressed as the fold change relative to the mean value of the control group ± SEM. Significant effects are denoted by * (p<0.05) and # (p<0.1). 148   149   Figure 7.6 Summary of changes in protein level in the nucleus accumbens in animals treated with chronic ropinirole in comparison to saline. In contrast to the dorsal striatum few alterations in protein level were observed in the NAc with the exception of a decrease in pCREB. Data are expressed as the fold change relative to the mean value of the control group ± SEM. Significant effects are denoted by * (p<0.05) and # (p<0.1).  Table 7.1 The sequence of the primers used to detect mRNA for target genes using qPCR  Forward (5’ → 3’) Reverse (5’ → 3’) DA D1 GGCCTTTGGGTCCCTTTTGT ATCACGCAGAGGTTCAGAATTGG DA D2 TAAGACGATGAGCCGCAGAA TGAACACACCGAGAACAATGG 5-HT22 CCTCAGCACTCGAGCCAAAC TGGACCGTTGGAAGAGCTTT GAPDH GGTGGACCTCATGGCCTACA GGCCTCTCTCTTGCTCTCAGTATC  Table 7.2A Latency to respond on the collect lever by trial type at baseline and for different time-points during saline administration. Data presented in time (s) shown as mean ± SEM Condition 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Baseline 0.96 ± 0.31   0.79 ± 0.13 0.79 ± 0.16 0.46 ± 0.15 0.73 ± 0.16 0.40 ± 0.09 0.56 ± 0.17 0.16 ± 0.06 Bin 1 0.36 ± 0.05 0.23 ± 0.05 0.65 ± 0.07 0.86 ± 0.08 0.64 ± 0.08 0.46 ± 0.06 0.60 ± 0.07 0.48 ± 0.07 Bin 2 0.54 ± 0.09 0.29 ± 0.05 0.65 ± 0.09 0.73 ± 0.08 0.64 ± 0.07 0.49 ± 0.06  0.55 ± 0.05 0.49 ± 0.06 Bin 3 0.64 ± 0.01 0.87 ± 0.02 0.86 ± 0.02 0.89 ± 0.02 2.18 ± 0.22 0.97 ± 0.03 0.60 ± 0.01 0.93 ± 0.03 Extinction N/A 0.94 ± 0.17 0.87 ± 0.10 0.80 ± 0.09 0.89 ± 0.10 1.15 ± 0.29 1.94 ± 1.00 0.64 ± 0.12 Reinstatement 0.63 ± 0.10 1.11 ± 0.19  0.80 ± 0.11 0.98 ± 0.15 1.07 ± 0.18 1.36 ± 0.46 0.76 ± 0.13 0.72 ± 0.18 Washout 0.45 ± 0.03 0.83 ± 0.11 0.59 ± 0.06 0.51 ± 0.10 0.51 ± 0.06 0.55 ± 0.20 0.35 ± 0.06 0.12 ± 0.04   150  Table 7.2B Latency to respond on the collect lever by trial type at baseline and for different time-points during ropinirole administration. Data presented in time (s) shown as mean ± SEM Condition 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Baseline 0.70 ± 0.11 0.91 ± 0.12 0.70 ± 0.11 0.53 ± 0.08 0.69 ± 0.10 0.45 ± 0.09 0.50 ± 0.11 0.33 ± 0.08 Bin 1 0.58 ± 0.06 0.28 ± 0.08 0.76 ± 0.15 0.82 ± 0.08 0.74 ± 0.08 0.57 ± 0.14 0.80 ± 0.08 0.64 ± 0.10 Bin 2 0.48 ± 0.11 0.17 ± 0.03 0.65 ± 0.09 0.51 ± 0.04 0.57 ± 0.08 0.3 ± 0.04 0.61 ± 0.07 0.35 ± 0.05 Bin 3 0.56 ± 0.12 0.47 ± 0.05 0.53 ± 0.07 0.57 ± 0.07 0.56 ± 0.08 0.88 ± 0.21 0.72 ± 0.32 1.11 ± 0.57 Extinction N/A 0.54 ± 0.07 0.49 ± 0.05 0.92 ± 0.36 0.51 ± 0.06 0.65 ± 0.09 0.82 ± 0.14 0.70 ± 0.16 Reinstatement 0.53 ± 0.09 0.64 ± 0.12 0.55 ± 0.09 0.43 ± 0.09 0.56 ± 0.10 0.49 ± 0.08 0.46 ± 0.09 0.30 ± 0.08 Washout 0.56 ± 0.06 0.51 ± 0.05 0.57 ± 0.07 0.23 ± 0.04 0.61 ± 0.08 0.30 ± 0.04 0.35 ± 0.07 0.10 ± 0.03  Table 7.3A Latency to respond at subsequent hole based on the statues of the previous hole for baseline and different time-points during saline administration. Data presented in time (s) shown as mean ± SEM Drug Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Baseline  0.89 ± 0.12 1.62 ± 0.36 0.97 ± 0.13 1.76 ± 0.37 Bin 1 0.88 ± 0.06 2.09 ± 0.34 0.92 ± 0.08 1.82 ± 0.21 Bin 2 0.81 ± 0.06 2.33 ± 0.56 0.89 ± 0.04 1.53 ± 0.13 Bin 3 1.55 ± 0.42  3.05 ± 0.96 1.10 ± 0.14 2.31 ± 0.40 151  Drug Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Extinction 0.88 ± 0.15 3.52 ± 0.51 1.06 ± 0.14 3.32 ± 0.53 Reinstatement 0.86 ± 0.13 1.66 ± 0.39 0.93 ± 0.13 1.57 ± 0.35 Washout 0.92 ± 0.29 2.04 ± 0.40 0.83 ± 0.08 1.61 ± 0.33  Table 7.3b Latency to respond at subsequent hole based on the statues of the previous hole for baseline and different time-points during ropinirole administration. Data presented in time (s) shown as mean ± SEM Drug Latency at Hole 3 when H2 on Latency at Hole 3 when H2 off Latency at Hole 4 when H3 on Latency at Hole 4 when H3 off Baseline 1.10 ± 0.15 2.07 ± 0.46 1.42 ± 0.32 1.80 ± 0.35 Bin 1 1.77 ± 0.32 2.74 ± 0.68 1.57 ± 0.26 3.30 ± 0.73 Bin 2 1.01 ± 0.10 1.35 ± 0.21 0.96 ± 0.08 1.18 ± 0.08 Bin 3 0.82 ± 0.53 1.11 ± 0.10 0.91 ± 0.07 1.32 ± 0.09 Extinction 1.29 ± 0.25 3.15 ± 0.44 1.06 ± 0.09 2.69 ± 0.47 Reinstatement 1.26 ± 0.16 2.33 ± 0.56 1.72 ± 0.42 2.05 ± 0.47 Washout 0.83 ± 0.09 1.21 ± 0.17 0.73 ± 0.05 1.22 ± 0.11   Table 7.4 Changes in mRNA or protein in rats treated with chronic ropinirole or saline from both early and late time points. Data are expressed as fold change from control ± SEM  OFC PFC  Ropinirole Saline Ropinirole Saline  Early Late Early Late Early Late Early Late DA D1 1.26 ± 0.29 1.25 ± 0.44 1.05 ± 0.17 2.83 ± 2.28 0.91 ± 0.17 0.57 ± 0.57 0.44 ± 0.44 1.09 ± 0.29 DA D2 1.25 ± 0.29 1.24 ± 0.44 1.05 ± 0.17 2.83 ± 2.28 1.15 ± 0.22 1.07 ± 0.21 1.22 ± 0.35 1.31 ± 0.42 5-HT2a 1.12 ± 0.22 1.06 ± 0.19 1.45 ± 0.56 1.49 ± 0.80 1.05 ± 0.14 1.00 ± 0.06 1.04 ± 0.13 1.04 ± 0.17 152  Chapter 8: Experiment 6: The β-adrenoreceptor blocker propranolol ameliorates compulsive-like gambling behaviour on a rodent slot machine task: implications for iatrogenic gambling.    8.1 Introduction  Parkinson’s disease is a progressive neurodegenerative disorder characterised by the loss of dopaminergic neurons in the substantia nigra (Fahn et al. 2004). The dopamine precursor levodopa (L-DOPA) is the benchmark treatment for PD.  However, over time the clinical effectiveness of L-DOPA wanes (Fahn 1999; Olanow et al. 2009). Adjunctive therapies with dopamine agonists successfully remediate the motoric symptoms of PD, but a small but significant sub-set of patients treated with these drugs develop impulse control or addictive disorders such as punding, compulsive sexual behaviour and gambling disorder (Voon et al. 2007b; Weintraub et al. 2015).  These psychiatric conditions have severe consequences on patients quality of life and as such limit the efficacy of dopamine agonist therapies, leading to a greater reliance on L-DOPA, with all of the associated motor side-effects and limited long-term efficacy (Calabresi et al. 2010).  A better understanding of the neurobiological aetiology underlying the development of iatrogenic impulse control/gambling disorders could theoretically lead to the development of efficacious adjunctive treatments, designed to preserve the therapeutic benefit of dopamine agonists for the motor symptoms of PD while negating the psychiatric side-effects. Animal models that recapitulate these dopamine-agonist induced increases in gambling-like behaviour could make a significant contribution in this regard.  153  Rodent models have previously demonstrated that repeated administration of D2-like agonists induce compulsive-like behaviour and augment impulsive decision making  (Eagle et al. 2014; Rokosik and Napier 2012; Szechtman et al. 1998). In regards to gambling-related decision making, we have demonstrated using the rSMT that rats share key behavioural features with human gamblers (Winstanley et al. 2011). Insofar as cues predictive of future reward are able to evoke reward expectancy in animals, in a similar manner to humans (Clark et al. 2012; Habib and Dixon 2010; Sescousse et al. 2016). Animals reward expectancy on the rSMT is critically mediated by the DA D2-like receptor family (Cocker et al. 2016; Cocker et al. 2014; Winstanley et al. 2011). Previously, we demonstrated that chronic administration of the D2-like agonist ropinirole, a drug that has previously been associated with iatrogenic GD in PD (Dodd et al. 2005), induces compulsive-like task engagement on the rSMT (Cocker 2016). This invigoration of behaviour could be considered analogous with the compulsive play exhibited by PD patients with ICD’s. However, these results were in intact animals. The loss of dopaminergic neurons has been suggested to result in an compensatory increase in the sensitivity of the remaining dopamine receptors, so-called ‘denervation supersensitivity’ (Lee et al. 1978). Thus, from a clinical perspective, it is unclear if this dopamine mediated increase in compulsive-like gambling engagement is applicable to iatrogenic GD in PD.  Additionally, we demonstrated using ex vivo analysis of striatal tissue that pseudo-pathological invigoration of rSMT performance was concomitant with an increase in the active form of GSK3β in the dorsal striatum and phosphorylated (active) cyclic adenomonophosphate (cAMP) response element binding protein (pCREB) in the nucleus accumbens.  These findings lead us to hypothesise that targeting one, or both, of these proteins could represent a novel treatment target for iatrogenic GD.  154  Both CREB and GSK3β have been implicated in a broad range of functions including modulating learning and memory (Beaulieu et al. 2007; Carlezon et al. 2005; Li and Gao 2011). Both are activated by dopamine and contribute to subjective responsivity to drugs of abuse such as cocaine (Beaulieu et al. 2007; Enman and Unterwald 2012; Miller et al. 2009; Nestler and Carlezon 2006; Self et al. 1998). However, any role for either protein in controlling gambling-related decision making, has to our knowledge, not been investigated.    Direct modulation of these signaling proteins in-vivo is difficult; although compounds exist that selectively inhibit the activity of both proteins, these approaches are currently unfeasible in human subjects and consequently lack utility as potential treatment options. However, existing pharmacotherapies that are, for the most part, safe and well tolerated could be utilised to modulate these intra-cellular signaling cascades. Li+, commonly used to treat affective disorders, is a potent GSK3β inhibitor and interestingly has shown some limited efficacy as a pharmacotherapy for gambling disorder (GD) (Beaulieu et al. 2005; Hollander et al. 2005; Pallanti et al. 2002) However, it is unclear if Li’s therapeutic potential as a GD treatment is derived from inhibiting GSK3β, or rather due to palliative effects on co-morbid affective disorders (van den Brink 2012).  The β-adrenoreceptor blocker propranolol, commonly used as a prophylactic treatment for migraine and also to treat hypertension, inhibits the phosphorylation of CREB in the NAC (Kabitzke et al. 2011).  There is some limited evidence that propranolol can alter risk-reward decision making (but see Campbell-Meiklejohn et al. 2011; Corwin et al. 1990; Rogers et al. 2004), although any efficacy as a treatment for iatrogenic or idiopathic gambling is currently unknown.  Here, we will initially determine if dopamine depletion in the dorsal striatum, an established model of early-stage PD in the rat, potentiates the compulsive-like gambling 155  behaviour following chronic administration of a D2-like agonist.  Secondly, we will investigate whether administration of Li or propranolol can attenuate the pseudo-pathological phenotype induced by ropinirole administration.    8.2 Additional methods Bilateral dorsolateral striatal 6-OHDA lesion surgery  Animals were divided into groups based on their baseline performance. Rats subsequently received either bilateral dopamine depletions of the dorsolateral striatum or sham lesions using standard stereotaxic techniques. Animals were anesthetized using 2% isoflurane in O2 and administered anafen and buprenorphine for systemic and local analgesia, respectively. Surgical coordinates were based on previous reports (Baunez et al. 2007); anterior posterior: +0.2mm; medial/lateral: ± 3.5mm and dorsal/ventral: -4.5mm taken from bregma, mid-line and skull respectively, with the incisor bar set to -3.3mm. Lesions were performed via infusion of the neurotoxin 6-hydroxydopmaine hydrochloride (6-OHDA), 12µg/3µl dissolved in sterile saline containing 0.2mg/ml ascorbic acid (Sigma-Aldrich, Oakville, Canada) made fresh daily. Sham lesions consisted of 3µl of vehicle. Infusions were administered in a volume of 3µl over 6 minutes at a rate of 0.5µl a minute using a duel-channel infusion pump (Harvard Apparatus, Holliston, USA). Injectors were left in place for an additional 2 minutes to allow the infusate to diffuse away from the injection site.  Animals were allowed to recover in their home cage for a week before behavioural testing resumed. During this time, water was available ad libitum and rats were fed 20-25g of standard chow per day.  Motor assessment: forelimb adjusting step test 156  6-OHDA lesions of the dorsal striatum reliably induce deficits on the Forelimb adjustment step test (Olsson et al. 1995; Rokosik and Napier 2012). Therefore, in order to parametrize both the deficits following 6-OHDA lesions and any therapeutic benefits of ropinirole, animals were given a forelimb adjusting step test 4-weeks post 6-OHDA surgery and again following 4-weeks of chronic ropinirole treatment.  The forelimb adjustment test was carried out in a similar manner to previous reports (Rokosik and Napier 2012), in brief, the experimenter suspended the rats rear legs and one forelimb requiring the rat to support itself on its unrestrained forelimb.  The rat was then 'dragged' on its unrestrained forelimb for 0.9m over 5 seconds in both abduction and adduction directions. Measurements were taken twice for each forelimb in each direction and the average number of adjusting steps was calculated. Lithium chloride administration  Following recovery from osmotic mini-pump implantation, animals were again matched for prior task performance and separated into groups. Half of the animals receiving ropinirole and half receiving saline were fed standard rat chow that had been reconstituted with 0.3% Li+ chloride added.  The remaining animals were fed a matched control diet (Envigo, Maddison, WI). Animals were fed 14g/rat/day immediately following behavioural testing, and administration continued for 14-days.  Throughout Li+ administration animals had ad libitum access to 0.9% saline to minimise any ionic imbalances that may have arisen as a result of the diuretic properties of, and potential toxicity associated with, Li+ administration (Maletzky and Shore 1978; Smith and Amdisen 1983). This protocol, when used chronically, has previously been demonstrated to result in therapeutically relevant levels of circulating Li+ (Levesque 2013).  Measurement of serum lithium levels 157  Following two weeks of Li+ diet administration blood samples were taken to measure plasma Li+ levels. Blood samples were taken from the jugular vein whilst animals were under anaesthesia (2% isoflurane in O2). Samples (2ml) were taken immediately after behavioural testing. Plasma Li+ concentrations were determined using colorimetric analysis conducted using a Dimension Xpand plus integrated chemistry system (Siemans, Erlangen, Germany) at the UBC hospital laboratory. Animals were allowed to recover for 2-days in their home cage before testing resumed.  Propranolol administration  During the final 7-days of ropinirole administration, the effects of repeated injections of the β-adrenoreceptor antagonist propranolol were assessed. The groups established during diet manipulations were reversed, such that animals that had previously received the control diet were administered propranolol, and animals previously fed a diet containing Li+ received saline injections.  Propranolol (0.5mg/ml/kg) (Sigma-Aldrich) was made up fresh daily and dissolved in sterile saline.  The dose was calculated as the salt, and selected as it is behaviourally silent on the 5CSRT, but attenuates yohimbine induced impulsivity (Adams 2016). All injections were delivered via the intraperitoneal route and administered 10-minutes prior to behavioural testing.  Data analyses:  The effect of 6-OHDA lesions were analysed with similarly structured ANOVA’s with group (2-levels; 6-OHDA vs. sham) as a between subjects’ factor. During ropinirole administration data was grouped into session bins. Animals completed 12 sessions on the rSMT with either Li+ or control diet, these sessions were separated into bin 1 (sessions 1-4), bin 2 (sessions 5-8) and bin 3 (sessions 9-12). Subsequently animals completed a further 5 sessions of the rSMT with either propranolol or saline administration. All ropinirole analyses remained the 158  same as above with the exception that drug (2-levels: ropinirole vs. saline) was added as a between subjects’ factor. During bins 1-3 diet (2-levels: Li+ vs. control) was also used as a between subjects’ factor, whereas for the propranolol sessions, injection type (2-levels; propranolol vs. saline) was used instead. Where applicable follow-up analyses used independent samples T-tests to examine differences between drug conditions. Follow-up analyses were only conducted when there was a main effect on the original ANOVA. Data from the forelimb adjustment step test was analysed using an independent samples T-Test with surgery (6-OHDA or sham) as a grouping variable. Based on the a priori assumption that ropinirole would differentially alter performance in lesioned, but not intact animals, data from both ropinirole and saline treated animals were analysed separately via independent samples T-tests.    8.3 Results Sixteen animals were excluded from the baseline analysis due to failure to meet established performance criteria, defined as at least 50 trials completed and >50% accuracy on clear loss trials ({0,0,0}). The total number of animals included in the baseline analysis was therefore 50. These animals were matched for performance and divided into two groups, such that 25 animals received bilateral 6-OHDA infusions into the dorsal striatum, the other 25 received sham surgery.  Two animals from the 6-OHDA group developed post-surgical complications following stereotaxic surgery and were euthanized, yielding a total of 48 animals for lesion analysis. A further two animals experienced post-surgical complications following mini-pump implantation, thus the total number of animals included in all subsequent analysis was 46.     159  Baseline behaviour  Prior to 6-OHDA lesions animals were matched for their baseline performance and split into two matched groups. There were no statistical differences between these groups on any behavioural variable and consequently the baseline data for both groups was pooled (surgery group: all F’s <1.42). Animals choice of the collect lever varied significantly across the different trial types (Fig 8.1a: trial type: F7,336 = 216.26, p<0.0001). Similar to previous reports the likelihood of erroneous collect lever responses decreased contingent with the number of illuminated apertures within the array (Fig 8.2b: lights illuminated: F3,144 = 706.83; 3- vs. 2-lights: F1,48 = 167.14, p<0.0001; 2- vs. 1-light: F1,48 = 455.75, p<0.0001 1- vs. 0-lights: F1,48 = 112.67, p<0.0001).  