Neurocomputational mechanisms of reward-based online mood regulation in adolescents with bipolar disorder and major depressive disorder

Backgrounds: The overlapping symptoms between bipolar disorder (BD) in the depressive period and major depressive disorder (MDD) pose a substantial challenge in diagnosis and treatment. A prevailing hypothesis suggests that mood dysregulation may be linked to impairments in the dopamine reward system in the brain, but the underlying neurocomputational differences between BD and MDD remain elusive. In this study, we investigate whether atypical reward processing affects subjective mood in adolescents with BD and MDD. Our findings aim to elucidate the behavioral and neural differences between the two groups, facilitating more accurate and timely diagnosis and intervention.Methods: Forty-five adolescents (aged ≤ 19 years) diagnosed with BD (N = 25) or MDD (N = 20) were asked to complete a risky gambling task while their brain responses were recorded using functional magnetic resonance imaging (fMRI). Several computational models were constructed to uncover the associations between various reward components (e.g., reward prediction errors, RPE) and trial-wise fluctuation in subjective mood during the gambling task.Results: We found that adolescents with BD exhibited a greater behavioral propensity for uncertain options, as compared to those with MDD. Computational modeling and mediation analysis suggest a triple relationship between RPE-mood association, decision rationality, and symptom severity. Using fMRI, we further observed distinct brain response patterns in a distributed mood regulation network between adolescents with BD and MDD.Conclusions: Our findings highlight the critical role of RPE in mood regulation and suggest more potential engagement of the model-free control system in BD as compared to MDD. These results provide new insights into the diagnosis and rehabilitation of BD and MDD in adolescents.