Learned associations between rewarding stimuli and environmental cues which predict their availability play an important role in guiding behaviour. These learned associations are thought to be encoded by neuroadaptations in disperse sets of strongly activated neurons, termed neuronal ensembles, located throughout motivationally-relevant brain areas. However to date, the nature of the adaptations which occur selectively on neuronal ensembles encoding appetitive associative memories remain largely unknown. Using the Fos-GFP mouse, which expresses green fluorescent protein (GFP) in recently activated neurons, we investigated the intrinsic and synaptic excitability of neurons activated following exposure to stimuli associated with food (sucrose) or drug (cocaine) exposure. We observed that in the nucleus accumbens (NAc) shell, but not orbitofrontal cortex, neurons activated following exposure to a food-associated stimulus were more intrinsically excitable than surrounding, non-activated neurons. These neurons also demonstrated increased spontaneous excitatory transmission suggestive of potentiated synaptic strength. Following extinction of the food-cue association, NAc shell neurons activated following cue exposure were no longer more excitable than surrounding neurons. This suggests that the intrinsic excitability of striatal neurons activated by a food-associated cue is dynamically modulated by changes in associative strength. We also examined the intrinsic excitability of striatal neurons (including neurons in the NAc shell, core and dorsal striatum) activated by cocaine-associated stimuli. Interestingly, NAc shell neurons activated by cocaine-associated stimuli were not more excitable compared to the surrounding neurons regardless of extinction learning experience, possibly indicating differences between drug and food conditioning. Similar results were obtained for dorsal striatal neurons. However, NAc core neurons activated by cocaine-associated stimuli displayed an enhanced excitability which persisted following extinction, indicating that core and shell neuronal ensembles differentially encode the cocaine associative memories. Overall, by selectively recording from stimuli-activated neurons, this work reveals novel adaptations at the intrinsic and synaptic levels on neuronal ensembles following appetitive learning with both food and drug rewards.

Previous research has implicated the role of food associated cues or pre-task reward exposure with eating behaviour. Eating behaviour (specifically overeating) has itself been associated with subtypes of impulsivity. To date, no research has examined the direct relationship between (food) reward-associated cues, or rewarding food exposure on behavioural impulsivity – a possible underlying mechanism. This thesis aimed to examine how behavioural impulsivity may be modulated by external cues, or by hedonic reward consumption, and how this interacts with eating attitudes (TFEQ). Experiment 1 examined the aims explicitly, giving participants a hedonic preload (or nothing) before they completed impulsivity tasks. Those who received a preload were more impulsive in terms of their impulsive choice, and inhibitory control than those who had not received a preload. This effect did not replicate in experiment 3, where 2 further conditions were added, a non-hedonic preload, and an anticipation condition, but no differences were found between the groups. Experiment 4 conditioned rewarding cues to novel stimuli, and presented them before the behavioural tasks, but again, no difference was found between the groups. This thesis discusses the theoretical and methodological concepts, which may explain some of these null findings. Experiment 2 aimed to examine how the reinforcing value of food (RRV) may be associated with types of impulsivity. However, no relationship was found between RRV and impulsivity, but RRV was consistent in predicting ad libitum food intake, as shown in previous studies. Chapter 6 of this thesis is a meta-analysis of our laboratory's research linking delayed discounting (DDT), the TFEQ, and cue exposure paradigms. The analysis showed that those in high in dietary disinhibition (TFEQ-D) who were shown food cues, or consumed a hedonic preload were more impulsive on the DDT than those high in TFEQ-D that did not consume anything. The key limitations of this thesis are discussed, most notably the lack of statistical power in the experimental studies conducted. The general discussion of this thesis discusses the important implications of this finding in understanding modulation of behavioural impulsivity.