In contrast to previous reports however, there were significant differences in animals collect lever responses between the different types of 2- and 1-light trials (2-light trials: F2,96 = 2.96, p=0.06; 1-light trials: F2,96 = 16.89, p<0.0001).  Animals appeared more likely to attempt to collect reward when the last light in the array remained on, indicating that reward related signals presented spatially proximal to the animals’ decision point were able to disproportionately control behaviour (last light status: F1,48 = 74.73, p<0.0001). However, the total number of illuminated apertures in the array still appeared to be a better predictor of animals’ lever choice behaviour, as animals were still far more likely to select the collect lever on 2-light trials when the last light stayed off, in comparison to 1- light trials when the last stayed on (({1,1,0} vs {0,0,1}) trial type: F1,48 = 27.45, p<0.0001). The latency to respond on the collect lever varied significantly by trial type (table 8.1: trial type: F7,336 = 7.49, p<0.0001). Erroneous responses were significantly slower on near-miss trials in comparison to any other trial type, indicating that these sorts of trials evoked more cognitive deliberation (lights illuminated: F3,144 = 5.09, p=0.002; 3- vs 2-lights: F1,48 = 31.83, p<0.0001 2- vs. 1-light: F1,48 = 61.29; 1- vs. 0-lights: 160  F1,48 = 0.71, NS). In concordance with previous reports, animals were quicker to respond at the subsequent hole if the light in the preceding hole had remained on, indicating that win-related served to invigorate performance (table 8.2) (light status: F1,48 = 27.09, p<0.0001). Animals completed an average of 70.20 ± 1.69 trials per session (Table 8.3: sham group: 70.03 ± 2.10; 6-OHDA group 70.38 ± 2.68).   Effect of 6-OHDA lesions in the dorsal striatum Dopamine depletion in the dorsal striatum via infusion of 6-OHDA infusions did not alter performance on the rSMT (Fig 8.2a, b: surgery group: F1,46 all F’s<1.45, NS). As there were no statistical differences in performance between 6-OHDA lesioned and intact animals, animals from both groups were pooled together for all subsequent analysis. Effect of chronic administration of the D2/3 agonist ropinirole and dietary lithium on rSMT performance Animals fed a diet containing 0.3% Li+ chloride had serum Li+ levels of 0.32 mmol/l ± 0.02 (SEM). No meaningful correlations were observed between subjective Li+ levels and any behavioural measures (data not shown).  Similar to our initial characterisation of the effects of ropinirole on task performance (Cocker 2016), animals displayed an increase in erroneous collect responses during the first two session bins that abated over time (Fig 8.3a, c, e: drug X trial type: bin 1: F7,266 = 2.25, p=0.03; bin 2 F7,266 = 2.56, p=0.01; bin 3: F7,266 = 0.76, NS).  Again consistent with our previous data, these alterations in performance did not appear contingent on the number of lights illuminated (Fig 8.3b, d, f: drug X lights illuminated: bin 1:  F3,114 = 0.67, NS; bin 2: F3,114 = 0.99, NS; bin 3: F3,114 = 0.24, NS) but were affected by the illumination status of the last light, such that ropinirole-treated rats were more likely to make an erroneous collect response during the first two session bins if the last light set to on (drug X last light status: bin 1: 161  F1,38 = 5.28, p=0.03; bin 2; F1,38 = 4.98, p=0.03: bin 3; F1,38 = 2.0, NS).  Neither prior surgery or dietary Li+ administration altered animals’ choice of the collect lever (surgery group: all F’s <0.06, NS: Fig 8.3a-f: diet group: all F’s <1.86, NS).   In contrast to our previous dataset, ropinirole administration produced a general increase in the speed with which animals selected the collect lever, and this varied between different trial types only in the first session bin, and to a lesser extent in the third (table 8.1: drug X trial type: bin 1; F7,266, = 3.02, p=0.005; bin 2: F7,266 = 0.95, NS; bin 3: F7,266 = 1.76, p=0.01). In contrast during session bins 2 and 3 the D2/3 agonist produced a general quickening of responding towards the collect lever (table 8.1: drug group: bin 1 F1,38 = 0.07, NS; bin 2: F1,38 = 4.31, p=0.05; bin 3: F1,38 = 11.38, p=0.002). Animals response latencies were not altered by either prior surgery or Li+ administration (surgery group: all F’s <0.78, NS: diet group: all F’s <1.2, NS).  The tendency animals displayed at baseline to respond quicker at the apertures if the preceding light had set to on was still present during chronic ropinirole or saline administration (table 8.2: light status: bin 1: F1,38 = 27.84, p<0.0001; bin 2: F1,38 = 19.26, p<0.0001; bin 3: F1,38 = 16.08, p<0.0001) yet chronic ropinirole treatment produced a hastening of responding, both when the previous light remained illuminated and when it was not, although this main effect of drug had dwindled by the third session bin (table 8.2: drug group: bin 1: F1,38 = 6.1, p=0.02; bin 2: F1,38 = 6.12, p=0.02; bin 3: F1,38 = 2.6, NS: drug X light status: bin 1: F1,38 = 5.85, p=0.02; bin 2: F1,38 = 2.88, p=0.1; bin 3 F1,38 = 2.95, p=0.09).  This is, again, largely consistent with our previous findings.  Although there was a group interaction for animals’ prior surgery and diet on animals’ latency to respond at the apertures during the second two session bins, this primarily appeared due to increased latency in Li+ treated animals receiving saline (surgery X drug X diet; bin 1: F1,38 = 1.95, NS; bin 2: F1,38 = 6.12, p=0.02; bin 3: F1,38 = 6.01, p=0.02). There were no main 162  effects observed for either surgery group or diet in isolation (surgery group: all F’s<2.14, NS; diet group: all F’s<1.77, NS).     Consistent with previous reports (Cocker 2016), animals’ engagement with the task was substantially invigorated following chronic ropinirole administration in comparison to saline treated controls, as indicated by a significant increase in the number of trials completed (Fig 8.5: drug group: bin 1: F1,38 = 14.02, p=0.001; bin2: F1,38 = 32.73, p<0.0001; bin 3: F1,38 = 39.33, p<0.0001).  This increase manifested during the first bin before plateauing during the second and third (drug X session: bin 1 F3,114 = 6.18, p=0.001; bin 2: F3,114 = 1.72, NS; bin 3: F3,114 = 1.0, NS). Neither striatal dopamine depletion nor administration of Li+ affected this elevation in trials completed induced by chronic ropinirole (surgery group: F’s<0.22; diet group: F’s< 2.27, NS). However, there was a significant interaction between surgery group, diet and ropinirole throughout Li+ administration (surgery X drug X diet group: bin 1: F1,38 = 7.23, p=0.01; bin 2: F1,38 = 11.15, p=0.002; bin 3: F1,38 = 4.44, p=0.04).  Unfortunately, this does not appear to be an overly meaningful effect.  Fewer animals were chronically administered saline, thus when animals were parsed into groups based on prior surgery, drug and diet treatment, there were relatively few animals in each saline condition.  Variability within these saline groups appeared to be driving this interaction (surgery X diet group: just ropinirole treated: all F’s <2.58, NS; just saline treated: bin 1: F1,28 = 7.12, p=0.01; bin 2: F1,28 = 11.85, p=0.002; bin 3: F1,28 = 6.88, p=0.01).    Effect of chronic administration of the D2/3 agonist ropinirole and co-administration of the β-adrenoreceptor antagonist propranolol  Behaviour during the 5-sessions of propranolol administration was, in large part, consistent with the behaviour reported during dietary Li+.  Chronic ropinirole or saline 163  administration did not produce differing effects on erroneous collect responses (fig 8.4a, b: drug X trial type: F7,259 = 0.53, NS).  Likewise, ropinirole maintained a facilitatory effect on the latency to select the collect lever (table 8.1: drug group: F1,38 = 4.94, p=0.03) and respond at the apertures (table 8.2: drug group: F1,38 = 5.95, p=0.02).  The co-administration of propranolol slowed animals responding across different trial types, although only at a trend level (injection X trial type: F7,259 = 1.94, p=0.06) but did not alter any other measure (injection group: all F’s <0.2, NS).  There were similarly no interactions as a function of animals’ prior surgical group (surgery group: all F’s < 0.53, NS).  As during Li+ administration, ropinirole continued to invigorate task performance, leading to an elevated number of trials completed in comparison to saline treated controls (Fig 8.5: drug group: F1,38 = 24.37, p<0.0001).  However, in contrast to Li+, injection of propranolol served to partially attenuate the number of trials completed (Fig 8.5: injection group: F1,38 = 3.39, p=0.07). This decrease in trials was selective to ropinirole treated animals (injection group: ropinirole treated: F1,28 = 3.59, p=0.07; saline treated: F1,10 = 0.98, NS). Interestingly, the effects of propranolol varied in ropinirole treated, but not saline treated animals during the 5-days of administration (session X injection: ropinirole treated: F4,112 = 7.97, p<0.0001; saline treated: F4,40 = 0.24, NS). Examining these effects further; independent samples t-tests revealed that propranolol lead to an attenuation of ropinirole induced invigoration, to at least a trend level during the first 3 administration sessions (ropinirole treated: injection: session 1: T30 = -1.68, p=0.1 session 2: T30 = -2.32, p=0.03; session 3: T30 = -1.73, p=0.09 session 4: T30 = -1.14, NS session 5: T30 = -1.14, NS).  Forelimb adjustment step test To verify that 6-OHDA lesions of the dorsal striatum induced Parkinsonian-like deficits both lesioned (n=23) and sham (n=25) animals were tested for forelimb stepping 4-weeks post 164  surgery. 6-OHDA lesioned animals made significantly fewer adjusting steps in comparison to intact animals (Fig 8.6a: T46 = -4.45, p<0.0001). These motor deficits were ameliorated following treatment with ropinirole (Fig 8.6b: T20 = -2.64, p=0.02), but did not result in any alterations in motor behaviour in the sham surgery group (Fig 8.6b: T22 = -0.8, NS).   8.4 Discussion Here, we provide further evidence that chronic administration of the D2-like agonist ropinirole invigorates behaviour on the rSMT. This robust and sustained increase in the number of trials completed could be considered as a behavioural phenotype for modelling compulsive gambling. Further, we demonstrate that this compulsive-like behavioural response was partially attenuated by a relatively low dose of the β-adrenoreceptor antagonist propranolol, as predicted by our previous analyses suggesting that a drug capable of inhibiting pCREB in the NAC may reverse the effects of ropinirole.  In contrast, dietary administration of the potent GSK3β inhibitor Li+ was without effect.  These findings suggest that propranolol may be an effective therapeutic strategy in targeting iatrogenic ICD’s that arise in PD patients following treatment with dopamine agonist therapies. Additionally, these data indicate that pharmacotherapies aimed at the adrenergic system may represent a novel avenue for developing treatments for particularly compulsive forms of gambling engagement.    One of the goals of this study was to elucidate whether the compulsive-like effects of chronic ropinirole on the rSMT would be augmented by the supposed denervation supersensitivity that arises during PD (Lee et al. 1978). In other words, do the neurobiological sequelae of PD confer vulnerability toward ICD’s, or do these behavioural perturbations arise de novo following dopamine agonist therapy.  Here, we used 6-OHDA lesions of the dorsolateral 165  striatum, a commonly used technique in the rat that been shown to mimic the loss of dopaminergic terminals similar to that observed in early PD (Baunez et al. 2007; Blesa et al. 2012; Deumens et al. 2002; Przedborski et al. 1995). 6-OHDA lesions did not alter animals’ performance on the rSMT, or response to ropinirole, in any way, a finding that is consistent with other investigations examining the effect of dopamine depletions on probabilistic decision making in rats (Rokosik and Napier 2012). It is worth noting that 6-OHDA lesioned animals showed persistent forelimb motor impairments that were remediated following ropinirole administration, intimating dopamine depletions did produce motoric deficits characteristic of PD. These data would imply that ICD’s arise directly as a result of treatment with dopamine agonists, as oppose to any precipitating vulnerability conferred by PD. Such a supposition is congruent with reports of ICD’s in conditions treated with dopamine agonists, but no loss of endogenous dopamine function such as fibromyalgia, prolactinoma and restless leg syndrome (Clark and Dagher 2014). The lack of effects observed following administration of dietary Li+ were somewhat surprising.  We have previously shown that chronic ropinirole robustly activates the β-arrestin mediated AKT-GSK3β intra-cellular signaling cascade (Cocker 2016). This activation was observable as a large and sustained decrease in pGSK3β; GSK3β is constitutively active, therefore a decrease in the inactive (phosphorylated form) demonstrates increased activation within the pathway (Beaulieu et al. 2009). Li+ is a potent GSK3β inhibitor and has been used, with at least partial success, to treat GD  (Beaulieu et al. 2005; Hollander et al. 2005; Pallanti et al. 2002), thus it seemed probable given our previous findings that it would likely attenuate the compulsive-like task involvement observed following ropinirole. However, 0.3% Li+ administered in the animals’ diet, a dosing regime previously shown to improve visual spatial 166  attention and attenuate yohimbine induced impulsivity (Levesque 2013), had no effect on the rSMT.  Ergo, the robust increase in the active form of GSK3β following chronic ropinirole does not appear to contribute to compulsive-like task engagement. It should be noted that the level of serum Li+ in the animals was slightly lower, although approaching, the well established therapeutic range of 0.4-1.5mmol/l (see Severus et al. 2008 for review). However, this therapeutic index is related to the treatment of bipolar disorder, not the inhibition of GSK3β. Li+ has a very high affinity for GSK3β and thus it is likely that, even at the slightly reduced concentrations obtained here, it would still produce an inhibition of GSK3β activity, although we have no direct measure here (Davies et al. 2000). Significantly, examining data only from animals who were within the specified therapeutic range, or correlating serum Li+ levels with behaviour, did not reveal any relationships with alterations in behaviour. Additionally, preliminary data from our lab shows that administration of the selective GSK3β inhibitor SB-216763 does not ameliorate the effects of ropinirole on a rodent betting task (Tremblay 2016). Thus, although chronic ropinirole reliably and potently activates the β-arrestin mediated signaling cascade, the resultant increase in active GSK3β does not appear to have any overt behavioural effects.   Another potential limitation related to Li+ administration, and a possible factor that may be informative in elucidating why some animals did not attain ‘therapeutic’ serum levels, is that Li+ is aversive and animals will not readily consume it. Certainly, it took several days of exposure before animals began to eat the entire daily amount of 14g chow/rat. Consequently, any attenuation of ropinirole-induced compulsivity by Li+ administration may have been masked by a compensatory increase in appetitive motivation to supplement reduced chow intake. However, such an explanation is unlikely for a number of reasons. Firstly, there were no changes in the 167  number of trials completed throughout Li+ administration, despite animals consuming the entire amount as administration progressed. Secondly, amongst animals implanted with an osmotic mini-pump delivering saline, there were no statistical differences between animals maintained on either Li+ or control diet.  Finally, there were no differences in latencies to respond at the array, collect reward or respond on the collect lever between Li+ and control diet, which, if Li+ induced hunger was increasing the incentive motivation of food reward, we would arguably expect.   In contrast propranolol was, at least partially, successful in remediating the ropinirole induced increase in the number of trials completed. The transitory nature of the decrease in trials following propranolol administration may raise concerns in regards to the therapeutic potential of this intervention.  However, the method of delivery may have lead to the short-lived effects, as the pulsatile nature of acute injections could have lead to compensatory changes in β-adrenoreceptor availability or sensitivity.  However, β-adrenoreceptor blockers such as propranolol undergo extensive first pass metabolism, and thus a continuous delivery system may have lead to lower levels of bioavailability (Shand 1976). Another possibility is that the dose was too low to ameliorate the already pronounced effects of ropinirole. The dose chosen had previously been shown to inhibit the pro-impulsive effects of yohimbine, but in that case yohimbine was administered acutely (Adams 2016), whereas here ropinirole was chronically administered and had been ‘on-board’ for several weeks.  Nevertheless, these initial effects are encouraging and warrant further studies to ascertain whether more effective dosing regimes of propranolol could completely attenuate the compulsive behavioural dysfunction observed following DA agonist therapies.  Propranolol, in addition to blocking β1 and β2 receptors, also has inhibitory effects on the norepinephrine transporter (NET) and reduces synthesis of tyrosine hydroxylase (Tuross and 168  Patrick 1986). Thus the effects observed following propranolol may have arisen as a result of these more direct effects at adrenergic receptors, rather than through reducing pCREB in the NAC.  However, acute administration of amphetamine, which inhibits catecholamine transporters, produced less pronounced effects on rSMT performance then the D2-like agonist quinpirole, which would indicate that any effects of propranolol are likely to be indirect (Winstanley et al. 2011).   Lastly, the selective attenuation in the number of trials completed could have arisen as a result of non-specific motor effects, although, as mentioned, the dose of propranolol used was selected specifically as it is behaviourally silent on the 5CSRTT (Adams 2016). Nevertheless, differing pharmacological compounds can have pronounced differences dependant on the behavioural task used (see Cocker and Winstanley 2015 for discussion), so a general motoric effect of propranolol cannot be excluded based on the dose chosen alone. However, critically, there were no differences in the number of trials completed in the animals chronically administered saline before repeated injection of either propranolol or saline. Moreover, there were no alterations in the latency for animals to respond either at the apertures or the levers following administration of propranolol, indicating that the reductions in trials was likely not due to general motor sedation.  