Joelle LeMoult

Loneliness is common and, while generally transient, persists in up to 22% of the population. The rising prevalence and adverse impacts of chronic loneliness underline the need to understand its underlying mechanisms. Evolutionary models implicate negative interpretation biases towards social information in chronic loneliness, but not in transient or state loneliness. Yet little research has compared these constructs, investigated whether biases are specific to social stimuli, or examined the roles of anxiety or depression in these associations. The present study examined negative interpretation biases towards socially relevant and non-social stimuli in chronic and state loneliness and tested whether anxiety and depression moderate these associations. These aims were explored in 563 undergraduate students who participated in a two-hour online session. First, participants completed an interpretation bias task where they rated socially relevant and non-social ambiguous images as positive or negative. Interpretation bias was calculated as percent negative ratings. Next, participants underwent a state loneliness induction and completed the interpretation bias task a second time. Finally, participants self-reported chronic loneliness, anxiety, and depression. Linear mixed models were used to test hypotheses that chronic, but not state, loneliness would predict more negative ratings of social stimuli, with associations moderated by anxiety or depressive symptoms. Contrary to hypotheses, only state loneliness was associated with more negative ratings of social stimuli, with greater anxiety symptoms predicting more negative ratings of non-social stimuli. Implications of these findings for present theoretical models of loneliness are discussed.

View record

Rumination is a pattern of passive and repetitive thoughts about symptoms of one’s distress, as well as its causes and consequences. While rumination has been linked to onset and exacerbation of depression, recent research suggests that only one subtype of rumination, brooding, and not the other type, reflection, is responsible for these effects. Theoretical perspectives and empirical evidence suggest that biases in inhibition, shifting, and updating information in working memory underlie rumination. However, to date, we do not know which cognitive bias contributes the most to rumination, and more specifically, to brooding and reflection. In this study, I aimed to address this gap in the literature using an experience sampling methodology. To this end, I invited 267 undergraduate students to the lab for a baseline session. During this session, they rated their level of depression, their tendency to ruminate, and their affect. They also completed three cognitive tasks, Emotional Stroop Task, Affective Switching Task, and Emotional 2-Back task, that assessed their inhibition, shifting and updating biases, respectively. Next, participants rated their level of brooding, reflection, and negative affect at nine different time points within 48 hours of their most stressful midterm exam. I hypothesized that updating bias would be the main cognitive control bias to predict the level and trajectory of brooding after the exam. I further hypothesized that the level of brooding, and not reflection, at each time point would predict the intensity of negative affect at the following time point. I also conducted additional exploratory analyses to examine (1) the association of inhibition, shifting, and updating biases with reflection; and (2) the bidirectional association between rumination and negative affect. Contrary to my hypothesis, inhibition and shifting biases predicted brooding after the exam. Inhibition also played a role in reflection. Furthermore, brooding, and not reflection, predicted negative affect at the next time point. In addition, negative affect predicted both brooding and reflection at the next time point. The role of cognitive control biases in brooding and reflection, and its implications, are discussed.

View record

The process by which emotional experiences are managed is known as emotion regulation. Two types of emotion regulation strategies are commonly compared: rumination (focusing on one’s problems and feelings) and distraction (focusing away from one’s problems and feelings). Whereas rumination typically increases negative affect, distraction typically decreases negative affect. Past research has focused on emotion regulation as an intrapersonal endeavor (managing one’s own emotions), whereas interpersonal emotion regulation (IER; receiving support from another person to regulate one’s emotions) lacks the same degree of investigation. This study sought to compare the effects of intrapersonal emotion regulation (rumination, distraction) and IER (co-rumination, co-distraction) on affect and relationship quality and closeness. Participants completed the Fast Friends paradigm; following, participants privately recalled a stressful event. Finally, participants were randomized into one of four emotion regulation conditions: rumination, distraction, co-rumination, or co-distraction. Affect and relationship quality and closeness measures were completed throughout the study session. I predicted that rumination and co-rumination would increase negative affect compared to distraction and co-distraction. I also predicted that co-distraction would decrease negative affect more than distraction. Finally, I predicted that co-rumination would provide the highest relationship quality and closeness compared to all conditions. Results showed that negative affect did not differ after the emotion regulation conditions. However, negative affect decreased significantly more for participants in the distraction condition compared to participants in the rumination and co-rumination conditions. Finally, there were no differences in relationship quality and closeness across conditions. The limitations and implications of this study are discussed.

