Binge Eating Disorder (BED) is characterized by recurrent episodes of binge eating without the use of compensatory behaviors. Functional accounts of BED propose that negative affect is an antecedent to binge eating because binge eating serves to alleviate negative affect. However, previous studies investigating the association between negative affect and binge eating have yielded inconsistent findings, perhaps due to individual vulnerability factors that moderate the effects of negative affect on binge eating behavior. As one candidate, the current study investigated emotion regulation strategies that may be implicated in the maintenance of binge eating in BED, particularly under conditions of negative affect: brooding rumination, distress tolerance, and mood-related expectancies for eating. These emotion regulation strategies were: a) compared in 38 women with BED vs. 36 non-eating disordered female controls, b) examined in relation to markers of current binge eating severity among BED women, and c) used as predictors of caloric intake and urge to eat in response to a personally-relevant dysphoric mood induction upon presentation of snack foods in a "taste task." Results revealed that women with BED endorsed higher brooding rumination, more positive expectancies that eating serves to ameliorate negative affect, and lower distress tolerance than controls. Among women with BED, higher brooding rumination was associated with greater binge eating severity, and stronger expectancies that eating reduces negative affect were associated with more frequent binge eating episodes and greater urge to eat in response to depression. Surprisingly, better distress tolerance was associated with more frequent binge eating episodes. Women with BED consumed more calories and reported greater loss of control as well as a greater sense of guilt in response to the taste task relative to control participants. Contrary to hypothesis, there were no direct or indirect effects of any of the three emotion regulation strategies on change in urge to eat or calories consumed on the taste task following sad mood induction in BED women. In controls, better distress tolerance and stronger expectancies that eating alleviates negative affect were associated with decreased caloric intake on the taste task after mood induction. Overall, these findings highlight the importance of considering trans-diagnostic processes in BED as well as the need to identify other theoretically-relevant factors that contribute to the cognitive and behavioral features of BED. Limitations and directions for future studies are discussed.

Objective: Theoretical and empirical evidence suggests a relationship between Repetitive Thought (RT, e.g. rumination), and depression in patients with Coronary Heart Disease (CHD). To date, cross-sectional studies indicate that rumination is associated with depression in CHD, but additional prospective longitudinal research is required to determine if rumination predicts subsequent depression. This research therefore aimed to test the hypothesis that RT, specifically rumination, is a vulnerability factor for depression over time in a CHD population. It was predicted that RT at baseline would predict depression rates at three month follow-up after controlling for baseline depression and potential confounding factors. Methods: Inpatients and outpatients with a diagnosis of CHD completed self-report questionnaires at baseline (N = 101) and at three month follow-up (N = 85). The data was analysed using a hierarchical multiple regression. Results: Baseline rumination significantly predicted depression at the three month follow-up after controlling for baseline depression and potential confounding factors. Rumination accounted for 8.3% of the variance (p< .001). Subscales of brooding and reflection were also found to be individually predictive of follow-up depression explaining 4% of the variance (p< .005) and 7% of the variance (p< .001) respectively. Conclusion: Findings are consistent with previous prospective and cross-sectional research that indicates that rumination plays a unique role in the maintenance of depression in CHD patients and is an identifiable vulnerability factor.

Objectives: Due to a number of conceptual and methodological limitations, existing research has provided only equivocal evidence that deficits/biases in attentional control (AC) are causally implicated in depressive rumination and/or that Cognitive Control Training (CCT) can be used to remediate such vulnerabilities. By using a well-validated training task and ensuring adequate training exposure, the current study aimed to examine the hypothesis that daily CCT would reduce rumination and improve mood among participants with elevated ruminative disposition. Method: Using a multiple baseline design (MBD), eight high-ruminating university participants rated their daily levels of rumination and mood before and after the randomly-determined introduction of daily CCT, designed to enhance their level of AC. Daily ratings were compared before and after the introduction of CCT, using systematic visual analysis and randomisation tests for significance at the group level. Results: No evidence was found to support the hypothesis that daily CCT reduces rumination and/or improve mood. While participants improved in their performance within the CCT across the training period, there was no evidence of near- or far-transfer, visual analysis revealed no impact of the introduction of daily training, and all group-level analyses were non-significant (p ≥ .05). Conclusion: Despite addressing a number of conceptual/methodological concerns, the current study provides no further support for AC theories of rumination or the use of CCT-based treatments for depression. Such conclusions must be interpreted in light of other methodological limitations, however, including the use of a non-clinical sample and the use of MBD to detect delayed treatment effects.