Both here, and in previous reports, chronic ropinirole lead to a robust and sustained increase in the number of trials animals completed on the rSMT (Cocker 2016). We previously showed that this increase was accompanied by significant changes in two intracellular signaling proteins, pCREB in the NAC and GSK3β in the dorsal striatum. Here, chronic administration of dietary Li+ had no effect on task performance, but repeated injection of propranolol did partially attenuate the increase in trials completed. Ergo, these data imply that the behaviourally 169  significant intra-cellular signaling cascade is the increase in pCREB within the NAC. The process through which pCREB is fostering this potentially compulsive-like task engagement is ultimately unclear.  However, one potential explanation could be related to CREB’s established role in learning and memory (or Kandel 2012 for review; see Silva et al. 1998).  CREB is stimulated in the NAC by exposure to stimuli that promote the release of dopamine (Carlezon et al. 2005). CREB activity has been shown to modulate intrinsic activity within the NAC and appears to gate responsivity to both appetitive and aversive stimuli (Carlezon et al. 2005; Dong et al. 2006), with increases in CREB decreasing responsivity to both (Barrot et al. 2002). Theoretically, therefore, chronic exposure to ropinirole would lead to an increase in CREB and could prompt some form of cognitive reappraisal in regards to the salience of task-associated cues (Nestler and Carlezon 2006). Critically, the effects of CREB on mediating salience attributions may be compounded by an increase in tonic dopamine levels. Dopaminergic neurons fire in response to stimuli that predict reward, and also show a dip in cellular activation if a predicted reward is not delivered, or worse then expected.  Chronic dopamine agonism increases tonic dopamine levels, potentially obfuscating these temporary dips and impeding learning from negative outcomes (Goto et al. 2007; Grace 1991; Schultz 1998). Thus, a blunted sensitivity to environmental cues as a result of a chronic increase in pCREB, in conjunction with a failure to detect dips in dopamine following worse then expected outcomes, may render animals less sensitive to losing outcomes. Certainly, intact animals on the rSMT appear to experience loss trials as aversive or frustrating and such trials prompt disengagement with the task (Cocker et al. 2016). The increase in the number of trials completed, the decreased response latencies following stimuli indicative of a loss, and the general invigoration of behaviour following chronic ropinirole, could all suggest that cues predicting negative outcomes, that previously led 170  to animals disengaging from the task, have lost the potency to alter behaviour.  Inhibiting pCREB through propranolol administration may have restored animals’ attributions of salience to game-related stimuli, such that stimuli predictive of a loss were again experienced as aversive, consequently lead to reduced engagement with the task.  Dopamine agonists also appear to lead to a reduction in the potency with which negative outcomes mediate decision making in human subjects. Patients with PD tend to display a greater propensity to learn from negative outcomes whilst off dopaminergic medication, but when on medication the same patients demonstrate impairments at learning from negative outcomes (Frank et al. 2007; Frank et al. 2004). Therefore, dopamine replacement therapies may render PD patients vulnerable to the development of ICD’s as aversive outcomes such as monetary loss become ineffective at tempering risky behaviour. β-adrenoreceptor blockers mediating CREB activity could theoretically have the potential to augment the salience of negative outcomes, making these patients less susceptible to developing maladaptive behaviours.  In summary, these results provide further evidence that ICD’s appear to develop directly as a result of dopamine agonist therapies, as oppose to ‘unmasking’ latent dysfunction. Moreover, we provide evidence indicating that β-adrenoreceptor blockade may aid in ameliorating ICD’s.  This finding is consistent with our prediction from a previous report that agents capable of inhibiting pCREB in the NAC could be useful in remediating ropiniroles’ deleterious effects. Further studies aiming to confirm this putative mechanism, as well as delineate a more effective dosing regime of propranolol may be tremendously beneficial in developing novel treatment strategies for iatrogenic gambling.    171   Figure 8.1 Baseline rSMT performance (a,b) Animals showed optimal responding on win trials (1,1,1), choosing to collect the available reward nearly 100% of the time.  Similarly, animals showed a marked preference for the optimal response, now the roll lever, when no lights were illuminated (0,0,0), only responding on the collect lever approximately 20% of the time.  Erroneous collect responses increased to ~45% when 1 light was illuminated.  However, when two-lights were illuminated in the array, animals responded erroneously on the collect lever at a far greater then chance level (78.67% ± 3.39 (SEM)), indicating that rats, like humans, treat such stimuli as more indicative of a win than a loss, and are hence susceptible to the near-miss effect.  There were no differences between the saline and ropinirole groups. All data shown are the mean across five sessions ± SEM.  172   Figure 8.2 Effect of dopamine depletions in the dorsal striatum via bilateral infusion of 6-OHDA (a, b) Dopamine depletions of the dorsal striatum, a well-established model of Parkinson’s disease, produced no alterations on animals’ choice of the collect lever. All data shown are the mean across the last 5 sessions prior to osmotic mini-pump surgery ± SEM  173    174  Figure 8.3 Effects of dietary lithium and chronic ropinirole on reward expectancy (a, b) during the first and second (c, d) session bins, animals that had osmotic mini-pumps delivering the D2-like agonist ropinirole showed an increase in erroneous collection responses in comparison to animals receiving saline. As administration of ropinirole progressed the deleterious effects of ropinirole abated, such that there were no differences between saline and ropinirole treated animals in the third session bin (e, f). (a-f) The administration of 0.3% Li+ in the animals’ diet did not produce any alterations in choice of the collect lever in either ropinirole or saline treated animals.  All data shown are the mean across 4 sessions ± SEM    Figure 8.4 Effects of repeated injections of the β-adrenoreceptor blocker propranolol and chronic ropinirole on reward expectancy. (a, b) In a similar fashion to dietary Li+, repeated injection with propranolol did not alter animals’ choice of the collect lever. Ropinirole administration, consistent with earlier results also did not alter choice of the collect lever.  All data shown are the mean across 4 sessions ± SEM 175    Figure 8.5 Effects of chronic ropinirole, dietary lithium and propranolol on the number of trials completed. Ropinirole administration produced a robust and sustained increase in the number of trials completed. Dietary Li+ did nothing to attenuate this compulsive-like task engagement. In contrast injection of propranolol did significantly attenuate the number of trials completed, only in the ropinirole treated animals. These effects were most pronounced during the first 3 days of propranolol administration.   176    Figure 8.6 Effects of 6-OHDA and chronic ropinirole on the number of adjusting steps made during the forelimb adjusting step test procedure. Infusion of 6-OHDA into the dorsal striatum reduced the number of adjusting steps rats took during the step-test procedure, indicating that animals displayed Parkinsonian-like motor deficits. These deficits whilst still present in the animals implanted with mini-pumps delivering saline, but were ameliorated in animals treated with ropinirole.    Table 8.1 Latency to respond on the collect lever by trial type at baseline and following pharmacological challenges. Data presented in time (s) shown as mean ± SEM Condition 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Baseline 0.76 ± 0.06 1.10 ± 0.08 0.96 ± 0.08 0.94 ± 0.08 0.70 ± 0.06 0.67 ± 0.06 0.75 ± 0.07 0.58 ± 0.15 Sham surgery 0.78 ±   0.10 0.96 ±   0.1 0.80 ±   0.08 0.85 ±   0.08 0.67 ± 0.09 0.52 ±   0.07 0.53 ±   0.6 0.30 ± 0.08  6-OHDA  0.8 ±   0.08 1.24 ±   0.19 0.96 ±   0.1 0.92 ±   0.11 0.75 ±   0.1 0.77 ± 0.15 0.76 ±   0.13 0.31 ± 0.06 Con diet – Sal Bin 1 0.93 ± 0.21 1.45 ± 0.26 0.9 ± 0.22 1.12 ± 0.21 0.77 ± 0.16 0.74 ± 0.15 0.72 ± 0.23 0.38 ± 0.1 177  Condition 1,1,1 1,1,0 1,0,1 0,1,1 1,0,0 0,1,0 0,0,1 0,0,0 Con diet – Rop Bin 1 0.82 ± 0.08 1.14 ± 0.19 0.85 ± 0.1 0.90 ± 0.08 0.94 ± 0.2 0.75 ± 0.18 0.71 ± 0.1 0.36 ± 0.1 Li+ diet – Sal Bin 1 1.21 ± 0.27 1.28 ± 0.19 0.95 ± 0.22 1.07 ± 0.19 1.08 ± 0.23 0.81 ± 0.14 0.49 ± 0.07 0.4 ± 0.12 Li+ diet – Rop Bin 1 0.78 ± 0.13 1.27 ± 0.15 1.08 ± 0.14 1.0 ± 0.13 0.85 ± 0.11 0.88 ± 0.12 1.1 ± 0.17 0.47 ± 0.09 Con diet – Sal Bin 2 0.83 ± 0.22 1.16 ± 0.19 0.95 ± 0.19 1.10 ± 0.25 0.95 ± 0.19 0.47 ± 0.10 0.67 ± 0.19 0.68 ± 0.31 Con diet – Rop Bin 2 0.74 ± 0.07 0.86 ± 0.08 0.78 ± 0.08 0.72 ± 0.07 0.78 ± 0.08 0.43 ± 0.09 0.45 ± 0.08 0.18 ± 0.07 Li+ diet – Sal Bin 2 1.01 ± 0.28 1.05 ± 0.21 0.94 ± 0.17 0.86 ± 0.10 0.94 ± 0.17 0.57 ± 0.15 0.51 ± 0.14 0.23 ± 0.1 Li+ diet – Rop Bin 2 0.66 ± 0.11 0.82 ± 0.1 0.73 ± 0.11 0.75 ± 0.09 0.73 ± 0.11 0.43 ± 0.08 0.54 ± 0.08 0.18 ± 0.03 Con diet – Sal Bin 3 1.0 ± 0.3 1.34 ± 0.28 1.09 ± 0.37 1.13 ± 0.29 0.82 ± 0.29 0.64 ± 0.18 0.73 ± 0.14 0.57 ± 0.18 Con diet – Rop Bin 3 0.83 ± 0.1 0.76 ± 0.11 0.71 ± 0.08 0.63 ± 0.07 0.47 ± 0.08 0.36 ± 0.1 0.32 ± 0.08 0.18 ± 0.07 Li+ diet – Sal Bin 3 1.3 ± 0.41 1.19 ± 0.25  0.98 ± 0.2 1.04 ± 0.26 0.63 ± 0.16 0.51 ± 0.16 0.45 ± 0.18 0.38 ± 0.29 Li+ diet – Rop Bin 3 0.65 ± 0.12 0.62 ± 0.1 0.61 ± 0.11 0.65 ± 0.12 0.30 ± 0.08 0.31 ± 0.07 0.36 ± 0.07 0.06 ± 0.01 Sal – Sal 1.04 ± 0.31 1.11 ± 0.24 0.89 ± 0.16 0.91 ± 0.2 0.6 ± 0.13 0.63 ± 0.18 0.53 ± 0.14 0.38 ± 0.13 Sal – Propran 0.92 ± 0.23 1.33 ± 0.29  1.49 ± 0.51 1.49 ± 0.44 0.88 ± 0.25 0.62 ± 0.13 0.93 ± 0.34 0.44 ± 0.13 Rop – Sal 0.74 ± 0.16 0.69 ± 0.12 0.7 ± 0.12 0.72 ± 0.13 0.35 ± 0.07 0.39 ± 0.08 0.35 ± 0.06 0.15 ± 0.03 Rop - Propran 1.1 ± 0.21 0.79 ± 0.11 0.94 ± 0.15 0.93 ± 0.14 0.41 ± 0.08 0.43 ± 0.09 0.55 ± 0.1 0.44 ± 0.13           Table 8.2 Latency to respond at subsequent hole based on the statues of the previous hole for baseline and pharmacological challenges. Data presented in time (s) shown as mean ± SEM Condition H3 by H2 on H3 by H2 off H4 by H3 on H4 by H3 off Baseline 1.76 ± 0.16 3.40 ± 0.34 1.51 ± 0.29 2.46 ± 0.23 Sham Surgery 1.71 ± 0.58 3.11 ± 0.61 0.97 ± 0.08 2.08 ± 0.35 6-OHDA 1.51 ± 0.73 2.96 ± 0.38 1.41 ± 0.33 2.01 ± 0.25 178  Condition H3 by H2 on H3 by H2 off H4 by H3 on H4 by H3 off Con diet – Sal Bin 1 0.72 ± 0.08 2.98 ± 0.96 0.77 ± 0.16 2.75 ± 1.39 Con diet – Rop Bin 1 0.79 ± 0.10 1.35 ± 0.19 0.85 ± 0.09 1.27 ± 0.18 Li+ diet – Sal Bin 1 10.96 ± 8.23 6.55 ± 2.05 1.24 ± 0.19 2.29 ± 0.45 Li+ diet – Rop Bin 1 1.48 ± 0.48 1.9 ± 0.22 0.9 ± 0.08 1.46 ± 0.18 Con diet – Sal Bin 2 0.60 ± 0.07 3.17 ± 1.66 0.83 ± 0.15 1.67 ± 0.41 Con diet – Rop Bin 2 0.81 ± 0.14 1.52 ± 0.45 0.83 ± 0.09 1.19 ± 0.13 Li+ diet – Sal Bin 2 2.22 ± 0.89 2.40 ± 0.95 1.06 ± 0.22 1.73 ± 0.26 Li+ diet – Rop Bin 2 0.74 ± 0.11 1.01 ± 0.1 0.73 ± 0.07 1.73 ± 0.26 Con diet – Sal Bin 3 0.63 ± 0.06 2.67 ± 0.82 0.64 ± 0.07 1.91 ± 0.36 Con diet – Rop Bin 3 0.74 ± 0.11 1.38 ± 0.29 1.15 ± 0.26 1.90 ± 0.45 Li+ diet – Sal Bin 3 1.16 ± 0.25 2.33 ± 0.72 0.93 ± 0.15 3.57 ± 2.21 Li+ diet – Rop Bin 3 0.83 ± 0.19 1.12 ± 0.16 1.14 ± 0.35 1.54 ± 0.35 Sal – Sal 1.29 ± 0.27 2.02 ± 0.48 1.31 ± 0.44 2.69 ± 1.55 Sal – Propran 0.83 ± 0.15 2.7 ± 1.05 0.68 ± 0.09 2.24 ± 0.57 Rop – Sal 0.72 ± 0.1 0.94 ± 0.12 0.68 ± 0.08 1.03 ± 0.09 Rop - Propran 0.97 ± 0.26 1.21 ± 0.21 0.79 ± 0.09 1.56 ± 0.27             179  Chapter 9: General Discussion   9.1 Summary of experimental findings Here we used a rodent analogue of a simple slot machine in order to examine the cortico-limbic and neuromodulatory influences on gambling-like decision making.  Using this task, we show that animals share key behavioural features with human gamblers.  Comparing our findings to others in the literature, we suggest that gambling is a heterogeneous disorder with multiple underlying aetiologies and endophenotypic animal models may aid in elucidating differing vulnerabilities. We present novel evidence that the dopamine D4 receptor mediates attributions of salience to reward-related stimuli and that β-adrenoreceptor blockade could attenuate compulsive gambling. These results have implications for future treatment strategies for both iatrogenic and idiopathic GD.  Experiment 1 replicated previous findings that showed rats, like humans, demonstrate increased reward expectancy when multiple win-related stimuli are presented concurrently. We argue that this phenomenon is translationally analogous to the near-miss effect reliably observed in human gamblers. We demonstrated a critical role for the dopamine D2-like family of receptors in augmenting reward expectancy on this task, in that administration of the D2-like agonist quinpirole increased erroneous choice of the collect lever. We further delineated the pharmacological basis of the rSMT by demonstrating that highly selective D4 agents altered behaviour. A D4 agonist impaired performance on the task in a similar, albeit less robust, manner to quinpirole, whereas a D4 antagonist actually improved animals’ ability to differentiate winning from non-winning trials. Lastly, we showed that prior administration of a D4 antagonist partially 180  remediated the deleterious effects of quinpirole. Ultimately these data suggest a novel role for D4 receptors in mediating animals’ attributions of salience to reward associated stimuli and intimate that D4 receptors may represent a potential avenue for the development of novel pharmacotherapies for GD.  Experiment 2 aimed to clarify the results from experiment 1, by elucidating what role D4 receptors play in mediating attributions of salience. We therefore assessed the role of D4 receptors in controlling Pavlovian and instrumental forms of incentive motivation. Using Pavlovian conditioned approach (“sign-tracking”) and a conditioned reinforcement paradigm we demonstrated that D4 receptors were not involved in controlling incentive motivation on either of these simple behavioural tasks. Although a null result, experiment 2 provided the first evidence that D4 receptors were not involved in the expression of conditioned approach or conditioned reinforcement, and may question the translational validity of using simple tasks to model complex disorders such as gambling addiction.  Experiment 3 targeted the ACC, an area that has been shown repeatedly to be involved in risk-reward decision-making, and which also contains a relatively high density of D4 receptors (Holroyd and McClure 2015; Lewis 1992; Rushworth et al. 2007; Westlund et al. 1990). Temporary inactivation of the ACC via local infusion of GABA receptor agonists, increased erroneous responding on both 1- and 0-light trials; indicating that this area is critically involved in parsing the appropriate response when competing stimulus-outcome associations are activated. We also administered the D4 agonist PD168077 into the ACC, which caused impairments selectively on archetypal near-miss trials, wherein the first two, but not the final light, are concordant with a winning outcome ([1,1,0]).  It would therefore appear that D4 receptor-181  mediated transmission in this region is somehow specifically involved in emphasising the temporal or spatial proximity of reward-predictive stimuli.  Experiment 4 examined another prefrontal structure, the insular cortex. Similar to the ACC, the insula has been heavily implicated in both substance and behavioural addictions (Naqvi and Bechara 2009) and there is a relatively high density of D4 receptors within parts of the insula  (Rivera et al. 2008). In experiment 4 we investigated the effects of temporary inactivation of the granular and agranular insula regions in different rats. We showed that temporary inactivation of the AI, but not GI impaired rSMT performance when 2- or 1-lights were illuminated. In contrast, infusion of a D4 receptor agonist directly into the AI had no effect on task performance, but when infused into the GI resulted in a decrease in erroneous collect lever responses on trials when the last light set to off. In other words, D4 agonism in the GI improved animals’ task performance when the final aperture in the sequence predicted a loss. The results from experiment 4 compliment experiment 3, in that the ACC and the insula are often activated concomitantly suggesting overlapping or complimentary roles (Craig 2009). Our data indicate that, while both areas likely contribute to gambling-related decision making, there are key differences. In experiment 3 we suggest that the ACC acts to constrain maladaptive Pavlovian approach tendencies, whereas the AI is integrating and guiding response output when conflicting information is presented. The finding that a D4 agonist infused into the GI improved task performance indicates that D4 receptors may also mediate salience attributions to unpleasant or aversive stimuli dependant in this region.  In the GI, we suggest that activation of these receptors selectively augmented the salience of a potential time-out punishment.  Experiment 5 showed that the rSMT could be used to model problematic engagement with gambling. Chronic administration of the D2-like agonist ropinirole produced augmented 182  behavioural responding on the rSMT, as evidenced by a robust increase in the number of trials completed. Ropinirole is commonly used as an adjunctive therapy for PD, and as such these results may inform the underlying neurobiology of the iatrogenic gambling seen in a sub-set of PD patients following treatment with dopamine agonists (Dodd et al. 2005). This experiment not only revealed chronic D2-agonism led to compulsive-like gambling behaviour, but also pointed to a putative mechanism. Using Western blots, we demonstrated chronic ropinirole led to an increase in pCREB in the NAC and GSK3β activity in the dorsal striatum. Results from experiment 5 suggest that attenuating either of these two intra-cellular signaling cascades may hold therapeutic potential for iatrogenic gambling observed following dopamine agonist therapy and perhaps other forms of compulsive gambling.   