Investigators have suggested impulsivity consists of several behavioural subtypes including ‘reflection'- (decision-making without evaluation of information), ‘temporal'- (failure to delay gratification) and ‘motor'- (failure to inhibit a motor response) impulsivity. These facets of impulsivity are thought to be dissociable, but to share some common underlying processes. The current studies investigated such processes. Study 1 investigated speed and accuracy biases, using instructions and cognitive priming to challenge impulsivity. Study 2 & 3 challenged inhibitory control resources, via a dual task and alcohol challenge, to investigate the effect on impulsivity. Study 3 also investigated the effect of alcohol outcome expectancies on impulsivity. The factor structure of impulsivity was also investigated using exploratory factor analysis (study 4), to establish whether the primary measures of the proposed subtypes can indeed be categorised into these three factors. Study 4 also investigated the relationship of participant demographics to impulsivity. The studies support the suggestion of a distinct subtype of reflection-impulsivity. Inhibitory control processes do not appear to underlie performance, however biases in speed/accuracy trade-offs have implications for this subtype. Behavioural inhibitory control was found to be the primary process underlying motor-impulsivity whilst biases for speed/accuracy have implications for Go-responses. The factor analysis provided preliminary evidence that there may be two distinct facets of motor-impulsivity: action cancellation and action restraint. Inhibitory control processes were not found to underlie temporal-impulsivity on an experiential task. Biases for speed/accuracy were found to contribute to performance on pen-and-paper measures. However, subsequent factor analysis provided evidence that experiential tasks may actually be more closely related to a form of cognitive control, instead of temporal-impulsivity. In conclusion, the studies found that the three proposed factors of impulsivity differentially rely on inhibitory control processes and biases for speed/accuracy. However, factor analysis indicated that additional factors may be required to fully characterise impulsivity.

This thesis focuses on how recalling past behaviour affects attitudes, intentions and behaviour in the domain of moral decision-making. It extends the existing literature on moral licensing and moral cleansing by exploring whether different individual difference variables moderate such licensing and cleansing effects. Five empirical studies are reported. Chapter 1 provides an overview of the moral licensing and moral cleansing literature and also outlines research into the predictive effects of moral norms on behavioural intentions within the structure of the TPB. In chapter 2, the first study reveals evidence of moral cleansing effects: participants in an immoral condition were more likely to donate to charity than were participants in a moral condition or control group. Study 2 investigated whether self monitoring moderated licensing and cleansing effects. Chapter 3 (study 3) investigated the mediation effect of emotion and the moderation effect of moral identity regarding licensing and cleansing effects on attitudes, intentions and moral norm and behaviour. A partial mediation of condition and behaviour by negative emotions was identified. The results also indicated evidence of a cleansing effect. Moreover, in studies 1, 2 and 3 mediation of the moral norms – intention relationship via attitudes was examined. Moral norms were identified as a strong predictor of charitable donation intentions. In Chapter 4, environmental attitude was investigated as a moderator of the effect of individuals' past pro-environmental behaviour on TPB components. Internally motivated pro-environmental attitude was found to be a significant moderator. Chapter 5 draws upon the idea that conception of morality differs in different cultures and examined how different moral foundation values and cultural orientations affect moral attitudes and intentions in the UK and Iran. Surprisingly, moral norms were a more useful predictor of intention than were attitudes in both national cultures.

Previous research employing modulation of 5-HT2C receptors (5-HT2CRs) in rodents has identified a potential role in mediating cognitive flexibility. The work presented in this thesis explores the effects of systemic administration of the selective 5-HT2CR antagonist SB242084 on a range of Pavlovian and operant learning paradigms used to model cognitive flexibility and reward-based learning in mice. Based on a key design difference in previous research reporting discrepant outcomes, trial initiation requirements were manipulated. However the effect of SB242084 administration relative to vehicle-treatment was consistent with prior reports of impaired reversal performance following reductions in 5-HT2CR activity, regardless of whether trials were automatically or self-initiated. In contrast, performance on a probabilistic reversal learning task was enhanced by drug-treatment, raising the possibility that task difficulty mediates the effect of this manipulation on performance. A drug-related enhancement in the ability to overcome learned non-reward at the previously incorrect location was additionally demonstrated under probabilistic reversal conditions, with no effect on perseverance at the previously correct location. However, performance of drug-treated animals in two closely related tasks demonstrated impaired extinction learning but intact development of latent inhibition to a pre-exposed stimulus. The effect of SB242084 on incentive motivation was additionally explored, but did not impact upon the acquisition of a sign-tracking response to a conditioned stimulus, or a subsequent reversal; suggesting that 5-HT2CRs may be more critically involved in instrumental than Pavlovian learning. These experiments reveal a complex picture for the involvement of 5-HT2CRs in flexible cognition, however, systemic manipulations may not be optimal for dissecting their role. Therefore, a final study explored the expression of c-Fos immunoreactivity in response to reversal learning. A broad network was activated by elements of the reversal task, including regions of the prefrontal cortex and amygdala, providing a basis for future studies targeting components of this circuitry.

