Impaired social learning in patients with major depressive disorder revealed by a reinforcement learning model

Background/objective: Patients with major depressive disorder (MDD) have altered learning rates for rewards and losses in non-social learning paradigms. However, it is not well understood whether the ability to learn from social interactions is altered in MDD patients. Using reinforcement learning during the repeated Trust Game (rTG), we investigated how MDD patients learn to trust newly-met partners in MDD patients. Method: Sixty-eight MDD patients and fifty-four controls each played as ‘investor’ and interacted with ten different partners. We manipulated both the level of trustworthiness by varying the chance of reciprocity (10, 30, 50, 70 and 90%) and reputation disclosure, where partners’ reputation was either pre-disclosed or hidden. Results: Our reinforcement learning model revealed that MDD patients had significantly higher learning rates for losses than the controls in both the reputation disclosure and non-disclosure condition. The difference was larger when reputation was not disclosed than disclosed. We observed no difference in learning rates for gains in either condition. Conclusions: Our findings highlight that abnormal learning for losses underlies the social learning process in MDD patients. This abnormality is higher when situational unpredictability is high versus low. Our findings provide novel insights into social rehabilitation of MDD. (AU)

Impaired social learning in patients with major depressive disorder revealed by a reinforcement learning model.

Background/objective: Patients with major depressive disorder (MDD) have altered learning rates for rewards and losses in non-social learning paradigms. However, it is not well understood whether the ability to learn from social interactions is altered in MDD patients. Using reinforcement learning during the repeated Trust Game (rTG), we investigated how MDD patients learn to trust newly-met partners in MDD patients. Method: Sixty-eight MDD patients and fifty-four controls each played as 'investor' and interacted with ten different partners. We manipulated both the level of trustworthiness by varying the chance of reciprocity (10, 30, 50, 70 and 90%) and reputation disclosure, where partners' reputation was either pre-disclosed or hidden. Results: Our reinforcement learning model revealed that MDD patients had significantly higher learning rates for losses than the controls in both the reputation disclosure and non-disclosure condition. The difference was larger when reputation was not disclosed than disclosed. We observed no difference in learning rates for gains in either condition. Conclusions: Our findings highlight that abnormal learning for losses underlies the social learning process in MDD patients. This abnormality is higher when situational unpredictability is high versus low. Our findings provide novel insights into social rehabilitation of MDD.

Amygdala connectivity related to subsequent stress responses during the COVID-19 outbreak.

Introduction: The amygdala plays an important role in stress responses and stress-related psychiatric disorders. It is possible that amygdala connectivity may be a neurobiological vulnerability marker for stress responses or stress-related psychiatric disorders and will be useful to precisely identify the vulnerable individuals before stress happens. However, little is known about the relationship between amygdala connectivity and subsequent stress responses. The current study investigated whether amygdala connectivity measured before experiencing stress is a predisposing neural feature of subsequent stress responses while individuals face an emergent and unexpected event like the COVID-19 outbreak. Methods: Data collected before the COVID-19 pandemic from an established fMRI cohort who lived in the pandemic center in China (Hubei) during the COVID-19 outbreak were used to investigate the relationship between amygdala connectivity and stress responses during and after the pandemic in 2020. The amygdala connectivity was measured with resting-state functional connectivity (rsFC) and effective connectivity. Results: We found the rsFC of the right amygdala with the dorsomedial prefrontal cortex (dmPFC) was negatively correlated with the stress responses at the first survey during the COVID-19 outbreak, and the rsFC between the right amygdala and bilateral superior frontal gyri (partially overlapped with the dmPFC) was correlated with SBSC at the second survey. Dynamic causal modeling suggested that the self-connection of the right amygdala was negatively correlated with stress responses during the pandemic. Discussion: Our findings expand our understanding about the role of amygdala in stress responses and stress-related psychiatric disorders and suggest that amygdala connectivity is a predisposing neural feature of subsequent stress responses.

Effects of 12-week escitalopram treatment on resting-state functional connectivity of large-scale brain networks in major depressive disorder.

In this study, the effects of antidepressants on large-scale brain networks and the neural basis of individual differences in response were explored. A total of 41 patients with major depressive disorder (MDD) and 42 matched healthy controls (HCs) were scanned by resting-state functional magnetic resonance imaging separately at baseline and after a 12-week follow-up. The patients with MDD received escitalopram for 12 weeks. After treatment, patients were classified into those with MDD in remission [MDDr, endpoint 17-item Hamilton Depression Rating Scale (HAMD) total score ≤7] and those in nonremission (MDDnr). The human Brainnetome Atlas was used to define large-scale networks and compute within- and between-network resting-state functional connectivity (rsFC). Results showed the decreased subcortical network (SCN)-ventral attention network (VAN) connectivity at baseline increased in patients with MDD after 12-week treatment, and it was comparable with that of HCs. This change was only observed in patients with MDDr. However, the decreased within-network rsFC in SCN and default mode network (DMN) persisted in all patients with MDD, including those with MDDr and MDDnr, after treatment. The strength of SCN-VAN connectivity at baseline was significantly negatively correlated with the reduction rate of HAMD score in all patients with MDD. Thus, SCN-VAN connectivity may be an antidepressant target associated with depressive state changes and a predictor of treatment response to serotonin reuptake inhibitors. The within-network rsFC in SCN and DMN may reflect a trait-like abnormality in MDD. These findings provide further insights into the mechanism of antidepressants and their individual differences in response. The trial name is "Appropriate technology study of MDD diagnosis and treatment based on objective indicators and measurement" (URL: http://www.chictr.org.cn/showproj.aspx?proj=21377; registration number: ChiCTR-OOC-17012566).

The perception-behavior dissociation in the ultimatum game in unmedicated patients with major depressive disorders.

Patients with major depressive disorders (MDD) exhibit social dysfunction, as illustrated by the lower acceptance rate of unfair proposals in the Ultimatum Game (UG) among patients with MDD compared with a control group. However, the underlying mechanisms remain largely unknown. In this study, we explored whether patients with MDD had altered perceptions of fairness or altered perception-behavior linkages compared with the control group, using a multilevel moderated mediation framework. Sixty-eight unmedicated patients with MDD and 55 members of a control group were recruited. Using generalized linear mixed effects models and multilevel structural equation modeling, we investigated the differences in the linkages between fairness level, fairness perception, and acceptance behavior among the two groups playing the UG. The results showed that the patients with MDD had a lower acceptance rate of unfair proposals than the control group. Fairness perception mediated the relationship between fairness level and acceptance behavior for both groups of participants when they played with human proposers but not computer proposers. The mediation effect was stronger among the control group than among the MDD patients. The linkage between fairness perception and acceptance behavior was attenuated among the patients with MDD compared with the control group. In conclusion, MDD patients were impaired in their ability to flexibly adjust acceptance behavior based on fairness perception in social interactions. (PsycInfo Database Record (c) 2022 APA, all rights reserved).