Experiment 6 revealed multiple findings. First, we showed that inducing Parkinsonian-like symptoms in animals via dopamine depletions of the dorsal striatum did not alter performance on the rSMT. Secondly, we replicated the effects of experiment 5, confirming that chronic ropinirole produced a robust increase in the number of trials completed on the rSMT. Finally, experiment 6 showed that dietary administration of Li+ (which potently blocks the activity of GSK3β) had no effect on task performance, but repeated injections of the β-adrenoreceptor blocker propranolol (which inhibits the phosphorylation of CREB in the NAc) did attenuate the compulsive-like behaviour induced by chronic ropinirole. Ultimately, these data suggest that β-adrenoreceptor antagonists could be a novel treatment avenue for controlling ICD’s following dopamine agonist therapies.   183  9.2 Theoretical implications and considerations for future studies Taken together these results suggest that dopamine may be acting in a number of different ways to guide gambling-related decision making. Our results suggest a critical role for dopamine D4 receptors in stabilising representations of emotionally valent stimuli. Importantly these representations appear to differ based on the region. For instance, while D4 receptors in the ACC augment the salience of reward associated stimuli presented temporally proximally to a decision-point, the same receptor in the GI appears to increase the salience of a potential punishment. We have also shown that chronic D2-like agonism produces a compulsive-like pattern of behaviour, ostensibly mediated via increased activation CREB in the NAC.  Ultimately, these data indicate that the same neurotransmitter is having very different neuromodulatory effects depending on the level of activation, and the region different receptor subtypes are activated within. These results suggest a number of hypothesis to test in the future.  9.2.1 Investigate any role for the cholinergic and noradrenergic systems on the rSMT. Data from experiments 3 and 4 point to a key role for prefrontal regions in controlling performance on the rSMT. We have assumed these effects are mostly dependent on dopaminergic functioning, as a result of experiments 1, 5 & 6 as well as dopamine’s established role in mediating risk-reward decision making. Yet both cholinergic and noradrenergic activity within the PFC increase during attentional tasks, such as the 5-CSRTT (Dalley et al. 2001) and systemic modulation of either system alters decision making on operant tasks (Hosking et al. 2014b; Sun et al. 2012). Experiment 6 revealed that the β-adrenoreceptor antagonist propranolol attenuated the ropinirole induced increase in trials, but produced no effect when administered in isolation, which could indicate that adrenoreceptor modulation be a fruitful area to investigate in terms of pathological, as opposed to “healthy” choice. However, the dose used in experiment 6 184  was chosen specifically to be behaviourally silent (Adams 2016). Thus we cannot preclude a role for adrenoreceptor modulation on rSMT performance in general. Investigating the role of both cholinergic and noradrenergic systems could be undertaken either through systemic pharmacology or targeting specific structures within the PFC. Alternatively, the role of the cholinergic system could be assessed via designer receptors activated exclusively by designer drugs (DREADDS). DREADDS are modified versions of G-protein coupled human muscarinic receptors that do not respond to any endogenous ligands and show no constitutive activity, but can be activated by the highly selective exogenous ligand clozapine-n-oxide (CNO) (Pei et al. 2010). DREADDS are typically engineered to be ‘Cre-dependant’ i.e. the gene controlling their expression is only transduced in the presence of Cre recombinase.  Our lab has recently established colonies of choline acetyltransferase (ChAT): Cre and tyrosine hydroxylase (TH):Cre rats. Using these preparations, animals’ prefrontal vs subcortical cholinergic and noradrenergic function could be systemically up- or down-regulated in order to elucidate any effects on the rSMT.   9.2.2 Optogenetic examination of the putative circuitry underlying rSMT performance. The techniques we have used here allowed us to examine the role of specific neural regions on rSMT performance. However, inactivating select regions whilst useful, is somewhat crude and only allows for partial exploration of one loci that is almost certainly working in concert with numerous others. Future studies aiming to explore the neural circuitry underlying rSMT performance could use optogenetics. Optogenetics utilises microbial opsins virally transfected into target cells that can subsequently be manipulated with high temporal and spatial specificity within awake behaving animals using optical stimulation (Tye and Deisseroth 2012). We have previously suggested that prefrontal and amygdalar regions compete for control over striatal 185  directed behaviour during rSMT performance (Cocker 2015). Optogenetic stimulation would allow for physiologically relevant up or down regulation of these pathways.  9.2.3 Explore the relationship between dysfunctional reward expectancy and propensity toward drug self-administration. Despite similarities in the phenomenology of drug and behavioural addictions, it is still unclear whether they share a common neurobiological basis. Animal models that can fraction out differing precipitating vulnerabilities may be valuable in elucidating any commonalities between differing forms of addiction. Broadly, in addition to increases in impulsivity, individuals with addictive disorders tend to display perturbations in cost-benefit decision making (see Cocker and Winstanley 2015 for discussion). Our lab has recently shown that rats which make risky decisions at baseline on the rGT are uniquely and adversely affected by cocaine administration, in that their decisions becomes even riskier, and this correlates with enhanced cue-induced drug-seeking (Ferland and Winstanley 2016).  However, in contrast to the marked effects on the rSMT, chronic ropinirole- which we know can trigger GD in human subjects-  does not alter rGT performance.  Such data raise the intriguing possibility that the decision-making phenotype/cognitive processes associated with drug and gambling addiction may be behaviourally dissociable.  Investigating the impact of cocaine self-administration on performance of tasks such as the rSMT could be a valuable way of further exploring this hypothesis.   9.2.4 Sex differences in gambling. Slot machines, as mentioned earlier are a particularly compulsive form of gambling, with a disproportionate number of slot machine players presenting for treatment (Breen and Zimmerman 2002; Choliz 2010; Dowling et al. 2005).  Yet slot machine players represent a fairly thin demographic, with players most often reported to be middle-aged women (Blaszczynski and Nower 2002; Petry 2003).  These gamblers not only 186  display a later commencement of recreational gaming but also a faster onset of problem gambling, so called ‘telescoping’ (Morgan et al. 2010; Piazza et al. 1989a).  What underlies the late-onset and rapidity of this type of gambling is not immediately clear and to our knowledge there is no commensurable increase in gambling in men of a similar age.  Interestingly, levels of the dopamine transporter and dopamine receptor density decline with age (Kaasinen et al. 2000; van Dyck et al. 2002; Wong et al. 1984), although whether this yields an increased vulnerability towards compulsive gambling is not known. Similarly unclear is the effect, if any, of the decrease in sex hormones as a result of menopause.  Menopause has a relatively similar age of onset as this particularly compulsive form of gambling and is associated with increased vulnerability towards neuropsychiatric disorders such as schizophrenia and depression (Di Paolo 1994; Epperson et al. 1999).  Similar to GD, both these latter disorders have strong links with monoamine dysfunction and patients have been reported to exhibit hypofrontality (Galynker et al. 1998; Howes and Kapur 2009; Nestler et al. 2002).  The effects of sex and hormones on gambling-related decision making could be explored relatively easily on the rSMT. Systemic administration of dopaminergic agents in intact and ovariectomized females would be a crucial first step in identifying any potential role for sex hormones in mediating the influence of dopamine on gambling behaviour.    9.3 Limitations and critical considerations  Firstly, in comparison to other behavioural tasks the rSMT takes a very long time to train, often taking 100+ sessions in order for animals to reach stable asymptotic performance. This length of time presents possible confounds in regards to behavioural flexibility. Shortening the number of training sessions, or even reducing the number of lights from three to two might 187  augment the effects of pharmacological or surgical interventions. Additionally, a higher throughput would make the rSMT more efficient as a screening paradigm for pharmacotherapies. However, it is important to note that performance of the rSMT never becomes habitual, at least in so far as behaviour quickly abates when reward is no longer available, the typical method for assessing goal-directed vs. habitual responding (Balleine and O'Doherty 2010; Winstanley et al. 2011). Moreover, a shortened training time may induce ceiling effects in regards to animals’ erroneous lever choice, making the detection of drug effects more difficult. Reducing the number of apertures used would almost certainly shorten the training time, but would reduce the face validity of the task somewhat and near-misses would make up fully 50% of all trials, likely reducing the potency with which these trial types generate reward expectancy (Cote et al. 2003).  One intractable issue with the rSMT and animal studies in general is how losses are represented. Human gambling is predicated on the notion of wagering something of value, usually money, for the chance of winning a greater amount, or losing the wagered stake.  As animals are rewarded with sugar pellets in these gambling models, the only potential loss would be a caloric deficit, whereby animals expend energy unsuccessfully attempting to gain reward.  Animals performing these tasks are kept in a state of food deprivation, thus the use of time-out penalties reduces the total amount of time animals have to supplement their basic diet.  In other words, animals do not experience a tangible loss, but rather an opportunity cost.  Whether this type of loss adequately approximates the experience of a human gambler is not clear.  However, many laboratory based human studies are conducted in such a way that subjects are remunerated with a flat fee, thereby eliminating any potential for loss.  Alternatively, subjects may play a game for the prospect of increasing or decreasing the compensation they receive for participating.  Using either of these contingencies, participants cannot leave the session at a 188  monetary disadvantage.  Thus the use of food reward in animals appears to be at least as efficacious as rewards offered in many laboratory based studies using human volunteers, even if both are somewhat distinct from real-life gambling.  Similarly, there are critical differences in regards to reward between human and animal tasks.  Human gamblers tend to be rewarded with secondary reinforcers such as money, whereas primary rewards such as food are the mainstay of animal paradigms.  This creates the de facto assumption that these rewards are equivalent.  Whether using primary vs. secondary reinforcers evokes different motivational processes is difficult to know unequivocally.  Imaging studies have shown humans do show some reward-specific activation, but generally both primary and secondary rewards activate common brain areas including the VMPFC, amygdala and striatum (Sescousse et al. 2013).  Critically, studies using healthy human controls have shown that people will display similar levels of motivation for a liquid reward as they do for money (Hayden and Platt 2009)  A general consideration that is important to state is that although animal models can offer insight into the neurobiological basis of human disease, valid and efficacious animal models rely on two-way translational research, meaning limitations within human research may indirectly affect attempts to model human conditions in animals. One of the reoccurring notions throughout this thesis is the statement that gambling is a diverse activity and gamblers a heterogeneous population. Problem gambling is observed in both sexes, across all cultures and ethnicities, and encompasses a wide age range (Grant and Kim 2001; Welte et al. 2001), yet gamblers are often treated as a homogenous group.  One of the most authoritative reviews to date has delineated a conceptual framework characterizing gamblers by personality characteristics and precipitating 189  vulnerabilities (Blaszczynski and Nower 2002).  However, these categories are often overlooked and even within these sub-types there may be substantial differences.  A compelling inference is that divergent results in human research may be due to the presence or specificity of gambling related cues between different tasks, (Balodis et al. 2012a but see; Leyton and Vezina 2012). Whether salient stimuli tailored to the individuals’ game of choice would be useful is unclear.  Nevertheless, grouping gamblers by their primary game of choice may be a simple method of categorization that may help improve consistency, as evidence suggests that different forms of gambling appeal to different types of individuals (Dickerson 1993; Petry 2003).  Whether such a prima-facie grouping would be preferable to sampling based on demographic factors is debatable.  Furthermore, research using clinically-diagnosed patients with GD is often compromised due to difficulty recruiting sufficient subjects as the disorder has a relatively low base rate (1-2%), therefore parsing subjects by preferred mode of gambling is practically challenging.  Yet, what does seem clear is that the difference in demographics between users of different gambling games is striking (Petry 2003) and separating out gamblers by games of choice may partially control for variations in demographic factors by proxy. A more comprehensive understanding of the differing forms of human gambling could greatly increase the construct validity of animal models aiming to fractioning out various aspects of dysfunctional decision making.       190  9.4 Concluding remarks Here, we have shown that animals, like humans, show increased reward expectancy when presented with a losing outcome that superficially resembles a winning one. We have shown that this behaviour, potentially analogous to the near-miss effect can be perturbed by pharmacological and surgical interventions targeting the dopamine system. Moreover, we have demonstrated that the model can be modified to induce problem-like gambling. This task provides a valuable model of gambling-like behaviour and emphasised several avenues from which to explore novel pharmacotherapies for the treatment of gambling disorder.                               191  Bibliography   Abdolahi A, Williams GC, Benesch CG, Wang HZ, Spitzer EM, Scott BE, Block RC, van Wijngaarden E (2015) Damage to the insula leads to decreased nicotine withdrawal during abstinence. Addiction 110: 1994-2003. Abela AR, Chudasama Y (2013) Dissociable contributions of the ventral hippocampus and orbitofrontal cortex to decision-making with a delayed or uncertain outcome. The European journal of neuroscience 37: 640-7. Abela AR, Dougherty SD, Fagen ED, Hill CJ, Chudasama Y (2013) Inhibitory control deficits in rats with ventral hippocampal lesions. Cereb Cortex 23: 1396-409. Adams WK, Barrus, M. M., Zeeb, F. D., Cocker, P. J., Benoit, J & Winstanley, C. A (2016) Dissociable effects of systemic and orbitofrontal administration of adrenoceptor antagonists on yohimbine-induced motor impulsivity. Psychopharmacology  (submitted). Ainslie G (1975) Specious reward: a behavioral theory of impulsiveness and impulse control. Psychological bulletin 82: 463-96. Alessi SM, Petry NM (2003) Pathological gambling severity is associated with impulsivity in a delay discounting procedure. Behav Processes 64: 345-354. Allen GV, Saper CB, Hurley KM, Cechetto DF (1991) Organization of visceral and limbic connections in the insular cortex of the rat. The Journal of comparative neurology 311: 1-16. American Psychiatric Association., American Psychiatric Association. DSM-5 Task Force. (2013) Diagnostic and statistical manual of mental disorders : DSM-5, 5th edn. American Psychiatric Association, Washington, D.C. Anderson RI, Spear LP (2011) Autoshaping in adolescence enhances sign-tracking behavior in adulthood: impact on ethanol consumption. Pharmacol Biochem Behav 98: 250-60. Ariano MA, Wang J, Noblett KL, Larson ER, Sibley DR (1997) Cellular distribution of the rat D4 dopamine receptor protein in the CNS using anti-receptor antisera. Brain Res 752: 26-34. Asghari V, Sanyal S, Buchwaldt S, Paterson A, Jovanovic V, Van Tol HH (1995) Modulation of intracellular cyclic AMP levels by different human dopamine D4 receptor variants. J Neurochem 65: 1157-65. Asghari V, Schoots O, van Kats S, Ohara K, Jovanovic V, Guan HC, Bunzow JR, Petronis A, Van Tol HH (1994) Dopamine D4 receptor repeat: analysis of different native and mutant forms of the human and rat genes. Mol Pharmacol 46: 364-73. Baarendse PJ, Vanderschuren LJ (2012) Dissociable effects of monoamine reuptake inhibitors on distinct forms of impulsive behavior in rats. Psychopharmacology (Berl) 219: 313-26. Baarendse PJ, Winstanley CA, Vanderschuren LJ (2013) Simultaneous blockade of dopamine and noradrenaline reuptake promotes disadvantageous decision making in a rat gambling task. Psychopharmacology (Berl) 225: 719-31. Balleine BW, Dickinson A (1998) Goal-directed instrumental action: contingency and incentive learning and their cortical substrates. Neuropharmacology 37: 407-19. Balleine BW, O'Doherty JP (2010) Human and rodent homologies in action control: corticostriatal determinants of goal-directed and habitual action. Neuropsychopharmacology 35: 48-69. 