A better understanding of the psychological mechanisms underpinning addiction will facilitate its remediation. Some evidence suggests that the emotional properties of drug-paired stimuli themselves drive drug-procurement, while other evidence indicates that the expectation of reward elicited by the stimuli is sufficient to control drug-seeking. The current series of experiments aimed to explicate these seemingly contradictory data, by characterising the roles played in reward seeking by conditioned-stimulus-elicited emotion and expectation in non-dependent samples, before assessing their contribution in smokers. Further data suggest a role of personality in addictive behaviours, thus personality was assessed as a moderator of reward-seeking. Variations of a Pavlovian-to-instrumental transfer design, which tests the ability of reward-associated stimuli to modulate reward seeking, together with questionnaires of personality were applied. It was shown that outcome expectation was consistently necessary for cue-potentiated monetary-reward seeking, and similarly in smokers, cigarette outcome expectation was sufficient for cue-potentiated cigarette-reward seeking. Tentative evidence for the role of conditioned-stimulus emotional value in monetary-reward seeking was found, although this latter result requires scrutiny through additional research. Moderating influences of Extraversion and Neuroticism were found for cue-elicited emotion and outcome expectation, respectively. It is therefore proposed that reward expectancy is necessary for conditioned stimuli to control behaviour. The emotional properties of reward-predictive stimuli may be important for reward seeking in the absence of addiction, but when addiction to reward is present, control of reward seeking can occur via reward expectation only. Data from the role of personality, in moderating the effects of stimulus-elicited emotion or outcome expectation on reward-seeking behaviour, suggest that the control of behaviour by emotion may be facilitated by Extraversion, due to its propensity towards emotional processes, whereas control by expectation may be facilitated by Neuroticism, due to its inclination towards predictive learning.

Within classical economics, an agent is deemed “rational” if their preferences are both consistent and maximize utility of positive, subjective experience. Violations from this norm may occur as a result of utility devaluation (proceeding from risk or delay) or when an imbalance occurs between ‘liking' and ‘wanting'. The current studies investigate how changes in reward-contingencies, delay before reward receipt, and reinforcer devaluation contribute to such utility-based decision-making in human and rodent models. I examined the effects of devaluation through pre-exposure and outcome-contingency on the development and maintenance of sign- and goal-tracking responses in rats (chapter three) and humans (chapter four). Chapter four presents a novel, translational, eye-tracking procedure and correlates of such behaviour, including trait-impulsivity and discounting performance. In chapter five I presented probability- and temporal discounting using a mouse model, exploring the consequences of satiety-based devaluation, and changes in outcome-contingencies- and delays. In Chapter six I presented the same factors in a human sample using a novel laboratory-based procedure and sought to explore the impact of trait and state impulsivity and correlates of rapid discounting. Finally, in chapter seven I considered the relationship between trait impulsivity, alcohol use, smoking and discounting behaviour using a more conventional online questionnaire. The results represent some of the first to present probability discounting using a mouse model and characterisation of reliable sign- and goal-tracking performance in humans. The results of the latter studies broadly support previous findings in rats showing that lowered reward contingency diminishes goal-oriented, but enhances sign-oriented, responding. Chapter six establishes a human equivalent to rodent discounting paradigms through implicit learning that will allow future controlled studies in humans. Finally, chapter seven presents evidence for an association between delay discounting and trait impulsivity. The results of these studies support the assertion that irrational decision-making arises, in part, from changes in reward utility as a function of delay, probability, devaluation and individual differences. Furthermore the translational-homologous models presented allow for future biopsychological research into mechanisms underlying such behaviours.

