The efficient detection of information of negative valence in the environment is crucial to survival (e.g. to elicit an avoidance response). However, previous research remains inconclusive regarding the question of whether detection is more sensitive to information of negative compared to positive valence. In the present thesis I used a signal detection approach applied to an emotional-evaluation word task (requiring the participants to classify a briefly presented masked word into emotional or non-emotional categories) to address this question. The results established conclusively enhanced detection sensitivity to negative valence compared to positive valence of verbal information, under both supraliminal and subliminal conditions (Chapter 2) while ruling out any alternative accounts in terms of word frequency, idiosyncratic differences in valence ratings and different levels of arousal. The extent to which the enhanced negative valence detection depends on availability of attention was addressed in Chapter 3. Using a dual-task paradigm, participants performed the emotional detection task together with a letter-search task of either low or high perceptual load. A negative valence detection advantage was found in the low load but not high load conditions. These results established that attentional resources are critical for the enhanced detection of negative valence. The role of individual differences in trait anxiety in the effects of attention on valence detection was examined in Chapter 4. The results demonstrated that high trait anxiety was associated with enhanced detection of negative valence even under high load, whereas individuals with low trait anxiety were less sensitive to negative valence across both levels of load. The effects of monetary reward were addressed in Chapter 5. The results indicated that while reward enhanced detection sensitivity, the negative valence detection advantage remained unaffected. Overall the results establish conclusively a negative valence detection advantage that interacts with attention, trait anxiety, but not with reward.

Attachment theory has increasingly been applied to the understanding of individual differences in emotion regulation, however application of the attachment framework to anger is underspecified. The present thesis describes three studies reporting relationships between attachment insecurity and anger expression, taking into account attachment-related differences in anger regulation. Using multiple regression analyses, the results of Study 1a indicated that attachment anxiety was a significant independent predictor of trait anger, while attachment avoidance and self-esteem were not. Study 1b extended these findings by examining whether the use of specific anger regulation strategies mediated the relationship between attachment insecurity and dispositional aggression (physical aggression, verbal aggression and hostility). Attachment anxiety was indirectly related to physical aggression and hostility, through the use of maladaptive regulation techniques and a lack of anger control; while attachment avoidance had an indirect relationship with hostility through anger suppression. Study 2 utilised an anger induction procedure to investigate the relationship between attachment insecurity, self-reported and physiological responses, and subsequent aggressive behaviour. Neither attachment dimension was significantly associated with physiological reactivity to the anger induction, however attachment avoidance was negatively associated with changes in self-reported anger. Attachment avoidance was once again a significant predictor of anger suppression, while attachment anxiety significantly predicted variance in aggression. Overall, the findings indicate that attachment anxiety is a predictor of dispositional anger and aggression, whilst attachment avoidance predicts the use of suppression to regulate anger, reduced self-reported anger responsivity and increased hostility. Implications for both theory and practice are discussed with suggestions for attachment and emotion regulation-based anger management interventions.

Hate on the Internet, and particularly the role of Social Networking Sites (SNSs) in shaping online hate, the topic of investigation in this thesis, has been neglected in psychological and hate crime literature. Chapter 1 introduces the concept of online hatred (i.e. expressions of hateful content on the Internet), points out the gaps in current online hate research and emphasises the importance of studying this concept in an applied psychological context. Chapter 2 describes two studies, which together develop an explicit prejudice measure towards Roma and Travellers, as a) there was no specific one for these groups and b) they were the target groups in the study discussed within Chapter 3. The first study tested the general statement suitability of an existing prejudice measure (i.e. Levinson & Sandford’s 1944 Anti-Semitism scale) to measure non-group specific prejudice. Sixteen statements were identified as suitable to measure non-group specific prejudice. The second study tested the specific appropriateness of the 16 non-group specific statements for describing stereotypes associated with Roma and Travellers. Here, 10 out of the 16 statements were rated as appropriately describing stereotypes associated Roma and Travellers. Chapter 3 describes a study that aimed to investigate the persuasive effects (i.e. changes in levels of prejudice) of online discussions on small ‘like-minded’ groups of participants. Groups of participants with similar levels of prejudice (i.e. low, intermediate, high) towards Roma and Travellers (the target groups) discussed the eviction of a particular British Roma and Traveller site via instant messaging online. During the discussions, a confederate expressed views which aimed at either increasing or reducing prejudice towards Roma and Travellers. Results revealed that only participants with intermediate levels of prejudice towards Roma and Travellers were influenced by the discussions (i.e. participants became more prejudiced). Yet, participants with intermediate levels of prejudice resisted online influence which aimed at reducing their levels of prejudice. Overall, the results indicate that participants with intermediate levels of prejudice (i.e. weak attitudes) gave in to online influence, whereas those with low or high levels of prejudice (i.e. strong attitudes) resisted it. Chapter 4 explores how polarised and non-polarised YouTube users responded to racist online content. In particular, 71,000 user comments made in response to a video clip, depicting a woman on a London tram who racially abused ethnic minority passengers, were analysed using thematic analysis. The analysis revealed that the exposure to hateful online content does not lead to an automatic endorsement of hatred. More specifically, some YouTubers responded by trying to account for the woman’s behaviour, as well as with hateful comments in response to viewing the hateful video clip. It also revealed that many YouTubers attempted to position themselves and other users according to their stance on racism. In addition, most responses focused on a more complex issue, namely the contestation of Britishness in relation to immigration. Chapter 5 discusses a survey, exploring the permissibility of online hatred among Social Networking Site (SNS) users. The survey specifically explored SNS users’ perceptions towards what constitutes online hate, the association between online and offline hate and the role of online anonymity on expressing online hatred. Results were somewhat contradictory and thus not entirely clear. In particular, whilst most participants did not connect online with offline hatred, they blamed the victims of online hatred for their abuse but not the creators of hateful content. Further, they did not agree with legislation governing hate speech (i.e. pointing towards the permissibility of online hatred). Participants also rated hateful content to be criminal and admitted that online anonymity would aid their own anti-normative behaviour (i.e. implying online hatred is not permissible). Chapter 6 summarises the main findings from this thesis and discusses their implications, methodological considerations, and suggestions for future research.