Research is increasingly attempting to understand the developmental nature of depressive symptomology and its links with executive functioning (EF), repetitive negative thinking (RNT) and stress (e.g., Snyder & Hankin, 2016). Prospective studies are needed to explore the potential mechanisms underlying these associations. This study investigated whether EFs can predict changes in RNT, stress and depressive symptoms during a period of stress. One hundred and two undergraduates completed questionnaires measuring life events, trait and state RNT, depressive and anxious symptoms as well as behavioural EF tasks of cognitive switching and inhibitory control at baseline (Time 1). Follow-up questionnaires of RNT, stress, depression and anxiety were gathered approximately two months later (Time 2), during students’ formal examinations, a period of naturally elevated stress. Findings indicated no association between EF and RNT, depression or anxiety but found that the interaction between high levels of trait RNT and low levels of EF (switching) at baseline was a significant predictor of change in state RNT under stress. Findings are discussed in light of current research attempting to unpick associations between EF, RNT and depression in young adults.

LITERATURE REVIEW: A systematic review of the associations between effortful control, repetitive negative thinking and depression in adolescence. ABSTRACT: Background: Repetitive negative thinking (RNT) and the self-regulatory temperament, effortful control (EC), have been found to be important risk factors for the development of depressive symptoms. Furthermore, adolescence has been found to be a period of increased risk for developing depressive symptoms. The relationships between these risk pathways are not well understood during this period of development. Objective: This systematic review aimed to evaluate the literature exploring the relationships RNT and EC have in accounting for depressive symptoms in adolescents. In particular, whether RNT and EC are associated with depressive symptoms, and whether EC moderates the effects of RNT on depressive symptoms. Methods: Three databases and key journals were searched for studies measuring EC, RNT and depressive symptoms in 10-20 year olds. Study selection was undertaken by applying inclusion and exclusion criteria. The methodological quality of the included studies was assessed using a validated checklist. Inter-rater reliability was calculated for a random subsample of the search. Results: Thirteen studies were selected for inclusion. There was evidence indicating that RNT was correlated with depressive symptoms both concurrently and prospectively. The evidence for a relationship between lower EC and higher levels of depressive symptoms was mixed. High quality studies concluded that EC and depressive symptoms are associated concurrently, but not over time. There is evidence that lower EC predicts RNT over time, and even spanning over childhood. Some evidence was found for EC as a moderator between RNT and depressive symptoms and this was also found when the relationship was prospective. Conclusions: Whilst the reviewed literature had many strengths, there were large differences in how EC in particular, was measured. This resulted in a challenge synthesising the results and making clear conclusions. Future research would benefit from considering self-report and behavioural measures, and recognising the potential impact of stressful life events. EMPIRICAL PAPER: Investigating associations between repetitive negative thinking, stress, and effortful control, and the development and maintenance of depression in adolescence: A follow-up study. ABSTRACT: Background: Adolescence is a period of increased vulnerability for depressive symptoms (Twenge & Nolen-Hoeksema, 2002). Given the impact of emotional disorders on an individual, it is important to understand risk factors, and conversely, protective factors