Characterizing emotion dynamics in remitted depression: A network approach using ecological momentary assessment

This is the authors' abstract. We don't add key points for this paper.

Background: Maladaptive emotion dynamics and processes, such as emotional inflexibility and dominance of negative emotions, are characteristic of depression. The extent to which these abnormalities persist following depressive episodes, and represent a vulnerability factor for recurrent depressive episodes, remains unknown. The current study investigated whether emotion dynamics predict depressive symptoms in individuals with remitted depression (rMDD) and healthy controls (HC). Methods: 98 adults (HC: n = 50; rMDD: n = 48) completed a three-week ecological momentary assessment protocol, in which they responded to two items probing positive emotions and three items probing negative emotions six times daily. Contemporaneous and temporal networks were constructed using multilevel vector autoregressive models. Density was calculated as a measure of emotional inflexibility and In- and Out-Expected Influence were calculated as measures of centrality. Linear regression models examined if density predicted clinical outcomes at the 6-month follow-up assessment. Results: Individuals with rMDD had significantly denser temporal emotional networks than HC participants. Groups also showed differential patterns of the most influential nodes in temporal and contemporaneous networks. Greater temporal density and contemporaneous density was significantly associated with increased depressive symptoms at the 6-month follow-up, assessed through both clinician-rated and self-report measures. Conclusions: Abnormalities in emotion dynamics persist following remission from depression and can be used to predict future depressive symptoms, suggesting these may represent a vulnerability factor for depression. Future research should study if interventions based on emotion networks targeting maladaptive emotional processes are able to prevent future depression.