192  Balodis IM, Kober H, Worhunsky PD, Stevens MC, Pearlson GD, Potenza MN (2012a) Attending to striatal ups and downs in addictions. Biological psychiatry 72: e25-6. Balodis IM, Kober H, Worhunsky PD, Stevens MC, Pearlson GD, Potenza MN (2012b) Diminished frontostriatal activity during processing of monetary rewards and losses in pathological gambling. Biological psychiatry 71: 749-57. Barbas H, Pandya DN (1989) Architecture and Intrinsic Connections of the Prefrontal Cortex in the Rhesus-Monkey. Journal of Comparative Neurology 286: 353-375. Barrot M, Olivier JD, Perrotti LI, DiLeone RJ, Berton O, Eisch AJ, Impey S, Storm DR, Neve RL, Yin JC, Zachariou V, Nestler EJ (2002) CREB activity in the nucleus accumbens shell controls gating of behavioral responses to emotional stimuli. Proc Natl Acad Sci U S A 99: 11435-40. Barrus MM, Hosking JG, Zeeb FD, Tremblay M, Winstanley CA (2015) Disadvantageous decision-making on a rodent gambling task is associated with increased motor impulsivity in a population of male rats. J Psychiatry Neurosci 40: 108-17. Barrus MM, Winstanley CA (2016) Dopamine D3 Receptors Modulate the Ability of Win-Paired Cues to Increase Risky Choice in a Rat Gambling Task. J Neurosci 36: 785-94. Baunez C, Christakou A, Chudasama Y, Forni C, Robbins TW (2007) Bilateral high-frequency stimulation of the subthalamic nucleus on attentional performance: transient deleterious effects and enhanced motivation in both intact and parkinsonian rats. Eur J Neurosci 25: 1187-94. Beaulieu JM, Gainetdinov RR (2011) The physiology, signaling, and pharmacology of dopamine receptors. Pharmacol Rev 63: 182-217. Beaulieu JM, Gainetdinov RR, Caron MG (2007) The Akt-GSK-3 signaling cascade in the actions of doparnine. Trends Pharmacol Sci 28: 166-172. Beaulieu JM, Gainetdinov RR, Caron MG (2009) Akt/GSK3 signaling in the action of psychotropic drugs. Annu Rev Pharmacol Toxicol 49: 327-47. Beaulieu JM, Sotnikova TD, Marion S, Lefkowitz RJ, Gainetdinov RR, Caron MG (2005) An Akt/beta-arrestin 2/PP2A signaling complex mediates dopaminergic neurotransmission and behavior. Cell 122: 261-73. Bechara A (2003) Risky business: emotion, decision-making, and addiction. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 19: 23-51. Bechara A (2005) Decision making, impulse control and loss of willpower to resist drugs: a neurocognitive perspective. Nat Neurosci 8: 1458-63. Bechara A, Damasio AR, Damasio H, Anderson SW (1994) Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 50: 7-15. Bechara A, Damasio H, Damasio AR, Lee GP (1999) Different contributions of the human amygdala and ventromedial prefrontal cortex to decision-making. The Journal of neuroscience : the official journal of the Society for Neuroscience 19: 5473-81. Bechara A, Dolan S, Denburg N, Hindes A, Anderson SW, Nathan PE (2001) Decision-making deficits, linked to a dysfunctional ventromedial prefrontal cortex, revealed in alcohol and stimulant abusers. Neuropsychologia 39: 376-89. Belin D, Mar AC, Dalley JW, Robbins TW, Everitt BJ (2008) High impulsivity predicts the switch to compulsive cocaine-taking. Science 320: 1352-5. 193  Beninger RJ, Banasikowski TJ (2008) Dopaminergic mechanism of reward-related incentive learning: focus on the dopamine D(3) receptor. Neurotox Res 14: 57-70. Beninger RJ, Phillips AG (1981) The effects of pimozide during pairing on the transfer of classical conditioning to an operant discrimination. Pharmacol Biochem Behav 14: 101-5. Beninger RJ, Ranaldi R (1992) The effects of amphetamine, apomorphine, SKF 38393, quinpirole and bromocriptine on responding for conditioned reward in rats. Behav Pharmacol 3: 155-163. Berendse HW, Galis-de Graaf Y, Groenewegen HJ (1992) Topographical organization and relationship with ventral striatal compartments of prefrontal corticostriatal projections in the rat. J Comp Neurol 316: 314-47. Berger B, Thierry AM, Moyne MA (1976) Dopaminergic innervation of the rat prefrontal cortex: a fluoresence histochemical study. Brain Res 106: 133-145. Bernoulli D (1954) Exposition of a New Theory on the Measurement of Risk. Econometrica 22: 23-36. Berridge KC, Robinson TE (1998) What is the role of dopamine in reward: hedonic impact, reward learning, or incentive salience? Brain Res Brain Res Rev 28: 309-69. Bertz JW, Jackson EL, Barron DR, Woods JH (2016) Effects of sex and remifentanil dose on rats' acquisition of responding for a remifentanil-conditioned reinforcer. Behav Pharmacol 27: 137-47. Biesdorf C, Wang AL, Topic B, Petri D, Milani H, Huston JP, de Souza Silva MA (2015) Dopamine in the nucleus accumbens core, but not shell, increases during signaled food reward and decreases during delayed extinction. Neurobiol Learn Mem 123: 125-39. Billieux J, Van der Linden M, Khazaal Y, Zullino D, Clark L (2012) Trait gambling cognitions predict near-miss experiences and persistence in laboratory slot machine gambling. Br J Psychol 103: 412-27. Blaszczynski A (1999) Pathological gambling and obsessive-compulsive spectrum disorders. Psychological reports 84: 107-13. Blaszczynski A, Nower L (2002) A pathways model of problem and pathological gambling. Addiction 97: 487-99. Blesa J, Phani S, Jackson-Lewis V, Przedborski S (2012) Classic and new animal models of Parkinson's disease. J Biomed Biotechnol 2012: 845618. Blum K, Sheridan PJ, Wood RC, Braverman ER, Chen TJH, Cull JG, Comings DE (1996) The D2 dopamine receptor gene as a determinant of reward deficiency syndrome. Journal of the Royal society of medicine 89: 396-400. Boileau I, Nakajima S, Payer D (2015) Imaging the D3 dopamine receptor across behavioral and drug addictions: Positron emission tomography studies with [(11)C]-(+)-PHNO. Eur Neuropsychopharmacol 25: 1410-20. Boileau I, Payer D, Chugani B, Lobo D, Behzadi A, Rusjan PM, Houle S, Wilson AA, Warsh J, Kish SJ, Zack M (2013) The D2/3 dopamine receptor in pathological gambling: a positron emission tomography study with [11C]-(+)-propyl-hexahydro-naphtho-oxazin and [11C]raclopride. Addiction 108: 953-63. Boileau I, Payer D, Chugani B, Lobo DS, Houle S, Wilson AA, Warsh J, Kish SJ, Zack M (2014) In vivo evidence for greater amphetamine-induced dopamine release in pathological gambling: a positron emission tomography study with [(11)C]-(+)-PHNO. Mol Psychiatry 19: 1305-13. 194  Bouthenet ML, Souil E, Martres MP, Sokoloff P, Giros B, Schwartz JC (1991) Localization of dopamine D3 receptor mRNA in the rat brain using in situ hybridization histochemistry: comparison with dopamine D2 receptor mRNA. Brain Res 564: 203-19. Boyson SJ, McGonigle P, Molinoff PB (1986) Quantitative autoradiographic localization of the D1 and D2 subtypes of dopamine receptors in rat brain. J Neurosci 6: 3177-88. Brand M, Kalbe E, Labudda K, Fujiwara E, Kessler J, Markowitsch HJ (2005) Decision-making impairments in patients with pathological gambling. Psychiatry research 133: 91-9. Breen RB, Zimmerman M (2002) Rapid onset of pathological gambling in machine gamblers. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 18: 31-43. Brown JW, Braver TS (2005) Learned predictions of error likelihood in the anterior cingulate cortex. Science 307: 1118-21. Burgess S, Geddes J, Hawton K, Townsend E, Jamison K, Goodwin G (2001) Lithium for maintenance treatment of mood disorders. Cochrane Database Syst Rev: CD003013. Bush G, Vogt BA, Holmes J, Dale AM, Greve D, Jenike MA, Rosen BR (2002) Dorsal anterior cingulate cortex: a role in reward-based decision making. Proceedings of the National Academy of Sciences of the United States of America 99: 523-8. Cador M, Robbins TW, Everitt BJ (1989) Involvement of the amygdala in stimulus-reward associations: interaction with the ventral striatum. Neuroscience 30: 77-86. Caine SB, Koob GF (1993) Modulation of cocaine self-administration in the rat through D-3 dopamine receptors. Science 260: 1814-6. Calabresi P, Di Filippo M, Ghiglieri V, Tambasco N, Picconi B (2010) Levodopa-induced dyskinesias in patients with Parkinson's disease: filling the bench-to-bedside gap. Lancet Neurol 9: 1106-17. Campbell-Meiklejohn D, Wakeley J, Herbert V, Cook J, Scollo P, Ray MK, Selvaraj S, Passingham RE, Cowen P, Rogers RD (2011) Serotonin and dopamine play complementary roles in gambling to recover losses. Neuropsychopharmacology 36: 402-10. Cardinal RN, Daw N, Robbins TW, Everitt BJ (2002a) Local analysis of behaviour in the adjusting-delay task for assessing choice of delayed reinforcement. Neural networks : the official journal of the International Neural Network Society 15: 617-34. Cardinal RN, Parkinson JA, Hall J, Everitt BJ (2002b) Emotion and motivation: the role of the amygdala, ventral striatum, and prefrontal cortex. Neurosci Biobehav Rev 26: 321-52. Cardinal RN, Pennicott DR, Sugathapala CL, Robbins TW, Everitt BJ (2001) Impulsive choice induced in rats by lesions of the nucleus accumbens core. Science 292: 2499-501. Carlezon WA, Jr., Duman RS, Nestler EJ (2005) The many faces of CREB. Trends Neurosci 28: 436-45. Carter BL, Tiffany ST (1999) Meta-analysis of cue-reactivity in addiction research. Addiction 94: 327-40. Cavedini P, Riboldi G, Keller R, D'Annucci A, Bellodi L (2002) Frontal lobe dysfunction in pathological gambling patients. Biological psychiatry 51: 334-41. Ceci A, Brambilla A, Duranti P, Grauert M, Grippa N, Borsini F (1999) Effect of antipsychotic drugs and selective dopaminergic antagonists on dopamine-induced facilitatory activity in prelimbic cortical pyramidal neurons. An in vitro study. Neuroscience 93: 107-15. 195  Chamberlain SR, Sahakian BJ (2007) The neuropsychiatry of impulsivity. Current opinion in psychiatry 20: 255-261. Chase HW, Clark L (2010) Gambling severity predicts midbrain response to near-miss outcomes. J Neurosci 30: 6180-7. Chemel BR, Roth BL, Armbruster B, Watts VJ, Nichols DE (2006) WAY-100635 is a potent dopamine D4 receptor agonist. Psychopharmacology (Berl) 188: 244-51. Chen X, Gabitto M, Peng Y, Ryba NJ, Zuker CS (2011) A gustotopic map of taste qualities in the mammalian brain. Science 333: 1262-6. Chernoloz O, El Mansari M, Blier P (2009) Sustained administration of pramipexole modifies the spontaneous firing of dopamine, norepinephrine, and serotonin neurons in the rat brain. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 34: 651-61. Choliz M (2010) Experimental analysis of the game in pathological gamblers: effect of the immediacy of the reward in slot machines. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 26: 249-56. Christakou A, Robbins TW, Everitt BJ (2004) Prefrontal cortical-ventral striatal interactions involved in affective modulation of attentional performance: implications for corticostriatal circuit function. The Journal of neuroscience : the official journal of the Society for Neuroscience 24: 773-80. Chudasama Y, Passetti F, Rhodes SEV, Lopian D, Desai A, Robbins TW (2004) Dissociable aspects of performance on the 5-choice serial reaction time task following lesions of the dorsal anterior cingulate, infralimbic and orbitofrontal cortex in the rat: differential effects on selectivity, impulsivity and compulsivity (vol 146, pg 105, 2003). Behavioural brain research 152: 453-453. Chudasama Y, Robbins TW (2003) Dissociable contributions of the orbitofrontal and infralimbic cortex to pavlovian autoshaping and discrimination reversal learning: further evidence for the functional heterogeneity of the rodent frontal cortex. J Neurosci 23: 8771-80. Clark CA, Dagher A (2014) The role of dopamine in risk taking: a specific look at Parkinson's disease and gambling. Front Behav Neurosci 8: 196. Clark L (2010) Decision-making during gambling: an integration of cognitive and psychobiological approaches. Philosophical transactions of the Royal Society of London Series B, Biological sciences 365: 319-30. Clark L, Bechara A, Damasio H, Aitken MR, Sahakian BJ, Robbins TW (2008) Differential effects of insular and ventromedial prefrontal cortex lesions on risky decision-making. Brain : a journal of neurology 131: 1311-22. Clark L, Crooks B, Clarke R, Aitken MR, Dunn BD (2012) Physiological responses to near-miss outcomes and personal control during simulated gambling. J Gambl Stud 28: 123-37. Clark L, Lawrence AJ, Astley-Jones F, Gray N (2009) Gambling near-misses enhance motivation to gamble and recruit win-related brain circuitry. Neuron 61: 481-90. Clark L, Studer B, Bruss J, Tranel D, Bechara A (2014) Damage to insula abolishes cognitive distortions during simulated gambling. Proc Natl Acad Sci U S A 111: 6098-103. Cocker PJ, Dinelle K, Kornelson R, Sossi V, Winstanley CA (2012) Irrational choice under uncertainty correlates with lower striatal D(2/3) receptor binding in rats. J Neurosci 32: 15450-7. 196  Cocker PJ, Hosking JG, Murch WS, Clark L, Winstanley CA (2016) Activation of dopamine D receptors within the anterior cingulate cortex enhances the erroneous expectation of reward on a rat slot machine task. Neuropharmacology 105: 186-195. Cocker PJ, Le Foll B, Rogers RD, Winstanley CA (2014) A selective role for dopamine D(4) receptors in modulating reward expectancy in a rodent slot machine task. Biol Psychiatry 75: 817-24. Cocker PJ, Tremblay, M., Kaur, S. & Winstanley, C. A. (2016) The dopamine D2/3 agonist ropinirole invigorates performance and induces compulsive-like gambling behaviour on a rodent slot machine task. . Psychopharmacology  (submitted). Cocker PJ, Winstanley CA (2015) Irrational beliefs, biases and gambling: exploring the role of animal models in elucidating vulnerabilities for the development of pathological gambling. Behav Brain Res 279: 259-73. Cocker PJ, Winstanley, C. A. (2015) Towards a Better Understanding of Disordered Gambling: Efficacy of Animal Paradigms in Modelling Aspects of Gambling Behaviour. Current Addiction Reports 2: 240-248. Cole BJ, Robbins TW (1987) Amphetamine impairs the discriminative performance of rats with dorsal noradrenergic bundle lesions on a 5-choice serial reaction time task: new evidence for central dopaminergic-noradrenergic interactions. Psychopharmacology 91: 458-66. Comings DE, Gade-Andavolu R, Gonzalez N, Wu S, Muhleman D, Chen C, Koh P, Farwell K, Blake H, Dietz G, MacMurray JP, Lesieur HR, Rugle LJ, Rosenthal RJ (2001) The additive effect of neurotransmitter genes in pathological gambling. Clin Genet 60: 107-16. Comings DE, Gonzalez N, Wu S, Gade R, Muhleman D, Saucier G, Johnson P, Verde R, Rosenthal RJ, Lesieur HR, Rugle LJ, Miller WB, MacMurray JP (1999) Studies of the 48 bp repeat polymorphism of the DRD4 gene in impulsive, compulsive, addictive behaviors: Tourette syndrome, ADHD, pathological gambling, and substance abuse. Am J Med Genet 88: 358-68. Comings DE, Rosenthal RJ, Lesieur HR, Rugle LJ, Muhleman D, Chiu C, Dietz G, Gade R (1996) A study of the dopamine D2 receptor gene in pathological gambling. Pharmacogenetics 6: 223-34. Corwin J, Peselow E, Feenan K, Rotrosen J, Fieve R (1990) Disorders of decision in affective disease: an effect of beta-adrenergic dysfunction? Biol Psychiatry 27: 813-33. Cosme CV, Gutman AL, LaLumiere RT (2015) The Dorsal Agranular Insular Cortex Regulates the Cued Reinstatement of Cocaine-Seeking, but not Food-Seeking, Behavior in Rats. Neuropsychopharmacology 40: 2425-33. Cote D, Caron A, Aubert J, Desrochers V, Ladouceur R (2003) Near wins prolong gambling on a video lottery terminal. J Gambl Stud 19: 433-8. Craig AD (2002) How do you feel? Interoception: the sense of the physiological condition of the body. Nature reviews Neuroscience 3: 655-66. Craig AD (2009) How do you feel - now? The anterior insula and human awareness. Nature Reviews Neuroscience 10: 59-70. Cussac D, Newman-Tancredi A, Sezgin L, Millan MJ (2000) [H-3]S33084: a novel, selective and potent radioligand at cloned, human dopamine D-3 receptors. N-S Arch Pharmacol 361: 569-572. 197  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: 4908-14. Danna CL, Elmer GI (2010) Disruption of conditioned reward association by typical and atypical antipsychotics. Pharmacol Biochem Behav 96: 40-7. Davies SP, Reddy H, Caivano M, Cohen P (2000) Specificity and mechanism of action of some commonly used protein kinase inhibitors. Biochem J 351: 95-105. Day JJ, Wheeler RA, Roitman MF, Carelli RM (2006) Nucleus accumbens neurons encode Pavlovian approach behaviors: evidence from an autoshaping paradigm. The European journal of neuroscience 23: 1341-51. de Visser L, Homberg JR, Mitsogiannis M, Zeeb FD, Rivalan M, Fitoussi A, Galhardo V, van den Bos R, Winstanley CA, Dellu-Hagedorn F (2011) Rodent versions of the iowa gambling task: opportunities and challenges for the understanding of decision-making. Front Neurosci 5: 109. Defagot MC, Antonelli MC (1997) Autoradiographic localization of the putative D4 dopamine receptor in rat brain. Neurochem Res 22: 401-407. DeJong W (1994) Relapse prevention: an emerging technology for promoting long-term drug abstinence. The International journal of the addictions 29: 681-705. Delfabbro P (2004) The stubborn logic of regular gamblers: obstacles and dilemmas in cognitive gambling research. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 20: 1-21. Demiralp T, Herrmann CS, Erdal ME, Ergenoglu T, Keskin YH, Ergen M, Beydagi H (2007) DRD4 and DAT1 polymorphisms modulate human gamma band responses. Cereb Cortex 17: 1007-1019. Descarries L, Lemay B, Doucet G, Berger B (1987) Regional and laminar density of the dopamine innervation in adult rat cerebral cortex. Neuroscience 21: 807-824. Deumens R, Blokland A, Prickaerts J (2002) Modeling Parkinson's disease in rats: an evaluation of 6-OHDA lesions of the nigrostriatal pathway. Experimental neurology 175: 303-17. Di Ciano P, Cardinal RN, Cowell RA, Little SJ, Everitt BJ (2001) Differential involvement of NMDA, AMPA/kainate, and dopamine receptors in the nucleus accumbens core in the acquisition and performance of pavlovian approach behavior. J Neurosci 21: 9471-7. Di Ciano P, Pushparaj A, Kim A, Hatch J, Masood T, Ramzi A, Khaled MA, Boileau I, Winstanley CA, Le Foll B (2015) The Impact of Selective Dopamine D2, D3 and D4 Ligands on the Rat Gambling Task. PLoS One 10: e0136267. Di Paolo T (1994) Modulation of brain dopamine transmission by sex steroids. Reviews in the neurosciences 5: 27-41. Di Pietro NC, Mashhoon Y, Heaney C, Yager LM, Kantak KM (2008) Role of dopamine D1 receptors in the prefrontal dorsal agranular insular cortex in mediating cocaine self-administration in rats. Psychopharmacology 200: 81-91. Dickerson M (1993) Internal and external determinants of persistent gambling: Problems in generalising from one form of gambling to another. Journal of Gambling Studies 9: 225-245. Dixon MR, Marley J, Jacobs EA (2003) Delay discounting by pathological gamblers. Journal of applied behavior analysis 36: 449-458. 