Search

This thesis investigates how identity styles operate in a non-Western cultural context. Through four papers, it both tests and extends some of the theoretical assertions made by the social constructivist approach of identity styles (Berzonsky, 2011), which proposes that people construct both ‘who they think they are' and ‘the reality in which they live', through informational, normative or diffuse-avoidant identity orientations. Although the relationship among identity styles and well-being is well established in Western literature, there is a serious dearth of similar research in non-Western cultures such as Pakistan. Western theories tend to assume the universal generalisability of identity styles and their relationship with well-being. The primary aim of this research is to test the validity of this assumption in the cultural context of Pakistan. Paper 1 systematically examines the factorial structure of the Identity Styles Inventory (ISI-5) in a Pakistani sample. In confirmatory factor analysis, normative orientation items perform relatively poorly, leading to a possibility that the conception of normative orientation is not as universal as previously assumed. Paper 2 shows that well-being is predicted positively by information orientation and negatively by diffuse-avoidant orientation in the Pakistani sample. Normative orientation remained as non-significant predictor of well-being. Identity commitment and satisfaction of identity motives partially mediate these links. Paper 3 explores the indigenous processes of identity formation through qualitative semi-structured interviews. Normative orientation is found to operate at a much more complex level than assumed previously. Participants described many different ways of being normative, making this a less automatic, mindless and effortless process in Pakistani culture than assumed previously based on Western research models. Paper 4 focuses on the construction and psychometric testing of new measures of normative orientation suitable for use with the samples from Pakistan. Together, these studies illustrate the value of using indigenous perspectives to enrich Western-based understandings of identity formation.

The notion that unconscious Bayesian inference underlies perception is gaining ground. Predictive coding approaches posit that the state of the world is inferred by integrating, at each level of the perceptual hierarchy, top-down prior beliefs about sensory causes and bottom-up prediction errors. In this framework, percepts correspond to a top-down stream of beliefs that best 'explain away' sensory signals. Although such frameworks are gathering empirical support, subjective facets of perception remain unexplained from these perspectives. This thesis combines behavioural, computational and neuroimaging methods to examine how subjective visual confidence can be accounted for in a predictive coding framework. Experiment one shows that, behaviourally, perceptual expectations about target presence or absence both liberalise confidence thresholds and increase metacognitive accuracy. These results are modelled in a signal detectiontheoretic framework as low-level priors shifting the posterior odds of being correct. Using EEG, experiment two reveals that influence of expectations on decision and confidence oscillates with the phase of pre-stimulus alpha oscillations. This means that prior to target onset, both objective and subjective decisions have been rhythmically biased by the periodic recruitment of expectations to visual areas. Using fMRI, experiment three shows that in the post-stimulus period, expectations and sensory signals are integrated into confidence judgements in right inferior frontal gyrus (rIFG). Furthermore, this process recruits orbitofrontal cortex and bilateral frontal pole, which represent top-down influences, and occipital lobe, which represents bottom-up signals. Together, these results suggest that expectations shape subjective confidence by biasing the posterior probability of the perceptual belief.

Computerised cognitive behavioural therapy (CCBT) is a clinically effective method of delivering CBT which may help address the under – treatment of common mental health disorders (CMHDs) in the population. However, concerns regarding acceptability, attrition rates and the therapeutic alliance are obstacles to widespread population dissemination. This thesis aimed to address these implementation issues by applying concepts from human – computer interaction (HCI) and attachment theory to the field of CCBT. Chapter 1 presents a meta – analysis investigating the effectiveness of CCBT for CMHDs and moderators of this effect. Chapter 2 presents a systematic review and analysis conducted to examine predictors of CCBT engagement. A process – based model of engagement with CCBT developed from the findings of this review is also presented. Adult attachment is known to influence engagement and alliance in face to face therapies, but research has not explored whether these relationships are mirrored in CCBT. Four empirical studies intended to address this question. Study 1 used a student population based survey to explore the acceptability of CCBT in a student population and the associations with adult attachment. Results demonstrated adult attachment was not associated with acceptability of CCBT. Study 2a utilised an open trial of a supported CCBT program to investigate whether adult attachment would predict engagement and alliance in vivo. Results showed attachment did not predict these outcomes. Study 2b utilised an open trial with a non – supported online CCBT program. Results indicated attachment security was positively associated with program engagement and alliance. It is proposed a combination of attachment system activation and perceiving computers as social actors account for these findings. Study 3 used a randomised, experimental paradigm to test the benefits of security priming in CCBT. Security priming produced higher levels of program engagement and better working alliance compared to neutral primes. Furthermore these effects were not moderated by dispositional attachment styles. These results demonstrate something so uniquely human, dispositional attachment orientations, founded on the intimate bonds we form in infancy and in adulthood , extend their influence into the experience of unguided CCBT, a solely human – computer interaction. Unguided – CCBT, a highly cost effective intervention with the potential for considerable public health impact, may benefit from incorporating security priming techniques in program designs to maximise engagement and alliance. Engagement and alliance is attainable in CCBT and paying attention to the attachment styles of program users may present a distinctive opportunity to overcome these implementation barriers.

Age-related decline in cognitive performance and brain structure can be offset by increased exercise. Little is known, however, about the cognitive and brain structural consequences of long-term high-effort endurance exercise. In a cross-sectional design, we recruited older adults who had been engaging in high-effort endurance exercise over at least twenty years, and compared their cognitive performance and brain structure with a non-sedentary control group similar in age, sex, education, IQ, depression levels, and other lifestyle factors. We hypothesized that long-term high-effort endurance exercise would protect against the age-related decline in memory, attention, and brain structure. Our findings, in contrast to previous studies, indicated that those participating in long-term high-effort endurance exercise, when compared without confounds to non-sedentary control volunteers, showed no differences on measures of speed of processing, executive function, incidental memory, episodic memory, working memory, or visual search. On measures of prospective memory, long-term exercisers performance suggested a self-imposed increase in effort, which did not impact on ability to complete the PM task. In complex attention tasks, they displayed a differential strategy to controls. Structurally, long-term exercisers only displayed higher diffuse axial diffusivity, an index of axonal integrity, than controls, but this did not correlate with any cognitive differences.

Obsessive-Compulsive Disorder (OCD) is characterised by intrusions into conscious thinking by repetitive, personally abhorrent, absurd and alien thoughts (obsessions) which lead to endless repetitive acts or rehearsal of irrational and sometimes bizarre mental and behavioural rituals (compulsions). Although a number of clinical constructs have been proposed to have a causal influence on OC symptoms, extremely little research exists examining how these constructs are related to each other and, further, how the relationships between these constructs influences each constructs relationship with OC symptoms (e.g. does a given construct have an independent role in influencing OC symptoms or is the constructs influence on OC symptoms being mediated by its influence on another intervening construct or constructs?). In the current thesis, the relationships between 5 clinical constructs were examined, as well as how the pattern of these relationships may affect each constructs influence on OC symptoms. A large questionnaire study suggested that the five constructs are best seen as separate constructs rather than indicators of one underlying construct. Separate experimental studies in which all 5 constructs were manipulated individually suggest that whilst inflated responsibility, intolerance of uncertainty and negative mood, respectively, all causally influence every other construct focused on in this thesis, not just right experiences and as many as can stop rules, respectively, do not casually influence any other construct. Subsequent regression analyses suggest that whilst some constructs directly influence OC symptoms, other constructs influence on OC symptoms are mediated by intervening constructs in the final model, whilst some constructs have little to no influence on OC symptoms when the influence of other constructs in the final model are taken into account. Implications of these findings for existing models of OCD, for studies examining the relationship between multiple constructs and OC symptoms and for the treatment of OCD are discussed.