No description available.

No description available.

The five experiments presented in this thesis aimed to investigate the relationship between subjective presence (the sense of 'being there' in a mediated environment) and subjective and physiological emotional responses. The investigation served as an initial step in the evaluation of emotion-based corroborative measures of presence. Two of the determinants of presence (Media Form and Media Content) were experimentally manipulated in order to create varying levels of presence and different types of subjective and physiological emotional responses. Varying levels of presence were created by manipulating Media Form characteristics in the context of a video display - specifically, the absence and presence of stereoscopic cues (Experiments I and 2) and small versus large eye-to-screen visual angles (Experiments 3 and 5). Media Content (video clips) with varying types of emotional impact was presented to participants via the video displays (Experiments I, 2, 3 and 5). The research showed that enhancements in Media Form increased both subjective presence and subjective arousal across all Content types. However, there was little evidence to show that Media Form manipulations affected physiological arousal, indicating limited utility for physiological measures of presence in this context. Media Content characteristics were investigated in more depth in Experiments 4 and 5. Differences in SUbjective presence between Media Contents shown to elicit different types of subjective emotion were investigated and correlations between subjective presence and subjective emotion were examined. The research indicated that the quality and intensity of emotional responses elicited by Media Content are potentially important determinants of subjective presence. The thesis concludes by suggesting that in further investigations of presence and emotion, which aim to identify corroborative measures of presence, it may be useful to analyse the relationship between different dimensions of presence (physical Space, Naturalness and Engagement) and different components of emotional responses (Subjective, Cognitive, Behavioural and Physiological).

The aim of this programme of research was to explore individual differences in the operation of interpretive and attentional biases within the framework of the fourfactor theory (Eysenck, 1997). Study one examined symptom interpretation and the influence of interpretive biases on future performance expectations in high-anxious, low-anxious and repressor individuals. Contrary to predictions, anxiety symptom interpretation was found not to differ across groups. In addition, high-anxious individuals showed no tendency towards pessimism, whereas repressors were overly-optimistic regarding their future performance. Study two examined the effect of bogus performance feedback on the exhibition of interpretive biases. The results showed that high-anxious individuals were initially overly pessimistic regarding future performance expectations whereas repressors were initially overly optimistic. However, positive feedback eliminated pessimistic expectations in high-anxious individuals and negative feedback eliminated optimistic expectationsin repressors. In study three, the influence of mood on attentional biases to negative and positive emotional stimuli was examined. High-anxious individuals' attentional bias to negative stimuli was apparent only during negative mood. No evidence of an opposite attentional bias in repressors was found irrespective of induced mood. Repressors showed an avoidance of positive stimuli during negative mood. Positive mood exerted no influence on the attentional biases of the groups to either positive or negative stimuli. Study four examined the influence of exercise-induced physiological activity on the attentional biases of high anxious and repressor groups. High anxious individuals demonstrated a bias towards negative stimuli during moderate levels of physiological activity. Repressors did not show avoidance of threat in any of the experimental conditions. However, high levels of exercise-induced physiological activity were associated with an avoidance of positive stimuli in repressors. In summary, the studies demonstrated that cognitive biases of high anxious individuals are not as stable as previously thought and they are influenced by feedback, prevailing mood and physiological activity. The findings fail to support the contention that repressors avoid processing threat. Paradoxically, these individuals appeared more likely to avoid processing positive stimuli. In light of these findings, this research suggests that Eysenck's (1997) four-factor theory requires modification.