The 21st century has seen a dramatic rise in Internet access and connectivity across the world. To date, only a small amount of research has been published on the subject of culture and Internet usage. This thesis investigates whether, and how, individuals from two different cultures (Saudi Arabia and the UK) engage with online social networks (OSNs) differently, and what might be the underlying psychological factors explaining such differences. A first qualitative study used interpretative phenomenological analysis (IPA; Smith, Flowers, & Larkin, 2009) to investigate motivations for using OSNs among Saudi and British participants. Both groups reported that they used OSNs to present a positive self-image, while desiring to maintain a sense of their ‘genuine' self in online interactions. For Saudi participants, OSNs also provided opportunities for selfexpression that were otherwise unavailable. British participants reported using OSNs for relationship maintenance. A second qualitative study also looked at motivations, but narrowed the focus to identity motives, applying motivated identity construction theory (Vignoles, 2011) to a thematic analysis of tweets written by citizens of Saudi Arabia and the United Kingdom. Motives for meaning, belonging, distinctiveness, continuity, efficacy, and self-esteem were all detectable in the tweets of both Saudi and British users. The manner in which these motives were pursued varied according to the cultural context of users within the affordances of the online context in which they were communicating. The research project then aimed to establish a way of measuring differences in online self-presentation strategies, by developing the online self-presentation strategies scale (OSPSS). Items were selected using exploratory structural equation modelling (ESEM). The scale was incorporated in a large-scale (N = 694) quantitative study of Saudi and British OSN users that measured self-presentation strategies, motivations of OSNs use and target audience. Mediation analyses were conducted to find out whether cultural differences in these dimensions were explained by two forms of cultural variation: relational mobility and Schwartz' theory of basic values. Self-enhancement vs. self-transcendence values and relational mobility, more than openness to change vs. conservation values, accounted for mean differences between the groups in motives, targeted audiences and self-presentation strategies. Together the studies reveal observable differences in the ways in which people from Saudi Arabia and the UK engage with OSNs. These are partially explained by the affordances that social media provide, which compensate for the unavailability of certain modes of expression and communication within offline cultural contexts, and by cultural differences in value priorities.

This thesis is the first study of the crowd psychology of the annual Hajj, or pilgrimage to Mecca (Makkah) in Saudi Arabia, to employ self-categorization theory (SCT). The thesis aims to document and understand the perspective of pilgrims from a social psychological point of view, since no one has done that before, as well as to understand the perceptions of the Hajj management. Specifically, the thesis focuses on crowd perceptions, feelings of safety and the reasons for these feelings, and relations between subgroups in the crowd and between pilgrims and management. A literature review in Chapter two highlights the history and culture of the Hajj and the issues in managing the Hajj. Academic perspectives on crowd psychology are discussed in chapter three. Chapters four and five present respectively a UK pilot study of pilgrims and a field pilot study of pilgrims and management. Chapter six (the main interview study with pilgrims) indicates that despite the inconveniences, participants felt safe, secure and wellbeing inside the Grand Mosque during Hajj. Chapter seven (the main interview study with Hajj management) explores the participants' understanding of crowd behaviour, crowd psychology and its relation to safety, danger and their own role. In Chapter eight (the major study of the thesis), a survey of 1194 pilgrims at the Hajj found that identification with the crowd predicted enjoyment of the crowd. Also, for those high in identification with the crowd, crowd density increased perceptions of safety. Perceived support was found to mediate these positive effects of social identity on feeling safe. Chapter nine critically explores the findings of the thesis and discusses them in relation to relevant literature. It also reflects on the implications of the study for the theory of crowd psychology, and considers what lessons there might be for the management of the Hajj. This chapter concludes the thesis and outlines suggestions for further research.

Planning for incidents involving mass decontamination has focused almost exclusively on technical aspects of decontamination, with little attempt to understand public experiences and behaviour. This thesis aimed to examine relevant theory and research, in order to understand public behaviour during incidents involving mass decontamination, and to develop theoretically-derived recommendations for emergency responders. As these incidents involve groups, it was expected that social identity processes would play an important role in public responses. A review of small-scale incidents involving decontamination is presented, along with a review of decontamination guidance documents for emergency responders. This literature shows that responder communication strategies play an important role in public experiences and behaviour, but that the importance of communication is not reflected in guidance documents. Theories of mass emergency behaviour, in particular the social identity approach, are reviewed, in order to generate hypotheses and recommendations for the management of incidents involving mass decontamination. It is hypothesised that effective responder communication will increase public compliance and cooperation, and reduce anxiety, mediated by social identity variables (e.g. perceptions of responder legitimacy, identification with emergency responders and other members of public, and collective agency). The empirical research presented in this thesis tests the hypotheses and recommendations derived from the social identity approach. The research includes: a responder interview study; three studies of volunteer feedback from field exercises; a visualisation experiment; and a mass decontamination field experiment. Findings show that effective responder communication consistently results in increased willingness to comply with decontamination, and increased public cooperation; this relationship is mediated by social identity variables. Results support the hypotheses, and show that an understanding of the social identity approach facilitates the development of effective responder communication strategies for incidents involving mass decontamination. Four theoretically-derived, and evidence-based, recommendations for emergency responders are generated as a result of this thesis.