198  Dodd ML, Klos KJ, Bower JH, Geda YE, Josephs KA, Ahlskog JE (2005) Pathological gambling caused by drugs used to treat Parkinson disease. Arch Neurol 62: 1377-81. Dong Y, Green T, Saal D, Marie H, Neve R, Nestler EJ, Malenka RC (2006) CREB modulates excitability of nucleus accumbens neurons. Nat Neurosci 9: 475-7. Dowling N, Smith D, Thomas T (2005) Electronic gaming machines: are they the 'crack-cocaine' of gambling? Addiction 100: 33-45. Droutman V, Read SJ, Bechara A (2015) Revisiting the role of the insula in addiction. Trends in cognitive sciences 19: 414-20. Dulawa SC, Grandy DK, Low MJ, Paulus MP, Geyer MA (1999) Dopamine D4 receptor-knock-out mice exhibit reduced exploration of novel stimuli. J Neurosci 19: 9550-6. Eagle DM, Noschang C, d'Angelo LS, Noble CA, Day JO, Dongelmans ML, Theobald DE, Mar AC, Urcelay GP, Morein-Zamir S, Robbins TW (2014) The dopamine D2/D3 receptor agonist quinpirole increases checking-like behaviour in an operant observing response task with uncertain reinforcement: a novel possible model of OCD. Behav Brain Res 264: 207-29. Enman NM, Unterwald EM (2012) Inhibition of GSK3 attenuates amphetamine-induced hyperactivity and sensitization in the mouse. Behav Brain Res 231: 217-25. Epperson CN, Wisner KL, Yamamoto B (1999) Gonadal steroids in the treatment of mood disorders. Psychosomatic medicine 61: 676-97. Eshel N, Nelson EE, Blair RJ, Pine DS, Ernst M (2007) Neural substrates of choice selection in adults and adolescents: development of the ventrolateral prefrontal and anterior cingulate cortices. Neuropsychologia 45: 1270-9. Evenden JL (1999) Varieties of impulsivity. Psychopharmacology (Berl) 146: 348-61. Evenden JL, Ryan CN (1996) The pharmacology of impulsive behaviour in rats: the effects of drugs on response choice with varying delays of reinforcement. Psychopharmacology 128: 161-70. Everitt BJ, Robbins TW (2005) Neural systems of reinforcement for drug addiction: from actions to habits to compulsion. Nat Neurosci 8: 1481-9. Everitt BJ, Wolf ME (2002) Psychomotor stimulant addiction: A neural systems perspective. Journal of Neuroscience 22: 3312-3320. Fahn S (1999) Parkinson disease, the effect of levodopa, and the ELLDOPA trial. Earlier vs Later L-DOPA. Arch Neurol 56: 529-35. Fahn S, Oakes D, Shoulson I, Kieburtz K, Rudolph A, Lang A, Olanow CW, Tanner C, Marek K, Parkinson Study G (2004) Levodopa and the progression of Parkinson's disease. N Engl J Med 351: 2498-508. Field M, Cox WM (2008) Attentional bias in addictive behaviors: a review of its development, causes, and consequences. Drug Alcohol Depend 97: 1-20. Fineberg NA, Potenza MN, Chamberlain SR, Berlin HA, Menzies L, Bechara A, Sahakian BJ, Robbins TW, Bullmore ET, Hollander E (2010) Probing compulsive and impulsive behaviors, from animal models to endophenotypes: a narrative review. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 35: 591-604. Fiorillo CD, Tobler PN, Schultz W (2003) Discrete coding of reward probability and uncertainty by dopamine neurons. Science 299: 1898-902. 199  Fitzpatrick CJ, Gopalakrishnan S, Cogan ES, Yager LM, Meyer PJ, Lovic V, Saunders BT, Parker CC, Gonzales NM, Aryee E, Flagel SB, Palmer AA, Robinson TE, Morrow JD (2013) Variation in the form of Pavlovian conditioned approach behavior among outbred male Sprague-Dawley rats from different vendors and colonies: sign-tracking vs. goal-tracking. PLoS One 8: e75042. Flagel SB, Akil H, Robinson TE (2009) Individual differences in the attribution of incentive salience to reward-related cues: Implications for addiction. Neuropharmacology 56 Suppl 1: 139-48. Flagel SB, Clark JJ, Robinson TE, Mayo L, Czuj A, Willuhn I, Akers CA, Clinton SM, Phillips PE, Akil H (2011) A selective role for dopamine in stimulus-reward learning. Nature 469: 53-7. Flagel SB, Robinson TE, Clark JJ, Clinton SM, Watson SJ, Seeman P, Phillips PE, Akil H (2010) An animal model of genetic vulnerability to behavioral disinhibition and responsiveness to reward-related cues: implications for addiction. Neuropsychopharmacology 35: 388-400. Flagel SB, Watson SJ, Akil H, Robinson TE (2008) Individual differences in the attribution of incentive salience to a reward-related cue: influence on cocaine sensitization. Behav Brain Res 186: 48-56. Flagel SB, Watson SJ, Robinson TE, Akil H (2007) Individual differences in the propensity to approach signals vs goals promote different adaptations in the dopamine system of rats. Psychopharmacology (Berl) 191: 599-607. Fletcher PJ, Higgins GA (1997) Differential effects of ondansetron and alpha-flupenthixol on responding for conditioned reward. Psychopharmacology (Berl) 134: 64-72. Floresco SB, Ghods-Sharifi S, Vexelman C, Magyar O (2006a) Dissociable roles for the nucleus accumbens core and shell in regulating set shifting. The Journal of neuroscience : the official journal of the Society for Neuroscience 26: 2449-57. Floresco SB, Magyar O, Ghods-Sharifi S, Vexelman C, Tse MT (2006b) Multiple dopamine receptor subtypes in the medial prefrontal cortex of the rat regulate set-shifting. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 31: 297-309. Floresco SB, St Onge JR, Ghods-Sharifi S, Winstanley CA (2008) Cortico-limbic-striatal circuits subserving different forms of cost-benefit decision making. Cogn Affect Behav Neurosci 8: 375-89. Floresco SB, Tse MT (2007) Dopaminergic regulation of inhibitory and excitatory transmission in the basolateral amygdala-prefrontal cortical pathway. J Neurosci 27: 2045-57. Fontenelle LF, Mendlowicz MV, Versiani M (2005) Impulse control disorders in patients with obsessive-compulsive disorder. Psychiatry and clinical neurosciences 59: 30-7. Forget B, Pushparaj A, Le Foll B (2010) Granular insular cortex inactivation as a novel therapeutic strategy for nicotine addiction. Biol Psychiatry 68: 265-71. Frank MJ, O'Reilly RC (2006) A mechanistic account of striatal dopamine function in human cognition: Psychopharmacological studies with cabergoline and haloperidol. Behav Neurosci 120: 497-517. Frank MJ, Samanta J, Moustafa AA, Sherman SJ (2007) Hold your horses: impulsivity, deep brain stimulation, and medication in parkinsonism. Science 318: 1309-12. 200  Frank MJ, Seeberger LC, O'Reilly R C (2004) By carrot or by stick: cognitive reinforcement learning in parkinsonism. Science 306: 1940-3. Fraser KM, Haight JL, Gardner EL, Flagel SB (2016) Examining the role of dopamine D2 and D3 receptors in Pavlovian conditioned approach behaviors. Behav Brain Res 305: 87-99. Gaboury A, Ladouceur R (1989) Erroneous Perceptions and Gambling. J Soc Behav Pers 4: 411-420. Galynker, II, Cai J, Ongseng F, Finestone H, Dutta E, Serseni D (1998) Hypofrontality and negative symptoms in major depressive disorder. J Nucl Med 39: 608-12. Gaspar P, Berger B, Febvret A, Vigny A, Henry JP (1989) Catecholamine innervation of the human cerebral cortex as revealed by comparative immunohistochemistry of tyrosine hydroxylase and dopamine-beta-hydroxylase. The Journal of comparative neurology 279: 249-71. Gazi L, Bobirnac I, Danzeisen M, Schupbach E, Bruinvels AT, Geisse S, Sommer B, Hoyer D, Tricklebank M, Schoeffter P (1998) The agonist activities of the putative antipsychotic agents, L-745,870 and U-101958 in HEK293 cells expressing the human dopamine D4.4 receptor. British journal of pharmacology 124: 889-96. Gazi L, Bobirnac I, Danzeisen M, Schupbach E, Langenegger D, Sommer B, Hoyer D, Tricklebank M, Schoeffter P (1999) Receptor density as a factor governing the efficacy of the dopamine D4 receptor ligands, L-745,870 and U-101958 at human recombinant D4.4 receptors expressed in CHO cells. Br J Pharmacol 128: 613-20. Gerstein D, Hoffman. J, Larison, C., Engel,am, L, Murphy. S, Palmer. A, Chuchro. L, Toce. M, Johnson. R, Buie. T & Hill. M. A. (1999) Gambling Impact and Behavior study. Report to the National Gambling Impact Study Commission. Glase SA, Akunne HC, Georgic LM, Heffner TG, MacKenzie RG, Manley PJ, Pugsley TA, Wise LD (1997) Substituted [(4-phenylpiperazinyl)-methyl]benzamides: selective dopamine D4 agonists. J Med Chem 40: 1771-2. Gonzalez-Ibanez A, Rosel P, Moreno I (2005) Evaluation and treatment of pathological gambling. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 21: 35-42. Goto Y, Otani S, Grace AA (2007) The Yin and Yang of dopamine release: a new perspective. Neuropharmacology 53: 583-7. Goudriaan AE, Oosterlaan J, de Beurs E, van den Brink W (2005) Decision making in pathological gambling: a comparison between pathological gamblers, alcohol dependents, persons with Tourette syndrome, and normal controls. Brain Res Cogn Brain Res 23: 137-51. Grabenhorst F, Rolls ET (2011) Value, pleasure and choice in the ventral prefrontal cortex. Trends in cognitive sciences 15: 56-67. Grace AA (1991) Phasic versus tonic dopamine release and the modulation of dopamine system responsivity: a hypothesis for the etiology of schizophrenia. Neuroscience 41: 1-24. Grace AA, Laviolette SR, Lipski WJ (2005) Amygdala modulation of conditioning within the prefrontal cortex is dependent on cannabinoid and dopamine D4 receptor activation. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 30: S13-S13. 201  Grant JE, Kim SW (2001) Demographic and clinical features of 131 adult pathological gamblers. The Journal of clinical psychiatry 62: 957-62. Grant JE, Kim SW (2006) Medication management of pathological gambling. Minnesota medicine 89: 44-8. Grant JE, Odlaug BL, Schreiber LR (2012) Pharmacological Treatments in Pathological Gambling. British journal of clinical pharmacology. Griffiths M (1999) Gambling Technologies: Prospects for Problem Gambling. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 15: 265-283. Gu Z, Yan Z (2004) Bidirectional regulation of Ca2+/calmodulin-dependent protein kinase II activity by dopamine D4 receptors in prefrontal cortex. Mol Pharmacol 66: 948-55. Habib R, Dixon MR (2010) Neurobehavioral evidence for the 'near-miss' effect in pathological gamblers. . J Exp Anal Behav 93: 313-328. Hall H, Sallemark M, Jerning E (1986) Effects of Remoxipride and Some Related New Substituted Salicylamides on Rat-Brain Receptors. Acta pharmacologica et toxicologica 58: 61-70. Hall J, Parkinson JA, Connor TM, Dickinson A, Everitt BJ (2001) Involvement of the central nucleus of the amygdala and nucleus accumbens core in mediating Pavlovian influences on instrumental behaviour. The European journal of neuroscience 13: 1984-92. Hamilton KR, Mitchell MR, Wing VC, Balodis IM, Bickel WK, Fillmore M, Lane SD, Lejuez CW, Littlefield AK, Luijten M, Mathias CW, Mitchell SH, Napier TC, Reynolds B, Schutz CG, Setlow B, Sher KJ, Swann AC, Tedford SE, White MJ, Winstanley CA, Yi R, Potenza MN, Moeller FG (2015) Choice impulsivity: Definitions, measurement issues, and clinical implications. Personal Disord 6: 182-98. Harrigan KA (2007) Slot Machines: Pursuing Responsible Gaming Practices for Virtual Reels and Near Misses. International Journal of Mental Health and Addiction 7: 68-83. Hayden BY, Platt ML (2009) Gambling for Gatorade: risk-sensitive decision making for fluid rewards in humans. Animal cognition 12: 201-7. Heidbreder CA, Gardner EL, Xi ZX, Thanos PK, Mugnaini M, Hagan JJ, Ashby CR, Jr. (2005) The role of central dopamine D3 receptors in drug addiction: a review of pharmacological evidence. Brain research Brain research reviews 49: 77-105. Heidbreder CA, Newman AH (2010) Current perspectives on selective dopamine D(3) receptor antagonists as pharmacotherapeutics for addictions and related disorders. Annals of the New York Academy of Sciences 1187: 4-34. Heinz A (2004) Correlation between dopamine D-2 receptors in the ventral striatum and central processing of alcohol cues and craving (vol 161, pg 1783, 2004). Am J Psychiat 161: 2344-2344. Henslin JM (1967) Craps and Magic. Am J Sociol 73: 316-330. Higley AE, Kiefer SW, Li X, Gaal J, Xi ZX, Gardner EL (2011) Dopamine D(3) receptor antagonist SB-277011A inhibits methamphetamine self-administration and methamphetamine-induced reinstatement of drug-seeking in rats. European journal of pharmacology 659: 187-92. Ho MY, Mobini S, Chiang TJ, Bradshaw CM, Szabadi E (1999) Theory and method in the quantitative analysis of "impulsive choice" behaviour: implications for psychopharmacology. Psychopharmacology 146: 362-72. 202  Hollander E, Pallanti S, Allen A, Sood E, Rossi NB (2005) Does sustained-release lithium reduce impulsive gambling and affective instability versus placebo in pathological gamblers with bipolar spectrum disorders? Am J Psychiat 162: 137-145. Holroyd CB, McClure SM (2015) Hierarchical control over effortful behavior by rodent medial frontal cortex: A computational model. Psychological review 122: 54-83. Hosking JG, Cocker PJ, Winstanley CA (2014a) Dissociable contributions of anterior cingulate cortex and basolateral amygdala on a rodent cost/benefit decision-making task of cognitive effort. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. Hosking JG, Lam FC, Winstanley CA (2014b) Nicotine increases impulsivity and decreases willingness to exert cognitive effort despite improving attention in "slacker" rats: insights into cholinergic regulation of cost/benefit decision making. PLoS One 9: e111580. Howes OD, Kapur S (2009) The dopamine hypothesis of schizophrenia: version III--the final common pathway. Schizophrenia bulletin 35: 549-62. Hyman JM, Whitman J, Emberly E, Woodward TS, Seamans JK (2013) Action and outcome activity state patterns in the anterior cingulate cortex. Cereb Cortex 23: 1257-68. Iida M, Miyazaki I, Tanaka K, Kabuto H, Iwata-Ichikawa E, Ogawa N (1999) Dopamine D2 receptor-mediated antioxidant and neuroprotective effects of ropinirole, a dopamine agonist. Brain research 838: 51-9. Ishii H, Ohara S, Tobler PN, Tsutsui K, Iijima T (2012) Inactivating anterior insular cortex reduces risk taking. J Neurosci 32: 16031-9. J AJ (2000) Canadian Gambling Behaviour and Attitudes. Gambling in Canada Research report  Jaber M, Robinson SW, Missale C, Caron MG (1996) Dopamine receptors and brain function. Neuropharmacology 35: 1503-19. Jentsch JD, Taylor JR (1999) Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related stimuli. Psychopharmacology 146: 373-90. Jones EG (1985) The Thalamus. Plenum Press, New York Joukhador J, Maccallum F, Blaszczynski A (2003) Differences in cognitive distortions between problem and social gamblers. Psychological reports 92: 1203-1214. Kaasinen V, Vilkman H, Hietala J, Nagren K, Helenius H, Olsson H, Farde L, Rinne J (2000) Age-related dopamine D2/D3 receptor loss in extrastriatal regions of the human brain. Neurobiology of aging 21: 683-8. Kabitzke PA, Silva L, Wiedenmayer C (2011) Norepinephrine mediates contextual fear learning and hippocampal pCREB in juvenile rats exposed to predator odor. Neurobiol Learn Mem 96: 166-72. Kahneman D, Tversky A (1979) Prospect Theory - Analysis of Decision under Risk. Econometrica 47: 263-291. Kallmen H, Andersson P, Andren A (2008) Are irrational beliefs and depressive mood more common among problem gamblers than non-gamblers? A survey study of Swedish problem gamblers and controls. Journal of Gambling Studies 24: 441-50. Kandel ER (2012) The molecular biology of memory: cAMP, PKA, CRE, CREB-1, CREB-2, and CPEB. Mol Brain 5: 14. Kassinove JI, Schare ML (2001) Effects of the "near miss" and the "big win" on persistence at slot machine gambling. Psychology of Addictive Behaviors 15: 155-158. 203  Kelley AE, Domesick VB (1982) The distribution of the projection from the hippocampal formation to the nucleus accumbens in the rat: an anterograde- and retrograde-horseradish peroxidase study. Neuroscience 7: 2321-35. Kelley AE, Stinus L (1984) The distribution of the projection from the parataenial nucleus of the thalamus to the nucleus accumbens in the rat: an autoradiographic study. Exp Brain Res 54: 499-512. Kennerley SW, Wallis JD (2009) Evaluating choices by single neurons in the frontal lobe: outcome value encoded across multiple decision variables. The European journal of neuroscience 29: 2061-73. Kennerley SW, Walton ME, Behrens TE, Buckley MJ, Rushworth MF (2006) Optimal decision making and the anterior cingulate cortex. Nat Neurosci 9: 940-7. Kertzman S, Lowengrub K, Aizer A, Vainder M, Kotler M, Dannon PN (2008) Go-no-go performance in pathological gamblers. Psychiatry research 161: 1-10. Khaled MA, Farid Araki K, Li B, Coen KM, Marinelli PW, Varga J, Gaal J, Le Foll B (2010) The selective dopamine D3 receptor antagonist SB 277011-A, but not the partial agonist BP 897, blocks cue-induced reinstatement of nicotine-seeking. Int J Neuropsychopharmacol 13: 181-90. Khan ZU, Gutierrez A, Martin R, Penafiel A, Rivera A, De La Calle A (1998) Differential regional and cellular distribution of dopamine D2-like receptors: an immunocytochemical study of subtype-specific antibodies in rat and human brain. J Comp Neurol 402: 353-71. Kirmayer LJ, Crafa D (2014) What kind of science for psychiatry? Front Hum Neurosci 8. Koffarnus MN, Newman AH, Grundt P, Rice KC, Woods JH (2011) Effects of selective dopaminergic compounds on a delay-discounting task. Behav Pharmacol 22: 300-11. Koob GF, Volkow ND (2010) Neurocircuitry of addiction. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 35: 217-38. Kramer MS, Last B, Getson A, Reines SA (1997) The effects of a selective D4 dopamine receptor antagonist (L-745,870) in acutely psychotic inpatients with schizophrenia. D4 Dopamine Antagonist Group. Archives of general psychiatry 54: 567-72. Kusumoto-Yoshida I, Liu H, Chen BT, Fontanini A, Bonci A (2015) Central role for the insular cortex in mediating conditioned responses to anticipatory cues. Proc Natl Acad Sci U S A 112: 1190-5. Kutlu MG, Burke D, Slade S, Hall BJ, Rose JE, Levin ED (2013) Role of insular cortex D(1) and D(2) dopamine receptors in nicotine self-administration in rats. Behavioural brain research 256: 273-8. Ladouceur R, Gaboury A, Dumont M, Rochette P (1988) Gambling - Relationship between the Frequency of Wins and Irrational Thinking. Journal of Psychology 122: 409-414. Ladouceur R, Jacques C, Ferland F, Giroux I (1999) Prevalence of problem gambling: a replication study 7 years later. Canadian journal of psychiatry Revue canadienne de psychiatrie 44: 802-4. Ladouceur R, Sylvain C, Boutin C, Lachance S, Doucet C, Leblond J, Jacques C (2001) Cognitive treatment of pathological gambling. The Journal of nervous and mental disease 189: 774-80. Lauzon NM, Ahmad T, Laviolette SR (2012) Dopamine D4 Receptor Transmission in the Prefrontal Cortex Controls the Salience of Emotional Memory via Modulation of Calcium Calmodulin-Dependent Kinase II. Cerebral Cortex 22: 2486-2494. 