Research shows that our personalities determine how we think, feel and behave. The aim of this research therefore was to examine individual differences in emotional responses on the basis of participants' scores for state-anxiety and the traits of anxiety, neuroticism and extraversion. Behavioural reaction times and neurological (EEG) were measured to emotional stimuli using dot-probe tasks. Further aims were to establish whether any differences found could be modulated by attentional training and to assess the use of alpha asymmetry in anxiety research. We found that trait-anxiety was the only trait to underpin differences in the response to emotion both behaviourally and neurologically. Although neuroticism and extraversion were responsible for differences in the response to emotion behaviourally they were not in the response to emotion neurologically. Our attentional training protocol had no effect on brain activity and a number of explanations are offered for this. When assessing the use of alpha asymmetry in anxiety research, differences based on trait-anxiety were evident supporting its use as a potential endophenotype for anxiety. We argue therefore that the most reliable measure of susceptibility and resilience to emotional disorders is an individual's level of trait-anxiety, both the cognitive and neurological response to emotional stimuli being evidently different dependent on an individual's score; this score also underlying differences in alpha asymmetry. Whilst evidence also indicates that differences in neuroticism and extraversion underlie individual differences in the cognitive response to emotion, this was not apparent neurologically suggesting that trait-anxiety is a more robust measure of susceptibility or resilience to emotional disorders and should therefore remain the focus of future research.

Feelings of shame can be difficult to manage. However, there has been limited research attention to the contextual factors which may make this particularly difficult, partly because shame has often been conceptualised as primarily an intrapsychic phenomenon. Alongside this, investigations of managing potentially shaming identities, often conceptualised as `stigma management', have generally not focused on emotion. To bridge this gap in the literature I draw on theoretical resources such as symbolic interactionism and social constructionism to conceptualise experiences of shame as constructed within interaction in particular social locations and from the discursive resources available. Within this framework I attempt to develop a more contextualisedu nderstandingo f managings hamet hrough two qualitative investigations. The first study exploresf ifty written accountso f experienceso f shame provided by a cross-section of students and employees at a British university. These are obtainedb y meanso f a questionnairew hich asksp articipantst o provide a narrative account of an experience of public shame and then answer several openended questions about this. The data is interrogated through two analyses. The first is a thematic analysis which explores the data as revealing something of the participants' subjective experiences. The second analysis draws on Foucauldian approachesto discoursea nalysisa nd treatst he data as a text which revealst he subject positionsm adea vailablef or resisting shame. The analysesp oint to the importanceo f a contextualiseda pproacht o understandingm anagings hame. In particular they illuminate the importance of positioning vis-ä-vis others, the use of shame as a form of social control, the varied meanings of shame and suggest that feelings of shame can be debilitating and difficult to manage, particularly where these are multi-layered and contradictory. The seconds tudy explorest he managemenot f potential shamei n relation to using mental health services through 22 interviews with service users. A thematic analysis of meaning-making which draws on both contextual constructionist and discursive approachesto data analysish ighlights the importanceo f the availability of both discursive and relational resources for avoiding and repairing shame. As for the first study, the results of the analysis suggest that some of the prior literature may have emphasised self-acceptance in repairing shame at the expense of theorising the need for a shift in the self-in-relation-to-others. I discuss the implications of these theoretical developments from both studies for mental health services and the development of approaches to helping individuals who struggle with shame which are not perniciously individualising and hence shaming. I also reflect on what the analyses indicate about the possibilities for combining methodologies which have apparently differing epistemological positions and the challenges involved in doing this.

This PhD is an investigation of vocal expressions of emotions, mainly focusing on non-verbal sounds such as laughter, cries and sighs. The research examines the roles of categorical and dimensional factors, the contributions of a number of acoustic cues, and the influence of culture. A series of studies established that naive listeners can reliably identify non-verbal vocalisations of positive and negative emotions in forced-choice and rating tasks. Some evidence for underlying dimensions of arousal and valence is found, although each emotion had a discrete expression. The role of acoustic characteristics of the sounds is investigated experimentally and analytically. This work shows that the cues used to identify different emotions vary, although pitch and pitch variation play a central role. The cues used to identify emotions in non-verbal vocalisations differ from the cues used when comprehending speech. An additional set of studies using stimuli consisting of emotional speech demonstrates that these sounds can also be reliably identified, and rely on similar acoustic cues. A series of studies with a pre-literate Namibian tribe shows that non-verbal vocalisations can be recognized across cultures. An fMRI study carried out to investigate the neural processing of non-verbal vocalisations of emotions is presented. The results show activation in pre-motor regions arising from passive listening to non-verbal emotional vocalisations, suggesting neural auditory-motor interactions in the perception of these sounds. In sum, this thesis demonstrates that non-verbal vocalisations of emotions are reliably identifiable tokens of information that belong to discrete categories. These vocalisations are recognisable across vastly different cultures and thus seem to, like facial expressions of emotions, comprise human universals. Listeners rely mainly on pitch and pitch variation to identify emotions in non verbal vocalisations, which differs with the cues used to comprehend speech. When listening to others' emotional vocalisations, a neural system of preparatory motor activation is engaged.