The following work presents an exploration of human search behaviour both from biological and computational perspectives. Search behaviour is defined as the movements made by an organism while attempting to find a resource. This work describes some of the principal procedures used to record movement, methods for analysing the data and possible ways of interpreting the data. In order to obtain a database of searching behaviour, an experimental setup was built and tested to generate the search paths of human participants. The test arena occupied part of a football field and the targets consisted of an array of 20 golf balls. In the first set of experiments, a random and regular distribution of targets were tested. For each distribution, three distinct conspicuity levels were constructed: a cryptic level, in which targets were painted the same colour as the grass, a semi-conspicuous level in which targets were left white and a conspicuous condition in which the position of each target was marked by a red flag, protruding one metre from the ground. The subjects tested were 9-11 year old children and their search paths were collected using a GPS device. Subjects did not recognise the spatial cues regarding the way targets were spatially distributed. A minimal decision model, the bouncing search model, was built based on the characteristics of the childrens search paths. The model produced an outstanding fit of the children's behavioural data. In the second set of experiments, a new group of children were tested for two new distributions obtained by arranging the targets in patches. Again, children appeared unable to recognise spatial information during the collection processes. The children's behaviour once again produced a good match with that of the bouncing search model. This work introduces several new methodological aspects to be explored to further understand the decision processes involved when humans search. Also, it illustrates that integrating biology and computational science can result in innovative research.

Using people actively involved in mountain climbing, this thesis explores people's motivations to participate in mountain climbing, an activity frequently characterised in terms of risk. Moreover, using a variety of both quantitative and qualitative methods the assessment of the role of risk as a motivation for mountain climbing is central to the thesis. The first study (N = 232) employed a theory of planned behaviour framework that incorporated beliefs about risk, together with other behavioural beliefs, as a means to investigate the motivations of mountain climbers. Although risk emerged as significant positive predictor of attitudes towards mountain climbing, it was the weakest of the four predictor variables. Study Two (N = 207) presents a psychometric analysis which mapped perceptions of eight types of climbing onto a three component (Challenge, Risk, and Enjoyment) representation of the characteristics associated with mountain climbing. The position of each type of climbing revealed some clear differences between these types in relation to each of the three dimensions. The results presented provide a useful insight into which particular types of climbing should be studied further to build upon the current understanding of the role and importance of risk to participation in mountain climbing. Study Three (N = 205) used a laddering methodology in order to identify the hierarchical relationship between motives reported by climbers who participate in three types of climbing. Individual cognitive maps were created for each type of mountain climbing. Inspection of both the cognitive maps and indices designed to reflect the importance of individual motives seem to suggest that the importance of risk to people's participation may be less than originally thought. Study Four (N = 37) was an on-line qualitative study which addressed mountain climbers‟ views concerning the popular yet controversial opinion that climbers are motivated by risk. Overall, risk appeared to acquire motivational status as a result of its instrumental relationship with other factors explicitly labelled as motivations for mountain climbing. Together, these findings suggest that, while risk occupies an important position within people's motivations to participate in mountain climbing, it is not risk per se that is key to people‟s participation. Moreover, the results presented hint at risk acting as a facilitator, something necessary to the fulfilment of other important motivations for mountain climbing.

Posttraumatic growth describes positive changes following challenging events. Although such changes are well documented there remain a number of important areas for further research, some of which are addressed in this thesis. In particular, this thesis aimed to clarify the relationship between growth and adjustment following health events, explore growth in different cultures (UK and Africa), and examine growth following childbirth using a prospective design. First, two systematic reviews were carried out to examine (i) growth following health events and (ii) maternal wellbeing in African women. The first meta-analytic review found that growth following cancer and HIV/AIDS was associated with higher levels of positive mental health, higher subjective physical health, and lower levels of negative mental health. Moderating variables were time since the event, age, ethnicity, and type of negative mental health outcome. The second review found that maternal psychological problems in African women have a similar or slightly higher prevalence than reported in developed countries. Risk factors were broadly comparable although some culture-specific factors were also found. Three research studies were conducted. The first study qualitatively explored 55 Gambian women's experiences of pregnancy, childbirth, and the postnatal period. Thematic analysis identified five themes: (1) transition to adulthood, (2) physical difficulties, (3) value of children in relation to others, (4) children as a strain, and (5) going through it alone. Prospective studies of growth following childbirth were then carried out in the UK (N=125) and The Gambia (N=101). Women completed questionnaires during their third trimester of pregnancy and up to 12 weeks after birth. A proportion of women in both countries reported growth following childbirth. In the UK, higher levels of growth were associated with caesarean sections and prenatal posttraumatic stress symptoms. In The Gambia, higher growth was associated with lower income, lower education, and higher postnatal social support.

This thesis in part addressed a call by Hodapp, Glidden & Kaiser (2005) to focus on identifying potential mediators and moderators of the relationship between growing up with a sibling with an autism spectrum condition (ASC) and adverse outcomes such as poorer sibling relationship quality (SRQ) and adjustment problems. It attempted to clarify previous inconsistent findings in the literature by considering autism as a dimensional disorder (with traits present on a continuum between the general population and those on the higher end reaching a clinical cut-off for diagnosis of ASC) and focusing on the autistic traits of (mainly) older siblings in a dyad. Specifically, these related to attention to detail, impaired mind reading ability, impaired social skills and impaired imagination. In addition, when looking at families with an ASC child, using a relatively homogenous sample group of typically developing younger siblings (YS) in middle childhood controlled for factors that potentially influenced past mixed findings. Hypotheses were extended to an analogous sample of young adults by asking dyads for both retrospective and current (adult) SRQ and self reported autistic traits. Themes emerging from the actual lived experiences of typically developing younger siblings of children with autism were also considered, and highlighted a number of difficulties faced by these siblings. The quantitative findings indicate that different elements of the social and communication domains relate to negativity compared to positivity in SRQ, and that there is a significant relationship between attention to detail (i.e. the non-social domain) and typical YS adjustment but not between YS adjustment and the social and communication domains. For young adults, reports of higher impairments in imagination were associated with retrospective reports of lower positivity in SRQ, and level of attention to detail was significantly associated with rivalry in adult SRQ. Overall these findings indicate that different autistic traits should be considered as separate influences on SRQ and adjustment.

Emotion is characterised by two-dimensions: emotional valence describes the extent to which an emotion is positive or negative, and arousal represents its intensity. Emotional content of verbal material affects cognitive processing, although research on word recognition has only recently taken emotion into account, primarily focusing on valence, while neglecting arousal. The present work aimed to disentangle the effects of valence and arousal during a lexical decision task, using reaction times (RTs), event-related potentials (ERPs) and BOLD responses in an event-related fMRI design. These methods were chosen to determine when affective features have an effect, and which neural systems are involved. The material for three experiments was based on a word corpus created by collecting ratings for emotional and lexico-semantic features. A first and novel finding was that arousal interacted with valence. Specifically, lexical decision times were slower for high-arousal positive stimuli (PH) and low-arousal negative ones (NL) compared to low-arousal positive (PL) and high arousal negative (NH) stimuli. ERPs also showed an interaction between 200-300 ms on the early posterior negativity (EPN), a component which is sensitive to emotional stimuli. At this processing stage people access their mental lexicon. Its amplitude was greater for PH and NL words, suggesting a higher processing load for conflicting stimuli. Positive valence and low arousal elicit an approach schema, whereas negative valence and high arousal elicit an avoidance schema (Robinson, Storbeck, Meier & Kirkeby, 2004). BOLD responses showed a similar interaction in the insula bilaterally, with increased activation for PH and NL words. This region is associated with integration of information on visceral states with higher-order cognitive and emotional processing, suggesting higher difficulty in integrating conflicting stimuli. Taken together, these studies indicate that emotion affects word processing during lexical access, and models of word recognition need to take into account both valence and arousal.

