Background: Major depressive disorder is associated with altered social functioning and impaired learning, on both the behavioural and the neural level. These deficits are likely related, considering that successful social interactions require learning to predict other people's emotional responses. Yet, there is little research examining this relation.
Methods: Forty-three individuals with high (HD; N = 21) and low (LD; N = 22) depression scores answered questions regarding their real-life social experiences and performed a social learning task during fMRI scanning. As part of the task, subjects learned associations between name cues and rewarding (happy faces) or aversive (fearful faces) social outcomes. Using computational modelling, behavioural and neural correlates of social learning were examined and related to real-life social experiences.
Results: HD participants reported reduced motivation to engage in real-life social activities and demonstrated elevated uncertainty about social outcomes in the task. Moreover, HD subjects displayed altered encoding of social reward predictions in the insula, temporal lobe and parietal lobe. Interestingly, across all subjects, higher task uncertainty and reduced parietal prediction encoding were associated with decreased motivation to engage in real-life social activities.
Limitations: The size of the included sample was relatively small. The results should thus be regarded as preliminary and replications in larger samples are called for.
Conclusion: Taken together, our findings suggest that reduced learning from social outcomes may impair depressed individuals' ability to predict other people's responses in real life, which renders social situations uncertain. This uncertainty, in turn, may contribute to reduced social engagement (motivation) in depression.
Keywords: Computational modelling; Depression; Faces; Learning; Social; fMRI.
Copyright © 2019 Elsevier B.V. All rights reserved.