to inform effective interventions. Repetitive negative thinking (RNT) and the self-regulatory temperament, effortful control (EC), have been found to be important risk factors for the development of depressive symptoms and require further exploration in adolescence. Objective: This study investigated whether RNT predicted changes in later depressive symptoms, and if so, whether this change was moderated by EC. The study examined these associations during emotional reactivity to a stressor (exams), and emotional recovery following the stressor. Methods: Two samples with similar designs and measures were combined in this study. Two-hundred-and-fifty-five females completed Baseline questionnaires measuring life events, RNT, EC, and depressive symptoms. One-hundred-and-ninety-nine participants were followed up prior to their exams (Pre-exam), and 115 participants were followed up after their exams (Post-exam). Results: The study first examined emotional reactivity to stress, finding that contrary to the literature, RNT did not predict depressive symptoms in response to stress, when controlling for Baseline depressive symptoms. EC did not significantly interact with RNT in predicting depressive symptoms. However, RNT was associated with emotional recovery from stress: RNT predicted levels of depressive symptoms following exams, when controlling for Pre-exam depressive symptoms. Furthermore, EC moderated this relationship, however contrary to the literature and predictions, this was not in the expected direction, with high levels of EC associated with high levels of depressive symptoms. Conclusions: These findings suggest that despite the strengths of the study design, including a large sample at Baseline and follow up over a period of stress, hypothesized associations were not found during emotional reactivity to stress, but hypothesized associations were found during emotional recovery from stress. Recommendations are made for future studies, including recruiting sufficient number of males to the study.

The Self-Regulatory Executive Function (S-REF) model (Wells & Matthew, 2015) states that psychological disorder results from an unhelpful thinking style called the Cognitive Attentional Syndrome (CAS). The CAS incorporates worry/rumination, threat monitoring and unhelpful thought control strategies and is controlled by erroneous metacognitive beliefs. The contribution of the S-REF model to the understating of rumination (paper one) and suicidal ideation (paper two) is evaluated within this thesis. Rumination is one component of the CAS. According to the S-REF model, its execution is guided by metacognitive beliefs. Paper one describes a systematic review and meta-analysis that was undertaken to establish the nature and strength of the relationships between rumination and metacognitive beliefs. Robust relationships, of moderate strength, were observed between rumination and beliefs about its benefits and between rumination and beliefs about its negative consequences. Future research should aim to delineate causality in the observed relationships and consider confounder variables. In paper two it is argued that the S-REF conceptualisation represents a promising new approach to the understanding of suicidal ideation. The importance of considering CAS processes was supported by the study findings. Worry and punishment-related thought control strategies, alongside rumination, predicted suicidal ideation. Distraction, social control and reappraisal strategies negatively predicted suicidal ideation. Some evidence that CAS processes were controlled by erroneous metacognitive beliefs was also obtained. However, this was not conclusive and should be considered again in a larger sample. Paper three describes the development of papers one and two in more detail, highlighting and justifying the important decisions made. Further reflections on methodology are also provided to demonstrate the learning achieved.