In the antisaccade task participants are required to saccade to the mirror image location of a sudden onset target. As such, the task provides a powerful tool with which to investigate the cognitive processes underlying goal-directed behaviour. In healthy participants antisaccade errors (prosaccades directed towards the target) occur on approximately 20% of trials, and increased antisaccade error rate is widely used as a measure of “cognitive disinhibition” in clinical settings. One aspect of antisaccade performance that has received relatively little attention is the large variability in error rate typically observed within healthy participants. Whilst there are many studies describing increased antisaccade error rates in patient populations, there has been comparatively little research into what individual differences might underlie the dramatic variations that are observed within healthy participants. This thesis presents five papers, each of which explores potential sources of variability in antisaccade performance in healthy participants. The first paper used a cueing manipulation to explore the extent to which individual differences in the ability to maintain the task goal in mind will influence antisaccade error rate. The second paper addressed the potential role of differences in motivation, by determining the extent to which antisaccade performance is moderated by a range of incentives. In paper 3, the role of strategic influences was investigated by altering the task instructions that participants were given. In paper 4, task instructions and working memory load were manipulated in order to determine their effect on antisaccade error awareness. The final paper, based on data gathered across the preceding experiments, explored the extent to which individual differences in factors such as working memory capacity, processing speed, and personality measures (schizotypy and impulsivity) correlated with antisaccade performance. Across all studies, the data is used to test predictions made by current parallel programming models of antisaccade performance. The data suggests that a range of “top-down” factors can influence antisaccade performance, but that the most important individual difference in explaining antisaccade error rate in healthy participants is prosaccade latency.

Contrasting approaches to explaining the social-cognitive contributors to bullying in schools have stressed the importance of a child‘s social goals in determining whether he or she will bully. In spite of this, the social goals of bullies and victims have not been adequately investigated in empirical research. This thesis aimed to address this issue by investigating the social goals associated with bullying/victimisation, determining whether these goals were able to predict bullying/victimisation even after other social processing biases and theory of mind had been taken into account, and considering the influence social goals have on children‘s response to provocation. In a series of six studies, 583 children from Primary schools in the UK completed several measures aimed at assessing their engagement in behaviours related to bullying and being victimised, their social goals (both as general interpersonal goals and also specific to hypothetical social scenarios), and other social-cognitive factors (including theory of mind). Although the pattern of results across studies was not always uniform, there was a general trend for bullying in boys to be associated with situation-specific goals that protected their physical dominance within their peer group, while bullying in girls was better predicted by an overall concern for maintaining an image of popularity. Interestingly, victimisation in boys was predicted by an inappropriate concern for others‘ feelings in certain scenarios, while victimisation in girls was associated with a low level of concern for behaving prosocially. Importantly, these kinds of social goals remained predictive of bullying and victimisation even after controlling for variance accounted for by theory of mind and other social information processing biases. Finally, social goals were found to mediate the relationship between bullying/victimisation and aggressive/submissive response strategies. Findings are discussed in relation to the existing literature as well as to their potential impact on intervention strategies.

This thesis breaks new ground by attending to two contemporary developments in art and science. In art, computer-mediated interactive artworks comprise creative engagement between collaborating practitioners and a creatively participating audience, erasing all notions of a dividing line between them. The procedural character of this type of communicative real-time interaction replaces the concept of a finished artwork with a ‘field of artistic communication'. In science, the field of creativity research investigates creative thought as mental operations that combine and reorganise extant knowledge structures. A recent paradigm shift in cognition research acknowledges that cognition is embodied. Neither embodiment in cognition nor the ‘field of artistic communication' in interactive art have been assimilated by creativity research. This thesis takes an interdisciplinary approach to examine the embodied cognitive processes in a ‘field of artistic communication' using a media artwork called Sim-Suite as a case study research strategy. This interactive installation, created and exhibited in an authentic real-world context, engages three people to play on wobble-boards. The thesis argues that creative processes related to Sim-Suite operate within a continuum, encompassing collaborative artistic creation and cooperative creative participation. This continuum is investigated via mixed methods, conducting studies with qualitative and quantitative analysis. These are interpreted through a theoretical lens of embodied cognition principles, the 4E approaches. The results obtained demonstrate that embodied cognitive processes in Sim-Suite's ‘field of artistic communication' function on a continuum. We give an account of the creative process continuum relating our findings to the ‘embedded-extended-enactive lens', empirical studies in embodied cognition and creativity research. Within this context a number of topics and sub-themes are identified. We discuss embodied communication, aspects of agency, forms of coordination, levels of evaluative processes and empathetic foundation. The thesis makes conceptual, empirical and methodological contributions to creativity research.

How a stimulus, such as an odour, is represented in the brain is one of the main questions in neuroscience. It is becoming clearer that information is encoded by a population of neurons, but, how the spiking activity of a population of neurons conveys this information is unknown. Several population coding hypotheses have formulated over the years, and therefore, to obtain a more definitive answer as to how a population of neurons represents stimulus information we need to test, i.e. support or falsify, each of the hypotheses. One way of addressing these hypotheses is to record and analyse the activity of multiple individual neurons from the brain of a test subject when a stimulus is, and is not, presented. With the advent of multi electrode arrays (MEA) we can now record such activity. However, before we can investigate/test the population coding hypotheses using such recordings, we need to determine the number of neurons recorded by the MEA and their spiking activity, after spike detection, using an automatic spike sorting algorithm (we refer to the spiking activity of the neurons extracted from the MEA recordings as MEA sorted data). While there are many automatic spike sorting methods available, they have limitations. In addition, we are lacking methods to test/investigate the population coding hypotheses in detail using the MEA sorted data. That is, methods that show whether neurons respond in a hypothesised way and, if they do, shows how the stimulus is represented within the recorded area. Thus, in this thesis, we were motivated to, firstly, develop a new automatic spike sorting method, which avoids the limitations of other methods. We validated our method using simulated and biological data. In addition, we found our method can perform better than other standard methods. We next focused on the population rate coding hypothesis (i.e. the hypothesis that information is conveyed in the number of spikes fired by a pop- ulation of neurons within a relevant time period). More specifically, we developed a method for testing/investigating the population rate coding hypothesis using the MEA sorted data. That is, a method that uses the multi variate analysis of variance (MANOVA) test, where we modified its output, to show the most responsive subar- eas within the recorded area. We validated this using simulated and biological data. Finally, we investigated whether noise correlation between neurons (i.e. correlations in the trial to trial variability of the response of neurons to the same stimulus) in a rat's olfactory bulb can affect the amount of information a population rate code conveys about a set of stimuli. We found that noise correlation between neurons was predominately positive, which, ultimately, reduced the amount of information a population containing >45 neurons could convey about the stimuli by ~30%.