204  Lauzon NM, Bishop SF, Laviolette SR (2009) Dopamine D1 versus D4 receptors differentially modulate the encoding of salient versus nonsalient emotional information in the medial prefrontal cortex. J Neurosci 29: 4836-45. Lauzon NM, Laviolette SR (2010) Dopamine D4-receptor modulation of cortical neuronal network activity and emotional processing: Implications for neuropsychiatric disorders. Behav Brain Res 208: 12-22. Lawrence AJ, Luty J, Bogdan NA, Sahakian BJ, Clark L (2009) Problem gamblers share deficits in impulsive decision-making with alcohol-dependent individuals. Addiction 104: 1006-15. Le Foll B, Gallo A, Le Strat Y, Lu L, Gorwood P (2009) Genetics of dopamine receptors and drug addiction: a comprehensive review. Behav Pharmacol 20: 1-17. Le Foll B, Goldberg SR, Sokoloff P (2005) The dopamine D3 receptor and drug dependence: effects on reward or beyond? Neuropharmacology 49: 525-41. Lee T, Seeman P, Rajput A, Farley IJ, Hornykiewicz O (1978) Receptor basis for dopaminergic supersensitivity in Parkinson's disease. Nature 273: 59-61. Leeman RF, Potenza MN (2012) Similarities and differences between pathological gambling and substance use disorders: a focus on impulsivity and compulsivity. Psychopharmacology (Berl) 219: 469-90. Lesieur HR (1979) The compulsive gambler's spiral of options and involvement. Psychiatry 42: 79-87. Lesieur HR, Blume SB (1987) The South Oaks Gambling Screen (SOGS): a new instrument for the identification of pathological gamblers. The American journal of psychiatry 144: 1184-8. Leung KS, Cottler LB (2009) Treatment of pathological gambling. Current opinion in psychiatry 22: 69-74. Levesque D (2013) Chronic lithium treatment influences impulse control and cognition and regulates gene expression. Retrieved from: Vancouver : University of British Columbia Library Masters thesis. Lewis DA (1992) The catecholaminergic innervation of primate prefrontal cortex. Journal of neural transmission Supplementum 36: 179-200. Leyton M, Vezina P (2012) On cue: striatal ups and downs in addictions. Biol Psychiatry 72: e21-2. Li YC, Gao WJ (2011) GSK-3 beta activity and hyperdopamine-dependent behaviors. Neuroscience and biobehavioral reviews 35: 645-654. Li YC, Wang MJ, Gao WJ (2012) Hyperdopaminergic modulation of inhibitory transmission is dependent on GSK-3 beta signaling-mediated trafficking of GABA(A) receptors. J Neurochem 122: 308-320. Limbrick-Oldfield EH, van Holst RJ, Clark L (2013) Fronto-striatal dysregulation in drug addiction and pathological gambling: Consistent inconsistencies? Neuroimage Clin 2: 385-93. Linnet J, Moller A, Peterson E, Gjedde A, Doudet D (2011) Dopamine release in ventral striatum during Iowa Gambling Task performance is associated with increased excitement levels in pathological gambling. Addiction 106: 383-390. 205  Liu X, Hairston J, Schrier M, Fan J (2011) Common and distinct networks underlying reward valence and processing stages: a meta-analysis of functional neuroimaging studies. Neuroscience and biobehavioral reviews 35: 1219-36. Lopez JC, Karlsson RM, O'Donnell P (2015) Dopamine D2 Modulation of Sign and Goal Tracking in Rats. Neuropsychopharmacology 40: 2096-102. Magno E, Foxe JJ, Molholm S, Robertson IH, Garavan H (2006) The anterior cingulate and error avoidance. The Journal of neuroscience : the official journal of the Society for Neuroscience 26: 4769-73. Maletzky BM, Shore JH (1978) Lithium treatment for psychiatric disorders. West J Med 128: 488-98. Marquis JP, Killcross S, Haddon JE (2007) Inactivation of the prelimbic, but not infralimbic, prefrontal cortex impairs the contextual control of response conflict in rats. European Journal of Neuroscience 25: 559-566. Marsden CA (2006) Dopamine: the rewarding years. British journal of pharmacology 147 Suppl 1: S136-44. Martin RJ, Usdan S, Cremeens J, Vail-Smith K (2013) Disordered gambling and co-morbidity of psychiatric disorders among college students: An examination of problem drinking, anxiety and depression. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming. Matsukawa N, Maki M, Yasuhara T, Hara K, Yu G, Xu L, Kim KM, Morgan JC, Sethi KD, Borlongan CV (2007) Overexpression of D2/D3 receptors increases efficacy of ropinirole in chronically 6-OHDA-lesioned Parkinsonian rats. Brain research 1160: 113-23. Matsumoto M, Hidaka K, Tada S, Tasaki Y, Yamaguchi T (1996) Low levels of mRNA for dopamine D4 receptor in human cerebral cortex and striatum. J Neurochem 66: 915-9. McFarland NR, Haber SN (2002) Thalamic relay nuclei of the basal ganglia form both reciprocal and nonreciprocal cortical connections, linking multiple frontal cortical areas. J Neurosci 22: 8117-32. McHugh MJ, Demers CH, Braud J, Briggs R, Adinoff B, Stein EA (2013) Striatal-insula circuits in cocaine addiction: implications for impulsivity and relapse risk. Am J Drug Alcohol Abuse 39: 424-32. Michalczuk R, Bowden-Jones H, Verdejo-Garcia A, Clark L (2011) Impulsivity and cognitive distortions in pathological gamblers attending the UK National Problem Gambling Clinic: a preliminary report. Psychol Med 41: 2625-2635. Miedl SF, Buchel C, Peters J (2014) Cue-Induced Craving Increases Impulsivity via Changes in Striatal Value Signals in Problem Gamblers. Journal of Neuroscience 34: 4750-4755. Millan MJ, Brocco M, Papp M, Serres F, La Rochelle CD, Sharp T, Peglion JL, Dekeyne A (2004) S32504, a novel naphtoxazine agonist at dopamine D3/D2 receptors: III. Actions in models of potential antidepressive and anxiolytic activity in comparison with ropinirole. J Pharmacol Exp Ther 309: 936-50. Millan MJ, Gobert A, Newman-Tancredi A, Lejeune F, Cussac D, Rivet JM, Audinot V, Dubuffet T, Lavielle G (2000) S33084, a novel, potent, selective, and competitive antagonist at dopamine D-3-receptors: I. Receptorial, electrophysiological and neurochemical profile compared with GR218,231 and L741,626. J Pharmacol Exp Ther 293: 1048-1062. 206  Millan MJ, Maiofiss L, Cussac D, Audinot V, Boutin JA, Newman-Tancredi A (2002) Differential actions of antiparkinson agents at multiple classes of monoaminergic receptor. I. A multivariate analysis of the binding profiles of 14 drugs at 21 native and cloned human receptor subtypes. J Pharmacol Exp Ther 303: 791-804. Miller JS, Tallarida RJ, Unterwald EM (2009) Cocaine-induced hyperactivity and sensitization are dependent on GSK3. Neuropharmacology 56: 1116-23. Milstein J, Dalley J, Robbins T (2010) Methylphenidate-induced impulsivity: pharmacological antagonism by {beta}-adrenoreceptor blockade. J Psychopharmacol 24: 309-21. Mizoguchi H, Katahira K, Inutsuka A, Fukumoto K, Nakamura A, Wang T, Nagai T, Sato J, Sawada M, Ohira H, Yamanaka A, Yamada K (2015) Insular neural system controls decision-making in healthy and methamphetamine-treated rats. Proc Natl Acad Sci U S A 112: E3930-9. Moodie C, Finnigan F (2005) A comparison of the autonomic arousal of frequent, infrequent and non-gamblers while playing fruit machines. Addiction 100: 51-9. Morgan PT, Desai RA, Potenza MN (2010) Gender-related influences of parental alcoholism on the prevalence of psychiatric illnesses: analysis of the National Epidemiologic Survey on Alcohol and Related Conditions. Alcohol Clin Exp Res 34: 1759-67. Morris SE, Cuthbert BN (2012) Research Domain Criteria: cognitive systems, neural circuits, and dimensions of behavior. Dialogues in clinical neuroscience 14: 29-37. Morrison SE, Bamkole MA, Nicola SM (2015) Sign Tracking, but Not Goal Tracking, is Resistant to Outcome Devaluation. Front Neurosci 9: 468. Mrzljak L, Bergson C, Pappy M, Huff R, Levenson R, Goldman-Rakic PS (1996) Localization of dopamine D4 receptors in GABAergic neurons of the primate brain. Nature 381: 245-8. 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. Cerebral Cortex 6: 470-481. Murch WS, Clark L (2015) Games in the Brain: Neural Substrates of Gambling Addiction. Neuroscientist. Naqvi NH, Bechara A (2009) The hidden island of addiction: the insula. Trends Neurosci 32: 56-67. Naqvi NH, Rudrauf D, Damasio H, Bechara A (2007) Damage to the insula disrupts addiction to cigarette smoking. Science 315: 531-4. Nashatizadeh MM, Lyons KE, Pahwa R (2009) A review of ropinirole prolonged release in Parkinson's disease. Clin Interv Aging 4: 179-86. Neighbors C, Lostutter TW, Cronce JM, Larimer ME (2002) Exploring college student gambling motivation. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 18: 361-70. Nestler EJ, Barrot M, DiLeone RJ, Eisch AJ, Gold SJ, Monteggia LM (2002) Neurobiology of depression. Neuron 34: 13-25. Nestler EJ, Carlezon WA, Jr. (2006) The mesolimbic dopamine reward circuit in depression. Biol Psychiatry 59: 1151-9. Noble EP (2000) Addiction and its reward process through polymorphisms of the D-2 dopamine receptor gene: a review. Eur Psychiat 15: 79-89. 207  Noel X, Brevers D, Bechara A (2013) A neurocognitive approach to understanding the neurobiology of addiction. Curr Opin Neurobiol 23: 632-8. Oak JN, Oldenhof J, Van Tol HH (2000) The dopamine D(4) receptor: one decade of research. Eur J Pharmacol 405: 303-27. Oei TPS, Gordon LM (2008) Psychosocial factors related to gambling abstinence and relapse in members of gamblers anonymous. Journal of Gambling Studies 24: 91-105. Olanow CW, Stern MB, Sethi K (2009) The scientific and clinical basis for the treatment of Parkinson disease (2009). Neurology 72: S1-136. Olsson M, Nikkhah G, Bentlage C, Bjorklund A (1995) Forelimb akinesia in the rat Parkinson model: differential effects of dopamine agonists and nigral transplants as assessed by a new stepping test. J Neurosci 15: 3863-75. Ongur D, Price JL (2000) The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb Cortex 10: 206-19. Paine TA, Asinof SK, Diehl GW, Frackman A, Leffler J (2013) Medial prefrontal cortex lesions impair decision-making on a rodent gambling task: reversal by D1 receptor antagonist administration. Behavioural brain research 243: 247-54. Pallanti S, Quercioli L, Sood E, Hollander E (2002) Lithium and valproate treatment of pathological gambling: A randomized single-blind study. J Clin Psychiat 63: 559-564. Parkes SL, Balleine BW (2013) Incentive memory: evidence the basolateral amygdala encodes and the insular cortex retrieves outcome values to guide choice between goal-directed actions. J Neurosci 33: 8753-63. Parkes SL, Bradfield LA, Balleine BW (2015) Interaction of insular cortex and ventral striatum mediates the effect of incentive memory on choice between goal-directed actions. J Neurosci 35: 6464-71. Patel S, Freedman S, Chapman KL, Emms F, Fletcher AE, Knowles M, Marwood R, McAllister G, Myers J, Curtis N, Kulagowski JJ, Leeson PD, Ridgill M, Graham M, Matheson S, Rathbone D, Watt AP, Bristow LJ, Rupniak NM, Baskin E, Lynch JJ, Ragan CI (1997) Biological profile of L-745,870, a selective antagonist with high affinity for the dopamine D4 receptor. J Pharmacol Exp Ther 283: 636-47. Patton JH, Stanford MS, Barratt ES (1995) Factor structure of the Barratt impulsiveness scale. Journal of clinical psychology 51: 768-74. Paulus MP, Stein MB (2006) An insular view of anxiety. Biological psychiatry 60: 383-7. Pei Y, Dong S, Roth BL (2010) Generation of designer receptors exclusively activated by designer drugs (DREADDs) using directed molecular evolution. Curr Protoc Neurosci Chapter 4: Unit 4 33. Petry NM (2001) Pathological gamblers, with and without substance use disorders, discount delayed rewards at high rates. Journal of abnormal psychology 110: 482-487. Petry NM (2003) A comparison of treatment-seeking pathological gamblers based on preferred gambling activity. Addiction 98: 645-55. Petry NM, Stinson FS, Grant BF (2005) Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. The Journal of clinical psychiatry 66: 564-74. Piazza NJ, Vrbka JL, Yeager RD (1989a) Telescoping of alcoholism in women alcoholics. The International journal of the addictions 24: 19-28. 208  Piazza PV, Deminiere JM, Le Moal M, Simon H (1989b) Factors that predict individual vulnerability to amphetamine self-administration. Science 245: 1511-3. Potenza MN (2006) Should addictive disorders include non-substance-related conditions? Addiction 101 Suppl 1: 142-51. Potenza MN (2007) Impulsivity and compulsivity in pathological gambling and obsessive-compulsive disorder. Rev Bras Psiquiatr 29: 105-6. Potenza MN (2008) Review. The neurobiology of pathological gambling and drug addiction: an overview and new findings. Philosophical transactions of the Royal Society of London Series B, Biological sciences 363: 3181-9. Potenza MN (2013) How central is dopamine to pathological gambling or gambling disorder? Frontiers in behavioral neuroscience 7. Potenza MN, Leung HC, Blumberg HP, Peterson BS, Fulbright RK, Lacadie CM, Skudlarski P, Gore JC (2003) An FMRI Stroop task study of ventromedial prefrontal cortical function in pathological gamblers. The American journal of psychiatry 160: 1990-4. Power Y, Goodyear B, Crockford D (2012) Neural correlates of pathological gamblers preference for immediate rewards during the iowa gambling task: an fMRI study. J Gambl Stud 28: 623-36. Primus RJ, Thurkauf A, Xu J, Yevich E, McInerney S, Shaw K, Tallman JF, Gallagher DW (1997) II. Localization and characterization of dopamine D4 binding sites in rat and human brain by use of the novel, D4 receptor-selective ligand [3H]NGD 94-1. J Pharmacol Exp Ther 282: 1020-7. Przedborski S, Levivier M, Jiang H, Ferreira M, Jackson-Lewis V, Donaldson D, Togasaki DM (1995) Dose-dependent lesions of the dopaminergic nigrostriatal pathway induced by intrastriatal injection of 6-hydroxydopamine. Neuroscience 67: 631-47. Pushparaj A, Hamani C, Yu W, Shin DS, Kang B, Nobrega JN, Le Foll B (2012) Electrical stimulation of the insular region attenuates nicotine-taking and nicotine-seeking behaviors. Neuropsychopharmacology 38: 690-8. Pushparaj A, Kim AS, Musiol M, Trigo JM, Le Foll B (2015a) Involvement of the rostral agranular insular cortex in nicotine self-administration in rats. Behavioural brain research 290: 77-83. Pushparaj A, Kim AS, Musiol M, Zangen A, Daskalakis ZJ, Zack M, Winstanley CA, Le Foll B (2015b) Differential Involvement of the Agranular vs Granular Insular Cortex in the Acquisition and Performance of Choice Behavior in a Rodent Gambling Task. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. Rachlin H (1990) Why Do People Gamble and Keep Gambling Despite Heavy Losses. Psychol Sci 1: 294-297. Raylu N, Oei TP (2004) The Gambling Related Cognitions Scale (GRCS): development, confirmatory factor validation and psychometric properties. Addiction 99: 757-69. Reavill C, Taylor SG, Wood MD, Ashmeade T, Austin NE, Avenell KY, Boyfield I, Branch CL, Cilia J, Coldwell MC, Hadley MS, Hunter AJ, Jeffrey P, Jewitt F, Johnson CN, Jones DN, Medhurst AD, Middlemiss DN, Nash DJ, Riley GJ, Routledge C, Stemp G, Thewlis KM, Trail B, Vong AK, Hagan JJ (2000) Pharmacological actions of a novel, high-affinity, and selective human dopamine D(3) receptor antagonist, SB-277011-A. J Pharmacol Exp Ther 294: 1154-65. 209  Redish AD, Schultheiss NW, Carter EC (2016) The Computational Complexity of Valuation and Motivational Forces in Decision-Making Processes. Curr Top Behav Neurosci 27: 313-33. Reid RL (1986) The Psychology of the near miss. Journal of Gambling Behavior 2: 32-39. Reuter J, Raedler T, Rose M, Hand I, Glascher J, Buchel C (2005) Pathological gambling is linked to reduced activation of the mesolimbic reward system. Nat Neurosci 8: 147-8. Reynolds SM, Zahm DS (2005) Specificity in the projections of prefrontal and insular cortex to ventral striatopallidum and the extended amygdala. The Journal of neuroscience : the official journal of the Society for Neuroscience 25: 11757-67. Richardson JV, Baron A (2008) Avoidance of timeout from response-independent food: effects of delivery rate and quality. J Exp Anal Behav 89: 169-81. Rivera A, Cuellar B, Giron FJ, Grandy DK, de la Calle A, Moratalla R (2002) Dopamine D4 receptors are heterogeneously distributed in the striosomes/matrix compartments of the striatum. J Neurochem 80: 219-29. Rivera A, Penafiel A, Megias M, Agnati LF, Lopez-Tellez JF, Gago B, Gutierrez A, de la Calle A, Fuxe K (2008) Cellular localization and distribution of dopamine D(4) receptors in the rat cerebral cortex and their relationship with the cortical dopaminergic and noradrenergic nerve terminal networks. Neuroscience 155: 997-1010. Robbins TW (2002) The 5-choice serial reaction time task: behavioural pharmacology and functional neurochemistry. Psychopharmacology 163: 362-380. Robbins TW, Cador M, Taylor JR, Everitt BJ (1989) Limbic-striatal interactions in reward-related processes. Neuroscience and biobehavioral reviews 13: 155-62. Robbins TW, Watson BA, Gaskin M, Ennis C (1983) Contrasting interactions of pipradrol, d-amphetamine, cocaine, cocaine analogues, apomorphine and other drugs with conditioned reinforcement. Psychopharmacology (Berl) 80: 113-9. Robinson MJ, Anselme P, Suchomel K, Berridge KC (2015) Amphetamine-induced sensitization and reward uncertainty similarly enhance incentive salience for conditioned cues. Behav Neurosci 129: 502-11. Robinson TE, Berridge KC (1993) The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain research Brain research reviews 18: 247-91. Robinson TE, Berridge KC (2001) Incentive-sensitization and addiction. Addiction 96: 103-14. Robinson TE, Berridge KC (2003) Addiction. Annu Rev Psychol 54: 25-53. Robinson TE, Berridge KC (2008) Review. The incentive sensitization theory of addiction: some current issues. Philos Trans R Soc Lond B Biol Sci 363: 3137-46. Roca M, Torralva T, Lopez P, Cetkovich M, Clark L, Manes F (2008) Executive functions in pathologic gamblers selected in an ecologic setting. Cognitive and behavioral neurology : official journal of the Society for Behavioral and Cognitive Neurology 21: 1-4. Rodriguez-Jimenez R, Avila C, Jimenez-Arriero MA, Ponce G, Monasor R, Jimenez M, Aragues M, Hoenicka J, Rubio G, Palomo T (2006) Impulsivity and sustained attention in pathological gamblers: influence of childhood ADHD history. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 22: 451-61. Rogers DC, Costall B, Domeney AM, Gerrard PA, Greener M, Kelly ME, Hagan JJ, Hunter AJ (2000) Anxiolytic profile of ropinirole in the rat, mouse and common marmoset. Psychopharmacology 151: 91-7. 