The objectives of the current experiment were to characterize patterns of daily rumination time, activity and milk production around the diagnosis of health disorders, and to determine if the addition of rumination and activity data to a commercial dairy farm fresh cow health monitoring program improves sick cow detection and diagnosis of disease during the first 30 DIM. Holstein animals (primiparous = 282, parous = 328) were enrolled in the experiment approximately 60 d before expected calving date, and were divided into two groups (Collar Monitoring-CM-, n=293 ; Control-C-, n=317). Electronic rumination and activity monitoring tags (SCR Engineers Ltd., Netanya, Israel) were fitted on cows neck at enrollment and were kept until approximately 80 ± 3 DIM. Farm personnel checked the cows and performed the diagnosis of disease following the routine of the dairy. Cows from both of the groups were sent to check based on the parameters used by the farm. Additionally, cows from group CM were checked based on the data provided by the tags. Serum calcium concentration was determined using blood samples collected from 0 to 4 DIM. BHBA concentration was determined twice using blood samples collected from 4 to 12 DIM and 7 to 20 DIM. Subclinical hypocalcemia (SCHC) and subclinical ketosis (SCK) were characterized as Ca <8.55 ng/dL, and BHBA >1,000 µmol/L in any blood sample, respectively. Daily rumination time (DRT), daily activity (ACT), and daily milk production patterns for cows with clinical disease showed differences with healthy cows around diagnosis (P $lt;0.05). Cows with subclinical disorders and calving problems had changes in DRT, ACT, and milk production patterns compared to healthy cows around calving (P <0.05). DRT and ACT patterns of regrouped cows were characterized by differences with non-regrouped cows around regrouping (P <0.05). No differences were found for DRT, ACT, and milk production between groups C and CM. The overall sensitivity (Se) of collars to identify health disorders was 56.4% (n = 402 cases), considering a positive outcome as at least 1 alert based on rumination and activity from -7 to +2 d relative to diagnosis. Se was higher for cows with more than one disorder (75.8%) than for cows with one disease only (45.5%) (P <0.001). No differences between groups were found for overall Se, and Se for cows with one disease. However, for cows with more than one disorder, Se was higher in group CM than C (P = 0.005). Overall specificity, positive predicted values, and negative predicted values were 74.5%, 46.4%, and 57.6%, respectively. The overall incidence of disease was 48%. No differences between groups were found for overall incidence of disease and each disease. Among primiparous, group CM (43.3%) had higher overall incidence of disease than group C (32.1%) (P = 0.05). Although were not differences for parous, incidence of metritis tended to be greater in group C than CM (P = 0.1). Incidence of SCK and SCHC was not different between groups. A higher percentage of animals from group CM than C received treatment (P = 0.04), and these differences were seen in primiparous (P = 0.03), but not in parous. However, a higher percentage of parous not diagnosed as sick from group CM received support treatments (drenching and fluids) compared to C. No differences were shown for culling rate, service rate until 150 DIM, conception rate at first service, and percentage of cows marked as do not breed between groups. DRT and ACT patterns for sick cows showed differences around diagnosis compared to healthy cows. The use of DRT and ACT data was able to identify sick cows in a commercial dairy farm. Results suggest that it may be also useful to identify cows needing attention before clinical signs are visible, improving the prevention of health disorders. Its usefulness may vary according to parity, disease, severity of disease and health compromise, and the intensity of the farm system for checking cows. Future research should evaluate different parameters and parameters thresholds based on rumination and activity data for identifying sick cows, and their efficiency in dairies with different degrees of intensity for checking animals health. / Os objetivos deste experimento foram caracterizar os padrões diários do tempo de ruminação, atividade, e produção de leite arredor do diagnostico de doenças, e determinar se a adição de dados de ruminação e atividade num programa de monitoramento de saúde de vacas de uma fazenda comercial