The aim of this thesis was to investigate some of the processes that contribute to the effective comprehension of text in an adult population. The thesis begins with an assessment of component skills that are of theoretical relevance to reading comprehension skill. Experiment One explored the relation between gist-based memory processes and reading comprehension skill. Weaknesses in semantic processing have been shown to contribute to comprehension difficulties both in childhood (e.g. Nation & Snowling, 1999), and adulthood (e.g. Perfetti, Yang & Schmalhoffer, 2007). Weekes, Hamilton, Oakhill & Holliday (2008) used the false memory (DRM) paradigm developed by Deese (1959); Roediger and McDermott (1995) to assess the relation between reading comprehension and memory processes in children. In the DRM, subjects memorise lists of semantically related words (e.g. bed, rest, awake) for later recall. During recall, it is typical to see intrusions of semantically related but non-presented items (e.g. ‘sleep' is often falsely recalled following presentation of the above). Weekes et al. (2008) found that children with comprehension difficulties produced fewer such intrusions than did good comprehenders, suggesting that poor comprehenders have difficulty extracting the central theme or ‘gist' from the word lists, a deficit that was attributed to weakness in semantic processing and memory. Experiment One demonstrated that this effect was not replicable in an adult population. Although there is evidence that deficits in semantic processing contribute to reading comprehension difficulties in adulthood, these appear to be too subtle to manifest themselves in the DRM paradigm. In Experiment Two, measures of vocabulary, word-level skills (orthography and decoding), working memory and verbal IQ were taken from a population of young adult readers. These measures were used as predictors of comprehension skill in multiple regression analyses. Moderate support for the Verbal Efficiency/Lexical Quality Hypothesis (Perfetti, 1985; 2007) was obtained, in that word-level skills and vocabulary size accounted for unique portions of variance in comprehension skill. Experiments Three and Four explored the processes involved in on-line reading comprehension and, specifically, in a comprehension task that demanded integration. In both experiments, subjects took part in a coherence judgement task (Ferstl, Guthke & von Cramon, 2002; Ferstl, 2006) in which they had to verify whether two sentences cohered with one another or not. Four conditions that resulted from crossing coherence and cohesion (i.e. the presence of a lexical connection), were used: Coherent and cohesive (where sentences cohered, and a cohesive tie made their coherence explicit); coherent and incohesive (where sentences cohered, but coherence had to be inferred on the basis of pragmatic information rather than lexical cohesion); incoherent and cohesive (where sentences that do not cohere were erroneously linked with a cohesive tie); and incoherent and incohesive (where sentences did not cohere, and were not erroneously linked with a cohesive tie). Typically, the paradigm elicits an interaction between coherence and cohesion in reading times for the second (target) sentence: Targets in coherent and cohesive trials are read more quickly than targets in coherent and incohesive trials; and targets in incoherent and incohesive trials are read more quickly than are targets in incoherent and cohesive trials. Experiment Three replicated this interaction, and demonstrated that variance in its size was predicted by working memory capacity, with high working memory readers showing larger interaction effect sizes than low capacity readers. The interaction was interpreted as a monitoring effect that was triggered by target sentences in the atypical conditions (i.e. incoherent and cohesive; coherent and incohesive). It was proposed that high capacity readers were better able to engage in this monitoring. Experiment Four sought to explore the semantic deficit hypothesis in relation to this effect, with the proposal that efficient semantic processes, rather than working memory capacity, contributed to variance in the size of the interaction. Performance on a semantic fluency task was found to predict unique variance in the size of the interaction effect, over and above that accounted for by working memory capacity. This finding suggests that the effect is better explained by semantic processing than by working memory capacity, and that the interaction may be better described as a semantic elaboration effect rather than a comprehension monitoring effect. The conclusion of this thesis is that reading comprehension in adult readers relies upon efficient and accurate lexical access, comprising both lower-level processes such as accurate word recognition and decoding skill, and higher-level processes of semantic elaboration and integration.

The metabotropic glutamate receptor, mGluR5, is densely expressed in brain regions involved in incentive learning processes. There is considerable evidence to suggest that following exposure to addictive drugs such as cocaine, adaptations in these brain areas may underlie the development and maintenance of behavioural responses related to addictive processes. The present thesis examines the role of mGluR5 in both incentive learning processes and some behavioural and neurobiological effects of cocaine. First, using a novel mutant mouse line in which mGluR5 is selectively knocked down in cells that express dopamine D1 receptors (D1R), I argue that this mGluR5 population is critically important for specific incentive learning processes. By blocking mGluR5 in wild-type mice with a selective antagonist, I then propose mGluR5 as necessary for the acquisition, but not the expression of an incentive association. Next, I present data showing that mGluR5 on dopaminoceptive neurons are not necessary for the „conditioned rewarding‟ properties of cocaine, measured in the conditioned place preference model, but do contribute to the psychomotor activating effects of cocaine. Finally, I present an immunohistochemistry study that examines cocaine-induced activation of the extracellular-signal related kinase (ERK) pathway. In the mGluR5 knock-down mice, activation of the ERK pathway in the striatum is disrupted following an acute injection of cocaine. Given the importance of the ERK pathway in establishing and maintaining long term memories, I propose that disruption of this pathway could contribute, in part, to some findings reported in the present thesis. Taken together, this thesis will argue that signalling through mGluR5 on D1R expressing neurons is important for the formation of incentive associations, and may contribute to neural adaptations necessary for the development and maintenance of behavioural responses related to addictive processes.

Tangible User Interfaces (TUIs) offer the potential for new modes of social interaction for children with Autism Spectrum Conditions (ASC). Familiar objects that are embedded with digital technology may help children with autism understand the actions of others by providing feedback that is logical and predictable. Objects that move, playback sound or create sound – thus repeating programmed effects – offer an exciting way for children to investigate objects and their effects. This thesis presents three studies of children with autism interacting with objects augmented with digital technology. Study one looked at Topobo, a construction toy augmented with kinetic memory. Children played with Topobo in groups of three of either Typically Developing (TD) or ASC children. The children were given a construction task, and were also allowed to play with the construction sets with no task. Topobo in the task condition showed an overall significant effect for more onlooker, cooperative, parallel, and less solitary behaviour. For ASC children significantly less solitary and more parallel behaviour was recorded than other play states. In study two, an Augmented Knights Castle (AKC) playset was presented to children with ASC. The task condition was extended to allow children to configure the playset with sound. A significant effect in a small sample was found for configuration of the AKC, leading to less solitary behaviour, and more cooperative behaviour. Compared to non-digital play, the AKC showed reduction of solitary behaviour because of augmentation. Qualitative analysis showed further differences in learning phase, user content, behaviour oriented to other children, and system responsiveness. Tangible musical blocks (‘d-touch') in study three focused on the task. TD and ASC children were presented with a guided/non-guided task in pairs, to isolate effects of augmentation. Significant effects were found for an increase in cooperative symbolic play in the guided condition, and more solitary functional play was found in the unguided condition. Qualitative analysis highlighted differences in understanding blocks and block representation, exploratory and expressive play, understanding of shared space and understanding of the system. These studies suggest that the structure of the task conducted with TUIs may be an important factor for children's use. When the task is undefined, play tends to lose structure and the benefits of TUIs decline. Tangible technology needs to be used in an appropriately structured manner with close coupling (the distance between digital housing and digital effect), and works best when objects are presented in familiar form.