210  Rogers RD, Lancaster M, Wakeley J, Bhagwagar Z (2004) Effects of beta-adrenoceptor blockade on components of human decision-making. Psychopharmacology (Berl) 172: 157-64. Rokosik SL, Napier TC (2012) Pramipexole-induced increased probabilistic discounting: comparison between a rodent model of Parkinson's disease and controls. Neuropsychopharmacology 37: 1397-408. Rossi M, Gerschcovich ER, de Achaval D, Perez-Lloret S, Cerquetti D, Cammarota A, Nouzeilles MI, Fahrer R, Merello M, Leiguarda R (2010) Decision-making in Parkinson's disease patients with and without pathological gambling. Eur J Neurol 17: 97-102. Rushworth MF, Buckley MJ, Behrens TE, Walton ME, Bannerman DM (2007) Functional organization of the medial frontal cortex. Curr Opin Neurobiol 17: 220-7. Samanez-Larkin GR, Hollon NG, Carstensen LL, Knutson B (2008) Individual differences in insular sensitivity during loss anticipation predict avoidance learning. Psychol Sci 19: 320-3. Sautel F, Griffon N, Levesque D, Pilon C, Schwartz JC, Sokoloff P (1995) A functional test identifies dopamine agonists selective for D3 versus D2 receptors. Neuroreport 6: 329-32. Schmidt TT, Rea E, Shababi-Klein J, Panagis G, Winter C (2013) Enhanced reward-facilitating effects of d-amphetamine in rats in the quinpirole model of obsessive-compulsive disorder. Int J Neuropsychopharmacol 16: 1083-91. Schultz W (1998) Predictive reward signal of dopamine neurons. J Neurophysiol 80: 1-27. Schultz W, Dayan P, Montague PR (1997) A neural substrate of prediction and reward. Science 275: 1593-9. Schultz W, Romo R (1990) Dopamine neurons of the monkey midbrain: contingencies of responses to stimuli eliciting immediate behavioral reactions. J Neurophysiol 63: 607-24. Seeman P, Van Tol HH (1993) Dopamine receptor pharmacology. Current opinion in neurology and neurosurgery 6: 602-8. Seif T, Chang SJ, Simms JA, Gibb SL, Dadgar J, Chen BT, Harvey BK, Ron D, Messing RO, Bonci A, Hopf FW (2013) Cortical activation of accumbens hyperpolarization-active NMDARs mediates aversion-resistant alcohol intake. Nat Neurosci 16: 1094-100. Self DW, Genova LM, Hope BT, Barnhart WJ, Spencer JJ, Nestler EJ (1998) Involvement of cAMP-dependent protein kinase in the nucleus accumbens in cocaine self-administration and relapse of cocaine-seeking behavior. J Neurosci 18: 1848-59. Sesack SR, Grace AA (2010) Cortico-Basal Ganglia reward network: microcircuitry. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 35: 27-47. Sescousse G, Caldu X, Segura B, Dreher JC (2013) Processing of primary and secondary rewards: a quantitative meta-analysis and review of human functional neuroimaging studies. Neuroscience and biobehavioral reviews 37: 681-96. Sescousse G, Janssen LK, Hashemi MM, Timmer MH, Geurts DE, Ter Huurne NP, Clark L, Cools R (2016) Amplified Striatal Responses to Near-Miss Outcomes in Pathological Gamblers. Neuropsychopharmacology. Severus WE, Kleindienst N, Seemuller F, Frangou S, Moller HJ, Greil W (2008) What is the optimal serum lithium level in the long-term treatment of bipolar disorder--a review? Bipolar Disord 10: 231-7. 211  Shaffer HJ, Hall MN (2001) Updating and refining prevalence estimates of disordered gambling behaviour in the United States and Canada. Canadian journal of public health Revue canadienne de sante publique 92: 168-72. Shaffer HJ, Korn DA (2002) Gambling and related mental disorders: a public health analysis. Annual review of public health 23: 171-212. Shand DG (1976) Pharmacokinetics of propranolol: a review. Postgrad Med J 52 Suppl 4: 22-25. Shao R, Read J, Behrens TEJ, Rogers RD (2013) Shifts in reinforcement signalling while playing slot-machines as a function of prior experience and impulsivity (vol 3, e235, 2013). Translational psychiatry 3. Shenhav A, Straccia MA, Cohen JD, Botvinick MM (2014) Anterior cingulate engagement in a foraging context reflects choice difficulty, not foraging value. Nat Neurosci 17: 1249-54. Shi CJ, Cassell MD (1998) Cortical, thalamic, and amygdaloid connections of the anterior and posterior insular cortices. The Journal of comparative neurology 399: 440-68. Silva AJ, Kogan JH, Frankland PW, Kida S (1998) CREB and memory. Annu Rev Neurosci 21: 127-48. Silveira MM, Murch WS, Clark L, Winstanley CA (2016) Chronic atomoxetine treatment during adolescence does not influence decision-making on a rodent gambling task, but does modulate amphetamine's effect on impulsive action in adulthood. Behav Pharmacol 27: 350-63. Singer T, Critchley HD, Preuschoff K (2009) A common role of insula in feelings, empathy and uncertainty. Trends in cognitive sciences 13: 334-40. Smith DF, Amdisen A (1983) Central effects of lithium in rats: lithium levels, body weight and water intake. Acta Pharmacol Toxicol (Copenh) 52: 81-5. St Onge JR, Floresco SB (2009) Dopaminergic modulation of risk-based decision making. Neuropsychopharmacology 34: 681-97. St Onge JR, Floresco SB (2010) Prefrontal cortical contribution to risk-based decision making. Cereb Cortex 20: 1816-28. St. Onge JR, Floresco SB (2009) Dopaminergic regulation of risk-based decision-making. Neuropsychopharmacology 34: 681-97. Steenbergh TA, Meyers AW, May RK, Whelan JP (2002) Development and validation of the Gamblers' Beliefs Questionnaire. Psychology of addictive behaviors : journal of the Society of Psychologists in Addictive Behaviors 16: 143-9. Sterzer P, Kleinschmidt A (2010) Anterior insula activations in perceptual paradigms: often observed but barely understood. Brain structure & function 214: 611-22. Stewart AO, Cowart MD, Moreland RB, Latshaw SP, Matulenko MA, Bhatia PA, Wang X, Daanen JF, Nelson SL, Terranova MA, Namovic MT, Donnelly-Roberts DL, Miller LN, Nakane M, Sullivan JP, Brioni JD (2004) Dopamine D4 ligands and models of receptor activation: 2-(4-pyridin-2-ylpiperazin-1-ylmethyl)-1H-benzimidazole and related heteroarylmethylarylpiperazines exhibit a substituent effect responsible for additional efficacy tuning. Journal of medicinal chemistry 47: 2348-55. Sulzer D, Sonders MS, Poulsen NW, Galli A (2005) Mechanisms of neurotransmitter release by amphetamines: a review. Prog Neurobiol 75: 406-33. Sun H, Cocker PJ, Zeeb FD, Winstanley CA (2012) Chronic atomoxetine treatment during adolescence decreases impulsive choice, but not impulsive action, in adult rats and alters 212  markers of synaptic plasticity in the orbitofrontal cortex. Psychopharmacology (Berl) 219: 285-301. Sun H, Green TA, Theobald DE, Birnbaum SG, Graham DL, Zeeb FD, Nestler EJ, Winstanley CA (2010) Yohimbine increases impulsivity through activation of cAMP response element binding in the orbitofrontal cortex. Biological psychiatry 67: 649-56. Sunahara RK, Guan HC, O'Dowd BF, Seeman P, Laurier LG, Ng G, George SR, Torchia J, Van Tol HH, Niznik HB (1991) Cloning of the gene for a human dopamine D5 receptor with higher affinity for dopamine than D1. Nature 350: 614-9. Sutton MA, Rolfe NG, Beninger RJ (2001) Biphasic effects of 7-OH-DPAT on the acquisition of responding for conditioned reward in rats. Pharmacol Biochem Behav 69: 195-200. Sylvain C, Ladouceur R, Boisvert JM (1997) Cognitive and behavioral treatment of pathological gambling: a controlled study. Journal of consulting and clinical psychology 65: 727-32. Szechtman H, Sulis W, Eilam D (1998) Quinpirole induces compulsive checking behavior in rats: a potential animal model of obsessive-compulsive disorder (OCD). Behav Neurosci 112: 1475-85. Tang L, Todd RD, Heller A, O'Malley KL (1994) Pharmacological and functional characterization of D2, D3 and D4 dopamine receptors in fibroblast and dopaminergic cell lines. J Pharmacol Exp Ther 268: 495-502. Tarazi FI, Baldessarini RJ (1999) Brain dopamine D(4) receptors: basic and clinical status. Int J Neuropsychopharmacol 2: 41-58. Taylor JR, Robbins TW (1984) Enhanced behavioural control by conditioned reinforcers following microinjections of d-amphetamine into the nucleus accumbens. Psychopharmacology (Berl) 84: 405-12. Taylor JR, Robbins TW (1986) 6-Hydroxydopamine lesions of the nucleus accumbens, but not of the caudate nucleus, attenuate enhanced responding with reward-related stimuli produced by intra-accumbens d-amphetamine. Psychopharmacology (Berl) 90: 390-7. Terrell D (1994) A Test of the Gamblers Fallacy - Evidence from Pari-Mutuel Games. J Risk Uncertainty 8: 309-317. Toneatto T, Blitz-Miller T, Calderwood K, Dragonetti R, Tsanos A (1997) Cognitive distortions in heavy gambling. Journal of gambling studies / co-sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming 13: 253-66. Tremblay AM, Desmond RC, Poulos CX, Zack M (2011) Haloperidol modifies instrumental aspects of slot machine gambling in pathological gamblers and healthy controls. Addict Biol 16: 467-84. Tremblay M, Silveira, M. M., Kaur, S., Hosking, J. G., Adams, W. K.,  Baunez, C., Winstanley, C. A. (2016) Increase in preference for uncertainty in rats following chronic dopamine D2/3 agonist ropinirole treatment for Parkinson’s Disease: Potential recruitment of the Akt/GSK3β signalling pathway. Society for neuroscience annual meeting abstract (submitted). Tunstall BJ, Kearns DN (2015) Sign-tracking predicts increased choice of cocaine over food in rats. Behav Brain Res 281: 222-8. Tuross N, Patrick RL (1986) Effects of propranolol on catecholamine synthesis and uptake in the central nervous system of the rat. J Pharmacol Exp Ther 237: 739-45. 213  Tye KM, Deisseroth K (2012) Optogenetic investigation of neural circuits underlying brain disease in animal models. Nature reviews Neuroscience 13: 251-66. Van Craenenbroeck K, Rondou P, Haegeman G (2010) The dopamine D4 receptor: biochemical and signalling properties. Cell Mol Life Sci 67: 1971-1986. van den Brink W (2012) Evidence-based pharmacological treatment of substance use disorders and pathological gambling. Curr Drug Abuse Rev 5: 3-31. van Dyck CH, Seibyl JP, Malison RT, Laruelle M, Zoghbi SS, Baldwin RM, Innis RB (2002) Age-related decline in dopamine transporters: analysis of striatal subregions, nonlinear effects, and hemispheric asymmetries. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry 10: 36-43. van Gaalen MM, Brueggeman RJ, Bronius PF, Schoffelmeer AN, Vanderschuren LJ (2006a) Behavioral disinhibition requires dopamine receptor activation. Psychopharmacology 187: 73-85. van Gaalen MM, van Koten R, Schoffelmeer AN, Vanderschuren LJ (2006b) Critical involvement of dopaminergic neurotransmission in impulsive decision making. Biological psychiatry 60: 66-73. van Holst RJ, Chase HW, Clark L (2014) Striatal connectivity changes following gambling wins and near-misses: Associations with gambling severity. NeuroImage Clinical 5: 232-9. van Holst RJ, Veltman DJ, Buchel C, van den Brink W, Goudriaan AE (2012) Distorted expectancy coding in problem gambling: is the addictive in the anticipation? Biol Psychiatry 71: 741-8. Van Tol HH, Bunzow JR, Guan HC, Sunahara RK, Seeman P, Niznik HB, Civelli O (1991) Cloning of the gene for a human dopamine D4 receptor with high affinity for the antipsychotic clozapine. Nature 350: 610-4. Verdejo-Garcia A, Lawrence AJ, Clark L (2008) Impulsivity as a vulnerability marker for substance-use disorders: Review of findings from high-risk research, problem gamblers and genetic association studies. Neuroscience and biobehavioral reviews 32: 777-810. Vincent SL, Khan Y, Benes FM (1993) Cellular distribution of dopamine D1 and D2 receptors in rat medial prefrontal cortex. J Neurosci 13: 2551-64. Vogt BA, Hof PR, Zilles K, Vogt LJ, Herold C, Palomero-Gallagher N (2013) Cingulate area 32 homologies in mouse, rat, macaque and human: cytoarchitecture and receptor architecture. The Journal of comparative neurology 521: 4189-204. Volkow ND (2005) What do we know about drug addiction? The American journal of psychiatry 162: 1401-2. Volkow ND, Fowler JS, Wang GJ, Swanson JM (2004) Dopamine in drug abuse and addiction: results from imaging studies and treatment implications. Mol Psychiatry 9: 557-69. Volkow ND, Fowler JS, Wang GJ, Swanson JM, Telang F (2007) Dopamine in drug abuse and addiction: results of imaging studies and treatment implications. Arch Neurol 64: 1575-9. Voon V, Fernagut PO, Wickens J, Baunez C, Rodriguez M, Pavon N, Juncos JL, Obeso JA, Bezard E (2009) Chronic dopaminergic stimulation in Parkinson's disease: from dyskinesias to impulse control disorders. Lancet Neurol 8: 1140-9. Voon V, Potenza MN, Thomsen T (2007a) Medication-related impulse control and repetitive behaviors in Parkinson's disease. Current opinion in neurology 20: 484-92. 214  Voon V, Thomsen T, Miyasaki JM, de Souza M, Shafro A, Fox SH, Duff-Canning S, Lang AE, Zurowski M (2007b) Factors associated with dopaminergic drug-related pathological gambling in Parkinson disease. Arch Neurol 64: 212-6. Wager TD, Phan KL, Liberzon I, Taylor SF (2003) Valence, gender, and lateralization of functional brain anatomy in emotion: a meta-analysis of findings from neuroimaging. NeuroImage 19: 513-531. Walker MB (1992) Irrational thinking among slot machine players. Journal of Gambling Studies 8. Wardle. H MA, Spence. S, Orford. J, Volberg. R, Jotangia. D, Griffths. M, Hussey. D & Dobbie. F.  (2010) British Gambling Prevalence Survey 2010. The Gambling Commission. Wedzony K, Chocyk A, Mackowiak M, Fijal K, Czyrak A (2000) Cortical localization of dopamine D4 receptors in the rat brain--immunocytochemical study. Journal of physiology and pharmacology : an official journal of the Polish Physiological Society 51: 205-21. Weintraub D, David AS, Evans AH, Grant JE, Stacy M (2015) Clinical spectrum of impulse control disorders in Parkinson's disease. Mov Disord 30: 121-7. Weintraub D, Potenza MN (2006) Impulse control disorders in Parkinson's disease. Current neurology and neuroscience reports 6: 302-6. Weller JA, Levin IP, Shiv B, Bechara A (2009) The effects of insula damage on decision-making for risky gains and losses. Social neuroscience 4: 347-58. Welte J, Barnes G, Wieczorek W, Tidwell MC, Parker J (2001) Alcohol and gambling pathology among US adults: Prevalence, demographic patterns and comorbidity. J Stud Alcohol 62: 706-712. Westlund KN, Carlton SM, Zhang D, Willis WD (1990) Direct catecholaminergic innervation of primate spinothalamic tract neurons. The Journal of comparative neurology 299: 178-86. Winstanley CA (2007) The orbitofrontal cortex, impulsivity, and addiction: probing orbitofrontal dysfunction at the neural, neurochemical, and molecular level. Annals of the New York Academy of Sciences 1121: 639-55. Winstanley CA (2011) The utility of rat models of impulsivity in developing pharmacotherapies for impulse control disorders. Br J Pharmacol 164: 1301-21. Winstanley CA, Cocker PJ, Rogers RD (2011) Dopamine modulates reward expectancy during performance of a slot machine task in rats: evidence for a 'near-miss' effect. Neuropsychopharmacology 36: 913-25. Winstanley CA, Eagle DM, Robbins TW (2006) Behavioral models of impulsivity in relation to ADHD: translation between clinical and preclinical studies. Clin Psychol Rev 26: 379-95. Winstanley CA, Theobald DE, Cardinal RN, Robbins TW (2004a) Contrasting roles of basolateral amygdala and orbitofrontal cortex in impulsive choice. The Journal of neuroscience : the official journal of the Society for Neuroscience 24: 4718-22. Winstanley CA, Theobald DE, Dalley JW, Glennon JC, Robbins TW (2004b) 5-HT2A and 5-HT2C receptor antagonists have opposing effects on a measure of impulsivity: interactions with global 5-HT depletion. Psychopharmacology 176: 376-85. Wolterink G, Phillips G, Cador M, Donselaar-Wolterink I, Robbins TW, Everitt BJ (1993) Relative roles of ventral striatal D1 and D2 dopamine receptors in responding with conditioned reinforcement. Psychopharmacology 110: 355-64. 215  Wong DF, Wagner HN, Jr., Dannals RF, Links JM, Frost JJ, Ravert HT, Wilson AA, Rosenbaum AE, Gjedde A, Douglass KH, et al. (1984) Effects of age on dopamine and serotonin receptors measured by positron tomography in the living human brain. Science 226: 1393-6. Worhunsky PD, Malison RT, Rogers RD, Potenza MN (2014) Altered neural correlates of reward and loss processing during simulated slot-machine fMRI in pathological gambling and cocaine dependence. Drug and alcohol dependence 145: 77-86. Xi ZX, Gilbert J, Campos AC, Kline N, Ashby CR, Jr., Hagan JJ, Heidbreder CA, Gardner EL (2004) Blockade of mesolimbic dopamine D3 receptors inhibits stress-induced reinstatement of cocaine-seeking in rats. Psychopharmacology 176: 57-65. Yan Y, Pushparaj A, Le Strat Y, Gamaleddin I, Barnes C, Justinova Z, Goldberg SR, Le Foll B (2012) Blockade of dopamine d4 receptors attenuates reinstatement of extinguished nicotine-seeking behavior in rats. Neuropsychopharmacology 37: 685-96. Yaxley S, Rolls ET, Sienkiewicz ZJ (1988) The responsiveness of neurons in the insular gustatory cortex of the macaque monkey is independent of hunger. Physiology & behavior 42: 223-9. Young JW, Geyer MA, Rissling AJ, Sharp RF, Eyler LT, Asgaard GL, Light GA (2013) Reverse translation of the rodent 5C-CPT reveals that the impaired attention of people with schizophrenia is similar to scopolamine-induced deficits in mice. Translational psychiatry 3: e324. Yuen EY, Yan Z (2009) Dopamine D4 receptors regulate AMPA receptor trafficking and glutamatergic transmission in GABAergic interneurons of prefrontal cortex. J Neurosci 29: 550-62. Yuen EY, Zhong P, Yan Z (2010) Homeostatic regulation of glutamatergic transmission by dopamine D4 receptors. Proceedings of the National Academy of Sciences of the United States of America 107: 22308-13. Zack M, Poulos CX (2007) A D2 antagonist enhances the rewarding and priming effects of a gambling episode in pathological gamblers. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology 32: 1678-86. Zeeb FD, Baarendse PJ, Vanderschuren LJ, Winstanley CA (2015) Inactivation of the prelimbic or infralimbic cortex impairs decision-making in the rat gambling task. Psychopharmacology (Berl) 232: 4481-91. Zeeb FD, Floresco SB, Winstanley CA (2010) Contributions of the orbitofrontal cortex to impulsive choice: interactions with basal levels of impulsivity, dopamine signalling, and reward-related cues. Psychopharmacology 211: 87-98. Zeeb FD, Robbins TW, Winstanley CA (2009) Serotonergic and dopaminergic modulation of gambling behavior as assessed using a novel rat gambling task. Neuropsychopharmacology 34: 2329-43. Zeeb FD, Soko AD, Ji X, Fletcher PJ (2016) Low Impulsive Action, but not Impulsive Choice, Predicts Greater Conditioned Reinforcer Salience and Augmented Nucleus Accumbens Dopamine Release. Neuropsychopharmacology. Zeeb FD, Winstanley CA (2011) Lesions of the basolateral amygdala and orbitofrontal cortex differentially affect acquisition and performance of a rodent gambling task. J Neurosci 31: 2197-204. 216  Zeeb FD, Winstanley CA (2013) Functional disconnection of the orbitofrontal cortex and basolateral amygdala impairs acquisition of a rat gambling task and disrupts animals' ability to alter decision-making behavior after reinforcer devaluation. J Neurosci 33: 6434-43. Zeeb FD, Wong AC, Winstanley CA (2013) Differential effects of environmental enrichment, social-housing, and isolation-rearing on a rat gambling task: dissociations between impulsive action and risky decision-making. Psychopharmacology (Berl) 225: 381-95. Zweifel LS, Fadok JP, Argilli E, Garelick MG, Jones GL, Dickerson TMK, Allen JM, Mizumori SJY, Bonci A, Palmiter RD (2011) Activation of dopamine neurons is critical for aversive conditioning and prevention of generalized anxiety. Nat Neurosci 14: 620-U112.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0307411/manifest

Comment

Related Items