melhora a detecção de vacas doentes e o diagnostico de doenças durante os primeiros 30 DEL. Animais Holstein (primíparas = 282, multíparas = 328) foram utilizados no experimento aproximadamente 60±3 dias antes da data esperada de parto, e foram divididos em dois grupos (Collar Monitoring-CM-, n=293; Control-C-, n=317). Dispositivos eletrônicos para o monitoramento da ruminação e atividade acoplados a colares (SCR Engineers Ltd., Netanya, Israel) foram colocados nas vacas no enrolamento e mantidos ate aproximadamente 80±3 DEL. O monitoramento de saúde das vacas e o diagnostico de doenças foram realizados pelos funcionários da fazenda seguindo a rotina do estabelecimento. Os animais dos dois grupos foram enviados para checagem de saúde baseados nos parâmetros utilizados pela fazenda. Adicionalmente, as vacas do grupo CM foram checadas baseadas na informação suprida pelos colares. A concentração de cálcio sérico foi determinada usando uma amostra de sangue coletada do dia 0 ao 4 em leite. A concentração de beta-hidroxibutirato (BHBA) foi determinada duas vezes usando amostras de sangue coletadas do dia 4 ao 12, e do 7 ao 20 do posparto. Hipocalcemia subclínica (SCHC) e cetose subclínica (SCK) foram caracterizadas como Ca <8.55 ng/dL, e BHBA >1000 µmol/L em qualquer amostra, respectivamente. Os padrões diários do tempo de ruminação (DRT), atividade (ACT), e produção de leite de vacas com doenças clinicas arredor do diagnostico mostraram diferencias comparados com vacas sadias (P <0.05). Vacas com alterações subclínicas e problemas de parto tiveram alterações nos padrões de DRT, ACT, e produção de leite arredor do parto, quando comparadas a vacas controle (P <0.05). Padrões de DRT e ACT de vacas reagrupadas se caracterizaram por diferencias com vacas não reagrupadas (P <0.05). Não foram achadas diferencias em DRT, ACT, e produção de leite entre os grupos C e CM. A sensibilidade (Se) dos colares para identificar problemas de saúde foi de 56.4% (n = 402 casos), considerando como evento positivo a ocorrência de pelo menos uma alerta baseada em ruminação e atividade dentro dos 7 dias prévios ate 2 dias apos o diagnostico de doença. A Se foi maior para vacas com mais de uma doença (75.8%) que em para vacas com uma doença somente (45.5%) (P <0.001). Não se acharam diferencias na Se geral, nem Se para vacas com uma doença somente entre grupos. Porem, a Se foi maior no grupo CM que no grupo C (P = 0.005) em vacas com mais de uma doença. A especificidade (Sp), valores da predição positiva (PPV), e valores da predição negativa (NPV) foram 74.5%, 46.4%, e 57.6%, respectivamente. A incidência de doença foi de 48%. Não houve diferencias entre grupos na incidência de doença, nem na incidência de cada doença. Entre as primíparas, o grupo CM (43.3%) teve maior incidência de doença do que o grupo C (32.1%) (P = 0.05). Embora não teve diferencia na incidência de doença entre grupos para multíparas, a incidência de metrite teve uma tendência a ser maior no grupo C do que no grupo CM (P = 0.1). A incidência de SCK e SCHC não foi diferente entre grupos. Maior percentagem de animais do grupo CM recebeu tratamento do que do grupo C (P = 0.04), e estas diferencias foram observadas em primíparas (P = 0.03), mas não em multíparas. Contudo, uma maior percentagem de animais não diagnosticados como doentes do grupo CM recebeu tratamentos de suporte, quando comparado ao grupo C. Não se acharam diferencias na taxa de descarte, taxa de serviço aos 150 DEL, taxa de concepção ao primeiro serviço, e percentagem de vacas de descarte reprodutivo entre grupos. Resumindo, os padrões de DRT e ACT de vacas doentes arredor do diagnostico de doença mostraram diferencias comparados com os de vacas sadias. O uso da informação de DRT e ACT foi capaz de identificar vacas doentes numa fazenda comercial. Os resultados sugerem que a utilização dos colares pode ser util para identificar vacas com necessidade de atenção antes da aparição de sinais clínicos visíveis, melhorando a prevenção de problemas de saúde. A utilidade da utilização do sistema pode variar de acordo a ordem de partos dos animais, doença em questão, severidade da doença e comprometimento de saúde do animal, e com a intensidade do sistema de monitoramento de saúde dos animais da fazenda. Próximas pesquisas deveriam avaliar diferentes parâmetros baseados na informação de ruminação e atividade para identificar vacas doentes, e a sua eficiência em fazendas com diferentes graus de intensidade para o monitoramento de saúde.