Expertise in topographic map reading is dependent on efficient processing of geographical information presented in a standardised map format. Studies have supported the proposition that expert map readers employ cognitive schemas in which prototypical configurations held in long term memory are employed during the surface search of map features to facilitate map comprehension. Within the experts' cognitive schemas, it is assumed that features are grouped according to spatial configurations that have been frequently encountered and these patterns facilitate efficient chunking of features during information processing. This thesis investigates the nature of information held in experts' cognitive schemas. It also proposes that features are grouped in the experts' schemas not only by their spatial configurations but according to the abstract and functional roles they perform. Three experiments investigated the information processing strategies employed by firstly, skilled map readers engaged in a map reproduction task and secondly, expert map readers engaged in a location comparison exercise. In the first and second experiments, skilled and novice map readers studied and reproduced a town map and a topographic map. Drawing protocols and verbal protocols provided insights into their information processing strategies. The skilled map readers demonstrated superior performance for reproducing contour related data with evidence of the use of cognitive schemas. For the third experiment, expert and novice map readers compared locations within map excerpts for similarities of boundary extents. Eye-gaze data and verbal protocols provided information on the features attended to and the participants' search patterns. The expert group integrated features into their cognitive schemas according to the abstract roles they performed significantly more frequently than the novices. Both groups employed pattern recognition to integrate features for some of the locations. Within a similar experimental design the second part of the third experiment examined whether experts also integrated the abstract roles of remote features and village grouping concepts within their cognitive schemas. The experts again integrated the abstract roles of physical features into their schemas more often than novices but this strategy was not employed for either the remote feature or grouping categories. Implications for map design and future Geographic Information Systems are discussed.

Previous research has shown that high trait-anxious children, relative to low trait-anxious children, are at an increased risk of developing fear due to threatening information (Field, 2006b; Field and Price-Evans, 2009). However, the mechanism that underlies this relationship remains unknown. Cognitive models of vulnerability to anxiety propose that biases in the processing of threat-relevant material play a part in the aetiology and maintenance of anxiety disorders (Beck and Clark, 1997; Eysenck, 1992) and as such could potentially explain the relationship between trait-anxiety and fear development in the face of ambiguous information in children. For example, high-anxious children tend to interpret ambiguous information in a more negative manner (interpretation bias) and remember ambiguous information as being more threatening than it was originally (memory bias) (see Hadwin and Field, 2010, for a review). Additionally, high-anxious children have been found to engage in negative cognitive rehearsal (Comer, Kendall, Franklin, Hudson, and Pimental, 2004). The experiments in this thesis investigated whether these cognitive biases underlie the relationship between trait anxiety and fear development in non-clinical children. In a series of three experiments, children (aged 8-11 years) were presented with some ambiguous information regarding two novel animals (the quoll and the cuscus) and before completing a cognitive rehearsal task were told that they would soon be asked to approach the animals. There were several findings: 1) High trait-anxious children were not significantly more likely than low trait-anxious children to display any of the cognitive biases tested (i.e., interpretation bias, memory bias or cognitive rehearsal). However, tentative evidence suggested that interpretation bias exacerbated the relationship between trait anxiety and fear; 2) Whether children cognitively rehearsed the ambiguous information or not had no significant impact on their fear for the animals, nor did the valence of their thoughts; 3) Children who interpreted the ambiguous information more negatively were more likely to become fearful of the animals and were also more likely to remember more negatively-biased and less positively-biased pieces of ambiguous information; 4) It was the lack of positively-biased memories not the increased number of negatively-biased memories that led children who interpreted the information more negatively to become more fearful of the animals as a result. The findings are discussed with reference to their implications for the theory and prevention of childhood fear: that positive interpretation and memory bias training may act to decrease or even help to prevent fear development in children.

A wealth of support has shown higher levels of state and trait impulsivity can be found among those individuals prone to developing problematic eating behaviors and obesity. Thus, upon commencing the investigations in this thesis, it was hypothesized that impulsivity is an individual difference implicated in overeating behaviour. Increasing information indicates that there are divisions within impulsivity subtypes. Prior to this thesis, studies in the field of eating behaviour had not distinguished between subtypes of impulsivity. This was problematic because it limited researchers ability to describe how impulsivity is specifically involved in the perpetuation of overeating behaviour. The purpose of this Thesis was to provide a methodical inquiry into the relationship between impulsivity, and its relation with overeating behaviour. This objective was achieved by separating three prioritites, first to determine if impulsivity was higher in women who self report overeating, second to define differences between impulsivity classifications and determine if there was a consistnent pattern between self reported overeating and a relation to a subtype of overeating behaviour, and third to designate a specific impulsivity subtype to individuals who self report overeating behaviour. Six Experimental Chapters explored these three priorities. Two exploratory correlational/regression analysis were used to refine our ability to operationalize measures of self reported overeating and impulsivity (Chapters 2 and 5). Chapter 3 and Chapter 5 were devoted to assess the impact that ingestion of palatable food, and the violation of cognitive boundaries of restraint, have on subsequent impulsivity. The two remaining investigations were structured to assess the impact that environmental factors have on impulsive behaviour. In Chapter 4, a Controlled versus Unrestricted eating environment were manipulated to determine whether overeaters benefit from a structured breakfast meal prior to completing a battery of impulsivity tasks. In Chapter 7, anticipation for a rewarding food item was manipulated in two conditions. In this final Chapter, the impact that anticipation for rewarding food in self reported overeaters was assessed. The battery of impulsivity tasks in this thesis include the Balloon Analogue Risk Task (BART), The Go No Go task, the Matching Familiar Figures task (MFFT), and two versions of the Delay Discounting Task (DDT). Impulsivity was classified along a spectrum of Reward Reactivity versus Inhibition subtypes, based on Evenden´s (1999) classification of impulsive behaviour. Participants tendency to overeat was based on a dual classification of tendency to restrain eating (Three Factor Eating Questionnaire- Restraint) with tendency to overeat (Three Factor Eating Questionnaire-Disinhibition subscale). The outcomes of the five experimental investigations in this thesis demonstrated a reliable pattern by which participants with high Disinhibition scores had significantly more impulsive responses on the MFFT task. These results indicated that inhibition impulsivity is the clearest individual difference to be found between healthy volunteers who self-report overeating. The role that Inhibition Impulsivity plays in the perpetuation of overeating behaviour is illustrated and discussed in each Experimental Chapter.

This thesis presents a rat model, by which the effects of repeated ethanol withdrawal on withdrawal severity was investigated, in relation to the cognitive and behavioural deficits associated with repeated episodes of withdrawal. Repeated ethanol withdrawal in the rat has been well established to model the effects of repeated episodes of human alcohol detoxification. This model has enabled the study of withdrawal severity and the role of the prefrontal cortex in the form of rat behaviour. Chronic ethanol consumption led to disrupted circadian rhythm especially in measures of wakefulness and NREM sleep. However, there were no cumulative effects of multiple ethanol withdrawals. These results were confounded by altered circadian rhythms observed in the control group. Repeated ethanol withdrawal induced significantly higher levels of C-Fos, a marker of neuronal activation, compared to a single withdrawal episode. In!addition, repeated ethanol withdrawal also induced significantly higher Zif 268 expression, a marker for neuronal plasticity, in the prelimbic cortex. These findings indicated a sensitivity of prefrontal cortical areas in response to repeated ethanol withdrawal. In assessing performance on a 2-choice serial reaction time task, repeated ethanol withdrawal resulted in more sessions to criterion, indicating possible learning deficits but only when the withdrawal experience occurred prior to behavioural training and testing. Repeated ethanol withdrawal did not significantly impair attentional set shifting ability on the intradimensional/extradimensional task. The findings of this current thesis suggest that repeated ethanol withdrawal did not produce significantly severe cognitive deficits as measured by behavioural tasks sensitive to prefrontal cortical damage, despite neurobiological activation of prefrontal areas.

