UCLA

Understanding emotion dynamics is important in order to ascertain transdiagnostic mechanisms underlying anxiety and mood disorders. Emotional inertia, the stability of emotion over time, and emotion network density, a measure of the interconnectedness of emotions, both represent the persistence of emotions over time. Previous literature suggests that emotional persistence, particularly of negative emotions, is related to depression, anxiety, and difficulty with emotion regulation. However, this research is limited in that studies examining idiographic network density, longitudinal relationships, and transdiagnostic, continuous, or objective outcome measures are scarce. Additionally, existing studies present mixed findings on the implications of positive emotional persistence. This dissertation contains three studies designed to address these shortcomings and examine relationships among emotional persistence and transdiagnostic psychological well-being in adults with moderate-to-severe depression, with the ultimate goal of illuminating mechanisms underlying anxiety and depressive disorders such that current treatments may be improved.Study 1 examined both cross-sectional and longitudinal relationships among emotional inertia, emotion network density, and self-reported well-being. Inertia and density were both calculated using self-reported affect ratings from ecological momentary assessment (EMA) surveys administered five times per day in two eight-day epochs. Inertia and density were calculated separately for negative and positive affect. Positive contemporaneous network density was positively associated with well-being six weeks later. Positive emotional inertia was inversely associated with well-being six weeks later. Neither negative inertia nor negative network density was associated with subsequent well-being, and emotional persistence was not related to concurrent reports of well-being.Study 2 examined cross-sectional associations between emotional persistence and heart-rate variability. Inertia and density were again calculated using EMA surveys, this time from three eight-day epochs. Heart-rate variability was measured via Apple Watch both during the day and at bedrest throughout all three epochs. Negative contemporaneous network density was inversely associated with both median daytime and median bedrest heart-rate variability. That is, participants with more densely connected negative emotions demonstrated lower heart-rate variability. Positive emotional persistence was not related to heart-rate variability.Study 3 examined relationships between emotional persistence and linguistic indicators of well-being. Measures of emotional persistence were the same as in Study 2. Natural linguistic processing was used to analyze transcripts of a ten-minute structured interview where participants answered questions about their previous week. Probability of both positive and negative sentiment was obtained using natural linguistic processing, and tendency towards positive sentiment was operationalized as the participants’ average difference between probability of positive and negative sentiment. Tendency towards positive sentiment was examined in relation to positive and negative emotional persistence. No metric of emotional persistence was associated with linguistic sentiment. However, participants who reported higher levels of average positive affect via EMA demonstrated higher tendency towards positive sentiment during the interview.Taken together, these findings provide insight into the implications of positive and negative emotional persistence. Emotional inertia and emotion network density seem to represent different, and at times opposing, presentations of emotional persistence. Broadly, this dissertation supports the literature suggesting that denser negative emotion networks are maladaptive and denser positive emotions reflect better psychological well-being. Contrasting results for negative and positive emotional persistence suggest that persistence of differently valenced emotions may relate to different components of psychological and physiological well-being. This study also highlights the methodological challenges of using objective measures of emotional functioning such as natural linguistic processing. Future research should continue to address these methodological limitations and investigate the mechanisms underlying the relationship between emotional persistence and psychological well-being to inform clinical interventions.