View record

Depression and anxiety disorders are the most prevalent classes of mental illness in youth and are associated with severe long-term impairment. Thus, a thorough understanding of factors contributing to the development of these disorders is paramount. Evidence indicates that maladaptive biological responses to stress increase risk for the onset of depression and anxiety, but the mechanisms underlying individual differences in stress responsivity are unclear. Importantly, evidence suggests that individual differences in cognitive disengagement biases, or difficulty disengaging from valenced material, may be associated with individual differences in biological responses to stress. The current study extends past research by investigating, for the first time, whether difficulty cognitively disengaging from valenced stimuli is associated with individual differences in biological responses to stress in a sample of preadolescents. A sample of 43 preadolescents completed two computer-based tasks to assess biases in both attentional disengagement and working memory disengagement. In addition, stress responsivity was indexed by levels of both cortisol and alpha-amylase in response to a psychosocial stressor, the Trier Social Stress Test for Children (TSST-C; Buske-Kirschbaum et al., 1997). It was hypothesized that disengagement biases for negative stimuli would be significantly associated with prolonged biological recovery from stress. Findings indicated that both attentional and working memory disengagement biases were associated with baseline levels of cortisol and alpha-amylase, and with the slope of biological recovery from stress; however, the direction of some findings were unexpected. In addition, working memory biases were associated with the slope of alpha-amylase reactivity to stress. The current results contribute to theoretical models of the relation between cognition and biological responses by identifying potential candidate cognitive mechanisms that are associated with individual differences in cortisol and alpha-amylase responses to stress. The current study also presents an important methodological advancement in the literature by examining the relative contribution of both attentional and working memory disengagement biases to the prediction of both neuroendocrine and autonomic nervous system responses to stress. Understanding the association between cognition and biological responses to stress may facilitate the development of personalized, transdiagnostic treatments for youth who show biological stress profiles that increase their risk of developing depression or anxiety.

View record

Depression is one of the most common psychiatric illnesses, and is the leading cause of disability worldwide. Research suggests that stress and subsequent responses to stress play a central role in exacerbating depressive symptoms and prolonging depressive episodes. It is, therefore, important to understand the factors that may be hindering recovery from stress, and to test what can promote effective recovery from stress. One adaptive response to stress is self-compassion. Numerous studies have demonstrated an inverse association between self-compassion and depressive symptoms, with a recent meta-analysis finding a large mean effect size (MacBeth & Gumley, 2012). Subsequently, recent research has suggested that self-compassion may be a resiliency factor that protects against both the development and maintenance of depressive episodes (Ehret, Joormann, & Berking, 2015). The extant literature, however, is limited by its reliance on correlational design and self-report data. The goal of the current study was to extend previous research by looking at the effect of experimentally induced self-compassion on both emotional and biological recovery from stress in depression. Participants experiencing elevated depressive symptoms completed a standardized psychosocial stressor – the Trier Social Stress Test (Kirschbaum, Pirke, & Hellhammer, 1993) – and were randomly assigned to one of two stress-response inductions: self-compassion or a no-strategy control condition. It was hypothesized that the self-compassion induction would be significantly more effective than the control condition at promoting recovery from stress, as indicated by self-report measures of affect and measurements of salivary cortisol, the primary stress hormone. Results suggested that the self-compassion induction was more effective than the control in reducing anxious affect immediately after the induction. However, the self-compassion induction did not have an effect on recovery from stress as measured by levels of depressed affect or salivary cortisol. By investigating self-compassion, the current study has the potential to improve our understanding of factors that promote psychobiological recovery from stress in depression.

View record