Peer interactions are recognised as playing a key role in the development of children, but we lack detailed analysis of individual differences in the early peer play of preschoolers. Five data sets are used to explore aspects of children's developing social competence between the ages of 2 and 5 years. Four of the five research investigations were carried out in day nurseries, and the remaining study was conducted in a reception class (children aged 5 years), all in a London Local Authority. The first paper explores core elements of peer play which can be identified through direct observation. It serves the dual purpose of highlighting children's real life experience of making social connections through peer interactions and exploring the key dimensions of verbal and nonverbal behaviour that support such connections. Papers 2 and 3 are mainly focused on exploring the different perspectives of parents and practitioners in their views of children's current social adjustment, with additional reports on child temperament from parents and reports on peer play from practitioners. Finally, Papers 4 and 5 explore in greater depth a range of potential predictors of young children's social competence, including temperament and socioemotional understanding. Being able to recognise individual differences in patterns of play specifically in terms of levels of interaction and disconnection led to the use of the Penn Interactive Peer Play Scale throughout the remaining studies. The notable differences in levels of successful interactivity underlined the need to measure children's effectiveness in using a range of abilities to establish and maintain engagement with play partners. The further studies involved a total of 516 practitioner reports and 179 parent reports on children's behaviour, social competencies and temperament, as well as 123 direct assessments of children's socio-emotional understanding. Matched parent and practitioner questionnaires were used to examine similarities and differences in adult perceptions and interpretations of children's peer play. Levels of convergence between parent and practitioner views of children's socio-behavioural functioning were found to change as children get older, from an early convergence on prosocial behaviours to a later convergence on problem behaviours. The results also highlighted the particular roles of temperament and socioemotional understanding in peer play. Effortful control was found to be a significant predictor of positive, interactive play. Furthermore, socio-emotional understanding – as assessed through the use of simple structured tasks and hypothetical scenarios – was found to predict patterns of interactive play, thereby evidencing the sociocognitive factors involved in effective peer interactions. Gender differences were also evident, suggesting that girls and boys may rely on different attributes and skills to forge social connections. The key findings are discussed with attention to their implications for effective practice in early years provision, developing our understanding of early social competence from different perspectives. Directions for further research are presented.

In human fear conditioning 'resistance to extinction' occurs when the removal of the aversive outcome fails to produce a reduction in conditioned responding. This phenomenon is important to understanding the persistence of anxiety disorders such as phobias. The research presented in this thesis examines factors that promote the acquisition and maintenance of learned fear response and attempts to differentiate between different explanations of the resistance to extinction phenomenon. To investigate the impact of different conditioning procedures (evaluative or classical conditioning) on the durability of the conditioned response (CR), event-related potential (ERP) methodology was employed. In addition, the role of the fear-relevance of the conditioned stimulus (CS), in supporting the acquisition and resistance to extinction of the CR, was explored. Evidence suggested that extinction effects are likely to reflect procedural differences in conditioning rather than different underlying learning processes. Extinction effects were dissociable across procedures, supporting the role of the type of unconditioned stimulus (US) in explaining past demonstrations of extinction when responses were indexed by physiological measures. Verbally transmitted, threat information heightened aversive US-expectancies and fear beliefs without the need for conditioning. Additionally, fear-beliefs were reduced without the need for extinction training when positive information was provided. Contrary to Davey's (1997) expectancy bias model, the results do not support the hypothesis that verbal information interacts with direct contingency experience to create fear responses; instead, information appears to be a direct pathway to fear. ERP measures for fear responses did not echo the effects of verbal information and contingency on fear-beliefs. However, the comparability of our ERP data, to other research using physiological measures of response, is discussed regarding the number of trials required to calculate the average ERP response. Due to averaging over a large number of trials the ERP measure may not be sensitive to fluctuations in response that may be dependent on information or contingency manipulations. In conclusion our data suggests the importance of verbal information as a pathway to fear and the role of cognitive factors in the prevention and treatment of fears.

Previous research has shown that we recognise faces similar in age to ourselves better than older or younger faces (e.g. Anastasi & Rhodes, 2006). The primary aim of this thesis was to investigate this phenomenon in young adults and children to gain further insight into the underlying perceptual, cognitive and/or social mechanisms involved in this apparent “own-age bias” (OAB) in face recognition. Chapter one confirmed that an OAB was present in both young adults and children, and the remaining chapters sought to address why this pattern may exist by drawing on the plethora of research into why a similar, potentially analogous bias occurs: the own-race bias (ORB). The ORB is the phenomenon that we are more accurate at recognising faces of our own race than those belonging to a different, less familiar race (see Meissner & Brigham, 2001 for review). Perhaps the best known explanation of the ORB is the Contact Hypothesis. This suggests that the own-race memory advantage is due to the fact that people tend to have more experience with faces from their own race and, as a direct result, develop greater expertise at recognising them (e.g. Chiroro & Valentine, 1995). The second chapter sought to investigate whether a similar explanation could be applied to the OAB, and found supporting evidence for this claim. The remaining studies examined what it is about contact with an age group that results in the superior recognition for faces of that age. By investigating perceptual expertise, social-categorisation and motivational explanations of the OAB, this thesis concluded that both quantity and quality of contact play an important role in the development of this bias. The findings of this thesis seem to be most consistent with a perceptual expertise account of the own-age bias in face recognition. However, it also seems likely that motivation to attend to faces (particularly with the goal of individuation) is likely to be a driving factor of this bias.

The main argument developed in this thesis is that individuals are not only affected by their own experiences of intergroup contact, but also by the contact they perceive is occurring around them. A pivotal reason for this is a change in the perception of ingroup norms regarding intergroup interactions, which can shape intergroup attitudes and make future positive intergroup contact experiences more likely. Chapter 1 provides a review of the literature of intergroup contact and makes the case for the necessity of a normative perspective on intergroup contact research. I Chapter 2, I provide a brief overview of the literature regarding the effect of social norms in intergroup relations, and describe in more detail the theoretical model proposed. Chapters 3 and 4 provide some initial evidence for the model using cross-sectional studies, showing that ingroup norms are associated with attitudes and with the intention of having intergroup contact, and that direct and indirect contact rely on different mechanisms to improve attitudes towards the outgroup. The effect of indirect contact is of a normative nature while the effect of direct contact is more emotional. Chapter 5 focuses on the distinction between ingroup norms about contact and ingroup norms about expressing prejudice, and the evaluation of moderators of the effect in ingroup norms. Chapter 6 reports two longitudinal studies, providing some evidence in support for a causal effect of ingroup norms about contact on several outcomes; and Chapter 7 reports a series of experiments, including results showing that ingroup norms predict the success of an actual interaction with outgroup members. In Chapter 8, I summarise the evidence for the theoretical model proposed and discuss some promising directions for future research, as well as the implications of these results for the reduction of segregation and the improvement of intergroup relations in general.