Self-compassion and self-forgiveness appear to have much in common, as both relate to one’s self-regard during challenging circumstances; however, their empirical relationship is largely yet to be explored. The present study examines theoretical and empirical areas of overlap and proposes a theory of their relationship, including its possible impact on health. Self-compassion and self-forgiveness were proposed to have a direct relationship that may be mediated by reduction in rumination, shame, and experiential avoidance. These factors together were also hypothesized to have a positive impact on health functioning. The current study tested these models in a sample of undergraduate students (n = 199). In parallel mediation analysis, selfcompassion and self-forgiveness were related to one another and this association was partially mediated by shame, only. As such, neither rumination nor experiential avoidance were included in subsequent analyses. In serial mediation analyses, self-compassion, shame, and selfforgiveness were found to affect health outcomes in various ways. For psychological distress, the association of self-compassion was partially mediated by shame and self-forgiveness, in an indistinguishable fashion. For mental health status, self-compassion was found to have a significant direct effect, only. For both somatic symptoms (full mediation) and physical health status (indirect only effect), the association of self-compassion operated through self-forgiveness both alone and linked with shame. As these two emerging areas in the psychological literature continue to mature, researchers should prioritize both integration between and nuances within these constructs in order to develop a more complete understanding of self-compassion, selfforgiveness, and their implications for health.

Motor-Vehicle crashes are the leading cause of death for teens and young adults. Research and public interventions have primarily examined the impact of external factors related to driving; however, less work has examined internal factors. Limited research has shown a negative association between trait forgiveness of others and both driving anger and driving aggression. The current study replicates previous findings and expands to include multiple dimensions of forgiveness and adverse driving outcomes as a dependent variable. It was predicted that multiple dimensions of forgiveness would be directly and indirectly related to adverse driving outcomes through the mediators of anger rumination, stress, and dangerous driving. Undergraduate students (N=759) at a regional university completed a series of self-report questionnaires online examining driving anger, driving aggression, multiple dimensions of forgiveness, adverse driving outcomes, anger rumination, stress, and dangerous driving behaviors. Hierarchical multiple regression analyses were used to replicate previous findings (analysis 1) and multiple serial mediations as expansion (analysis 2). In replication, trait forgiveness of others was shown to have a negative bivariate correlation with driving anger and driving aggression and to be a significant predictor of driving aggression above that of driving anger (analysis 1). Multiple serial mediation demonstrated an indirect only effect of multiple dimensions of forgiveness on adverse driving outcomes through the various mediators (analysis 2); however, varied relationships were observed. As a result, forgiveness of self and of uncontrollable situations demonstrated a significant negative effect on adverse driving outcomes through the various mediators. However, although, forgiveness of others was found to have a significant negative effect through anger rumination and dangerous driving behaviors in serial, it demonstrated a positive effect with stress as a mediator. The results support and replicate previous research and demonstrate a significant indirect only effect of multiple dimensions of forgiveness on adverse driving outcomes through the current mediators. The relationships were varied, however. Therefore, multiple dimensions of forgiveness continue to be meaningful variables related to driving anger, driving aggression, and adverse driving outcomes.

Previous research supports the idea that individuals high in trait anger tend to experience more hostile attribution bias. According to the Integrative Cognitive Model, cognitive factors, such as rumination, may increase the risk of hostile attribution bias and any subsequent aggressive behaviors. Sex differences are apparent in rumination and anger expression. The present research explored the potential role trait rumination plays in hostile attribution bias as well as potential conditional effects of sex on this relationship. Participants were asked to complete a number of self-report measures and vignettes of ambiguously hostile situations adapted to improve reliability. Hypotheses were largely supported and trait anger rumination was significantly predictive of hostile attribution bias; however, conditional effects of sex were non-significant. The adapted hostile attribution bias measure had improved reliability and may have utility for a survey-based method to assessing hostile attribution bias. Findings may further our understanding of hostile interpretations and potential for subsequent aggressive behaviors in high trait ruminators in ambiguous situations as well as lead to potential areas of intervention to reduce anger and anger rumination.