The aim of this thesis is to examine the relationship between the intrinsic dynamics of the body and its neural control. Specifically, it investigates the influence of musculoskeletal properties on the control signals needed for simple goal-directed movements in the framework of the equilibriumpoint (EP) hypothesis. To this end, muscle models of varying complexity are studied in isolation and when coupled to feedback laws derived from the EP hypothesis. It is demonstrated that the dynamical landscape formed by non-linear musculoskeletal models features a stable attractor in joint space whose properties, such as position, stiffness and viscosity, can be controlled through differential- and co-activation of antagonistic muscles. The emergence of this attractor creates a new level of control that reduces the system's degrees of freedom and thus constitutes a low-level motor synergy. It is described how the properties of this stable equilibrium, as well as transient movement dynamics, depend on the various modelling assumptions underlying the muscle model. The EP hypothesis is then tested on a chosen musculoskeletal model by using an optimal feedback control approach: genetic algorithm optimisation is used to identify feedback gains that produce smooth single- and multijoint movements of varying amplitude and duration. The importance of different feedback components is studied for reproducing invariants observed in natural movement kinematics. The resulting controllers are demonstrated to cope with a plausible range of reflex delays, predict the use of velocity-error feedback for the fastest movements, and suggest that experimentally observed triphasic muscle bursts are an emergent feature rather than centrally planned. Also, control schemes which allow for simultaneous control of movement duration and distance are identified. Lastly, it is shown that the generic formulation of the EP hypothesis fails to account for the interaction torques arising in multijoint movements. Extensions are proposed which address this shortcoming while maintaining its two basic assumptions: control signals in positional rather than force-based frames of reference; and the primacy of control properties intrinsic to the body over internal models. It is concluded that the EP hypothesis cannot be rejected for single- or multijoint reaching movements based on claims that predicted movement kinematics are unrealistic.

This thesis applies self-categorization theory (SCT) to the behavioural and experiential dimensions of spatiality and crowding. A literature review in Chapter 2 will highlight the study of ‘personal space' as a dominant theoretical approach to understanding variable crowding experiences. Several problems with the conceptualisation of ‘personal space' will be discussed and in Chapter 3, SCT will be introduced as a theoretical framework, which can explain spatiality and crowding in terms of identity and group-level processes. Chapter 4 investigates the impact of group relations on spatiality. In study 1, participants in minimal groups sought closer proximity to an in-group member than to an out-group member. Study 2 used national categories as the basis for identification. Although non-significant, the findings mirrored those of study 1. Additionally, the relationship between perceived difference and physical distance was moderated by the group context of the interaction. Chapter 5 investigates the impact of group context and interaction distance on participants' subjective experience. The hypothesised effect of these two independent variables on participants' experience was not supported. However, participants in studies 3 and 4 expressed a desire for closer proximity to in-group members than to out-group members. Chapter 6 investigates the effect of visualised crowding on subjective affect. In study 5, participants who visualised an in-group crowd reported higher positive affect and lower negative affect than those who visualised an out-group crowd. In study 6, the relationship between imagined ‘crowding' and affect was mediated by social identification. The relationship between physicality, social identity and ‘collective joy' was quantified in a field study (study 7), which also provided suggestive evidence of physical synchrony increasing social identification. A laboratory study (study 8) supported the suggestion that synchronised movement can increase social identification, thus leading to a more positive experience of close proximity. These finding are discussed in terms of a two-way embodiment model of social identity processes in crowds.

Attentional biases arising from classical conditioning processes may contribute to the maintenance of drug addictions and anxiety disorders. This thesis examined whether attentional mechanisms for conditioned stimuli (CS) would be dominated by affective properties (Lang, Greenwald, Bradley, & Hamm, 1993), or the uncertainty of the stimulus in predicting the outcome (Pearce & Hall, 1980). In chapter one affective and uncertainty-driven mechanisms of attention are discussed in relation to rewarding and aversive outcomes. In experimental chapter 2 methodological issues are addressed. In experimental chapters three and four attentional mechanisms are tested using a discriminative conditioning procedure with visual stimuli of varying predictive certainty (CS+,CS+/-,CS-) for a monetary or noise outcome (US). Attention was measured using an eye-tracker, and emotional conditioning and learning were measured using Likert scales. It was found that attention was mediated by uncertainty (chapter 3), but increasing the intensity of the outcome switched attention to affective-driven mechanisms for the noise outcome (chapter 4). In a further experiment this effect on attention remained for the noise outcome even under conditions promoting uncertainty-driven mechanisms (chapter 6). When cigarettes were the unconditioned stimuli instead of money in the appetitive conditioning, attention was also mediated by stimulus affect (chapter 5). In chapter 7 the data are discussed and it is concluded that when the outcome is highly emotionally salient, affective-driven mechanisms of attention dominate over uncertainty.

According to the attention network theory, attention is viewed as an organ system comprising specialised networks that carry out functions of alerting, orienting and executive control. The Attention Network Test (ANT) is a simple and popular experiment that measures the efficiencies and interactions of these three subcomponents of attention in a single task, and has been used for adults, children and attention deficit patients. In this thesis, cognitive modelling is used as a research tool to simulate the performance of subjects on the ANT, as well as variations of the ANT using ACT-R 6.0 cognitive architecture. All models are validated against human data using various goodness-of-fit criteria at multiple measures of the latency, accuracy and efficiency of the three networks. Once the simulation of healthy human performance on the ANT is established, modifications inspired by psychology literature are made to simulate the performance on ANT by children and patients affected with Alzheimer‘s disease (AD) and mild traumatic brain injury (mTBI). The implementation of networks, their interactions and impairments in the models are shown to be theoretically grounded. Based on the simulation results and the understanding gained through model processes, a number of novel predictions are made, behaviour of the networks and a few discrepancies in human data are explained. The model predicts that in the case of Alzheimer‘s disease, the orienting network may be impaired and cueing may have a positive effect on conflict resolution. Also, in the case of mTBI, it was predicted that the validity effect may be impaired only in the earlier weeks after the injury. For children, a possible relationship between processing speed and mechanism of inhibitory control is predicted. It is posited that there is not always a 'global clock' that controls processing speed and further different processes may be running with different processing times.

Sentiment analysis focuses upon automatic classiffication of a document's sentiment (and more generally extraction of opinion from text). Ways of expressing sentiment have been shown to be dependent on what a document is about (domain-dependency). This complicates supervised methods for sentiment analysis which rely on extensive use of training data or linguistic resources that are usually either domain-specific or generic. Both kinds of resources prevent classiffiers from performing well across a range of domains, as this requires appropriate in-domain (domain-specific) data. This thesis presents a novel unsupervised, knowledge-poor approach to sentiment analysis aimed at creating a domain-independent and multilingual sentiment analysis system. The approach extracts domain-specific resources from documents that are to be processed, and uses them for sentiment analysis. This approach does not require any training corpora, large sets of rules or generic sentiment lexicons, which makes it domain- and languageindependent but at the same time able to utilise domain- and language-specific information. The thesis describes and tests the approach, which is applied to diffeerent data, including customer reviews of various types of products, reviews of films and books, and news items; and to four languages: Chinese, English, Russian and Japanese. The approach is applied not only to binary sentiment classiffication, but also to three-way sentiment classiffication (positive, negative and neutral), subjectivity classifiation of documents and sentences, and to the extraction of opinion holders and opinion targets. Experimental results suggest that the approach is often a viable alternative to supervised systems, especially when applied to large document collections.

The mood-as-input hypothesis predicts that perseveration at an open-ended task is determined by “stop rules” for the task and by the valency of the mood. Stop rules define a person's goals in task attainment, e.g. stopping after doing as much as they can, or stopping when they no longer feel like continuing. This thesis will examine the combined effects of stop rules and specific negative moods (sadness, anxiety, anger) on perseverative worrying and checking tasks, and the influence of specific negative moods on personal performance standards. The final study explores the impact of experimentally induced mood on a worry task when the mood source is made highly salient i.e. attributed to an obvious event or source. On a perseverative checking task, different negative mood and stop rule combinations were found not to affect participant performance. However, using a personally-relevant worry task, participants in each specific negative mood condition persevered for longer using an “as many as can” rule compared with those using a “feel like continuing” rule. The opposite was found for participants in a happy mood. The effects of sadness and anxiety on personal performance standards and stop rule preference were also examined. Findings suggest a positive relationship between sad and anxious moods and “as many as can” stop rule preference. An attempt to manipulate mood attribution after inducing an angry mood showed marginally significant differences in attribution by the high and low manipulation groups, but no effects of mood attribution on task performance. These findings suggest that with a catastrophic worry task, participants in each specific negative mood condition using an “as many as can” stop rule persevered for longer compared with those using a “feel like continuing” stop rule. The implications of this work are discussed in relation to mood-as-input accounts of perseveration and models of mood.

Cognitive and behavioural theories of social anxiety and depression provide clear explanations for the links between these conditions and the strategies children use to cope with peer conflict situations. However, empirical research in the area has left several unresolved issues, warranting further investigation if we are to understand more fully the links between coping and emotional adjustment. This programme of research was designed to develop a comprehensive measure of children‟s coping, particularly in the context of peer stressors, and to examine the links between specific coping strategies and social anxiety and depression over time. In a series of seven studies, reported in four papers, a total of 833 primary and secondary school children completed measures of social anxiety, depression, coping, and a sociometric survey, as well as measures of goals and appraisals. In Paper 1, seven distinct coping strategies were revealed across several interpersonal situations that were related to children‟s feelings in distinct ways, and that meaningfully mapped onto differences between a mainstream school sample and a sample of pupils with emotional and behavioural difficulties. In Paper 2, six of the seven coping subscales identified in Paper 1 were confirmed and these specific ways of coping were differentially associated with social anxiety and depression. Generally, social anxiety and depression were longitudinally associated with distinctive profiles of coping strategies over a period of 9 months. In Paper 3, coping was found to have these differential associations with social anxiety and depression across a range of peer conflict situations, and there was also evidence of mediating effects of children‟s appraisals and goals. Finally, in Paper 4, coping was found to be predictive of changes in depression over one year, but associations between coping and emotional adjustment did not hold up over a two-year period. These findings are discussed in relation to the existing coping literature and theories of social anxiety and depression.

The role of γ-aminobutyric acid (GABA) in modulating nucleus accumbens (Acb) function was investigated using feeding in rats as a model of motivation. Feeding encompasses processes associated with ‘appetite', ‘satiety' and ‘reward'. Distinct neurotransmitter systems in the Acb have been demonstrated to control specific behavioural mechanisms subserving these motivational processes. The Acb has been described as an interface between ‘motivation' and ‘action'. It is involved in the modulation of feeding, sexual behaviour, defensive behaviours and drug seeking. Over 97% of Acb neurons are GABAergic and GABA mimetics robustly increase food intake in satiated animals. One current hypothesis suggests that GABAA and GABAB receptors gate control of ingestive motor responses via the same circuit but this circuit cannot control more complex goal-directed behaviours (Kelley et al., 2005). Temporal changes in the feeding related ‘behavioural satiety sequence' (BSS) correlate with mechanisms that override satiety. The BSS is sensitive to effects on consummatory behaviour. Satiated rats given chow following intra-Acb infusions of a GABAB receptor agonist fed voraciously and the BSS was delayed but all other behaviours were still present. The pattern was similar with fasting and a systemically administered benzodiazepine but, with a μ-opioid agonist (postulated to increase the incentive value of food), the peak in feeding was delayed. A GABAA receptor agonist induced feeding but all other behaviours were significantly reduced at all effective doses. In a second order operant schedule that independently measured appetitive and consummatory behaviour, the GABAA and opioid receptor agonists had no effect on responding but the GABAB receptor agonist increased reinforced responding in a dose dependent manner. Several other behavioural indices of motivation also increased, suggesting behaviourally selective effects. These results are inconsistent with the hypothesis described above. Intra-Acb GABA receptor subtype stimulation led to significant differences in the location and magnitude of activity in motivation related brain structures, particularly the lateral hypothalamus and amygdala, which were revealed using Fos-like immunoreactivity as a marker. The differential modulation of feeding behaviour via GABA receptor subtypes in the Acb is used to construct a modified model to explain endogenous inhibitory motivational control.

The research presented in this thesis examined the roles of recollection and familiarity in the own-race bias (ORB) in recognition memory for faces. In Paper 1, Jacoby's (1991) process-dissociation procedure was used to estimate the relative contributions of recollection and familiarity in recognizing own- and other-race faces. Recollection estimates were higher for own-race faces than for other-race faces, although this effect disappeared when deep or shallow encoding strategies were encouraged. In Paper 2, participants were shown to be less accurate at ignoring previously seen other-race distractors than own-race distractors. Papers 3 and 4 examined how accurately participants were able to remember contextual information about correctly recognized faces. In the encoding phase of an old/new recognition test, each target face was paired with one of several different backgrounds. At testing, old judgments were followed by context judgments, in which the participant attempted to identify with which background the face had been paired. The context judgments were consistently more accurate for correctly recognized own-race faces than for correctly recognized other-race faces. This effect was robust to experimental manipulations such as context reinstatement and divided attention. The overall conclusion from this thesis is that recollection is inferior for other-race faces compared to own-race faces. This recollection deficit means that it is more difficult to retrieve specific information about the circumstances in which other-race faces were encountered. The implications of this recollection deficit for real world behaviour are discussed, with particular reference to eyewitness memory.

Online social networks are understood to replicate the real life connections between people. As the technology matures, more people are joining social networking communities such as MySpace (www.myspace.com) and Facebook (www.facebook.com). These online communities provide the opportunity for individuals to present themselves and maintain social interactions through their profiles. Such traces in profiles can be used as evidence in deciding the level of trust with which to imbue individuals in making access control decisions. However, online profiles have serious implications over the reality of identity disclosure. There are many reasons why someone may choose not to reveal their true self, which sometimes leads to misidentification or deception. On one hand, the structure of online profiles allows anonymity, which gives users the opportunity to create a persona that may not represent their true identity. On the other hand, we often play multiple identities in different contexts where such behaviour is acceptable. However, realizing the context for each identity representation depends on the individual. As a result, some represented identities will be essentially real, if edited for public view, some will be disguised, and others will be fictitious or humorous. The millions of social network profiles, and billions of connections between them, make it difficult to formalize an automated approach to differentiate fact from fiction in online self-described identities. How can we be sure with whom we are interacting, and whether these individuals or groups are being truthful with the online identities they present to the rest of the community? What tools and techniques can be used to gather, organize, and explore the available data for informing the level of honesty that should be entrusted to an individual? Can we verify the validity of the identity automatically, based on the available information online? We aim to evaluate identity representation online and examine how identity can be verified in a less trusted online community. We propose a personality classifier model to identify a user‟s personality (such as expressive, valid, active, positive, popular, sociable and traceable) using traces of 2.2 million profile features collected from MySpace. We use data mining techniques and social network analysis to extract significant patterns in the data and network structure, and improve the classifier during the cycle of development. We evaluate our classifier model on profiles with known identities such as „real‟ and „fake‟. Our results indicate that by utilizing people‟s online, self-reported information, personality, and their network of friends and interactions, we are able to provide evidence for validating the type of identity in a manner that is both accurate and scalable.