Social reinforcement learning as a predictor of real-life experiences in individuals with high and low depressive symptomatology

The challenge of learning adaptive mental behavior

Many psychotherapies aim to help people replace maladaptive mental behaviors (such as those leading to unproductive worry) with more adaptive ones (such as those leading to active problem solving). Yet, little is known empirically about how challenging it is to learn adaptive mental behaviors. Mental behaviors entail taking mental operations and thus may be more challenging to perform than motor actions; this challenge may enhance or impair learning. In particular, challenge when learning is often desirable because it improves retention. Yet, it is also plausible that the necessity of carrying out mental operations interferes with learning the expected values of mental actions by impeding credit assignment: the process of updating an action's value after reinforcement. Then, it may be more challenging not only to perform-but also to learn the consequences of-mental (vs. motor) behaviors. We designed a task to assess learning to take adaptive mental versus motor actions via matched probabilistic feedback. In two experiments (N = 300), most participants found it more difficult to learn to select optimal mental (vs. motor) actions, as evident in worse accuracy not only in a learning but also test (retention) phase. Computational modeling traced this impairment to an indicator of worse credit assignment (impaired construction and maintenance of expected values) when learning mental actions, accounting for worse accuracy in the learning and retention phases. The results suggest that people have particular difficulty learning adaptive mental behavior and pave the way for novel interventions to scaffold credit assignment and promote adaptive thinking. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Reinforcement learning of altruistic punishment differs between cultures and across the lifespan

Altruistic punishment is key to establishing cooperation and maintaining social order, yet its developmental trends across cultures remain unclear. Using computational reinforcement learning models, we provided the first evidence of how social feedback dynamically influences group-biased altruistic punishment across cultures and the lifespan. Study 1 (n = 371) found that Chinese participants exhibited higher learning rates than Americans when socially incentivized to punish unfair allocations. Additionally, Chinese adults showed slower learning and less exploration when punishing ingroups than outgroups, a pattern absent in American counterparts, potentially reflecting a tendency towards ingroup favoritism that may contribute to reinforcing collectivist values. Study 2 (n = 430, aged 12-52) further showed that such ingroup favoritism develops with age. Chinese participants' learning rates for ingroup punishment decreased from adolescence into adulthood, while outgroup rates stayed constant, implying a process of cultural learning. Our findings highlight cultural and age-related variations in altruistic punishment learning, with implications for social reinforcement learning and culturally sensitive educational practices promoting fairness and altruism.

Loneliness is associated with more trust but worse trustworthiness expectations

Subjective feelings of loneliness emerge due to unsatisfactory social relationships, representing a major risk for mental and physical well-being. Despite its social nature, evidence on how loneliness affects social behaviours and expectations is lacking. Using Bayesian analyses and economic games, we show in three different studies that lonelier individuals trusted their partners to a greater extent despite less favourable trustworthiness expectations, showing a greater discrepancy between their trusting behaviours and their expectations of others' trustworthiness. Such discrepancy was reversed in extravert individuals who also reported to be less lonely. These results provide evidence on two opposing effects of loneliness as a motivator for social connections and promoter of social withdrawal, and demonstrate the moderating role of personality traits. This work contributes to a better understanding of how loneliness impacts social behaviour and social expectations, with important downstream clinical implications for varying health conditions associated with heightened feelings of loneliness.

Synaptic pruning during adolescence shapes adult social behavior in both males and females

Evolutionarily conserved, peer-directed social behaviors are essential to participate in many aspects of human society. These behaviors directly impact psychological, physiological, and behavioral maturation. Adolescence is an evolutionarily conserved period during which reward-related behaviors, including social behaviors, develop via developmental plasticity in the mesolimbic dopaminergic "reward" circuitry of the brain. The nucleus accumbens (NAc) is an intermediate reward relay center that develops during adolescence and mediates both social behaviors and dopaminergic signaling. In several developing brain regions, synaptic pruning mediated by microglia, the resident immune cells of the brain, is important for normal behavioral development. We previously demonstrated that during adolescence, in rats, microglial synaptic pruning shapes the development of NAc and social play behavior in males and females. In this report, we hypothesize that interrupting microglial pruning in NAc during adolescence will have persistent effects on male and female social behavior in adulthood. We found that inhibiting microglial pruning in the NAc during adolescence had different effects on social behavior in males and females. In males, inhibiting pruning increased familiar exploration and increased nonsocial contact. In females, inhibiting pruning did not change familiar exploration behavior but increased active social interaction. This leads us to infer that naturally occurring NAc pruning serves to reduce social behaviors toward a familiar conspecific in both males and females.

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.

Reduced effects of social feedback on learning in Turner syndrome

Turner syndrome is a genetic condition caused by a complete or partial loss of one of the X chromosomes. Previous studies indicate that Turner syndrome is associated with challenges in social skills, but the underlying mechanisms remain largely unexplored. A possible mechanism is a reduced social influence on learning. The current study examined the impact of social and non-social feedback on learning in women with Turner syndrome (n = 35) and a sex- and age-matched control group (n = 37). Participants were instructed to earn points by repeatedly choosing between two stimuli with unequal probabilities of resulting in a reward. Mastering the task therefore required participants to learn through feedback which of the two stimuli was more likely to be rewarded. Data were analyzed using computational modeling and analyses of choice behavior. Social feedback led to a more explorative choice behavior in the control group, resulting in reduced learning compared to non-social feedback. No effects of social feedback on learning were found in Turner syndrome. The current study thus indicates that women with Turner syndrome may be less sensitive to social influences on reinforcement learning, than the general population.

Impaired microglia-mediated synaptic pruning in the nucleus accumbens during adolescence results in persistent dysregulation of familiar, but not novel social interactions in sex-specific ways

Evolutionarily conserved, peer-directed social behaviors are essential to participate in many aspects of human society. These behaviors directly impact psychological, physiological, and behavioral maturation. Adolescence is an evolutionarily conserved period during which reward-related behaviors, including social behaviors, develop via developmental plasticity in the mesolimbic dopaminergic 'reward' circuitry of the brain. The nucleus accumbens (NAc) is an intermediate reward relay center that develops during adolescence and mediates both social behaviors and dopaminergic signaling. In several developing brain regions, synaptic pruning mediated by microglia, the resident immune cells of the brain, is important for normal behavioral development. In rats, we previously demonstrated that microglial synaptic pruning also mediates NAc and social development during sex-specific adolescent periods and via sex-specific synaptic pruning targets. In this report, we demonstrate that interrupting microglial pruning in NAc during adolescence persistently dysregulates social behavior towards a familiar, but not novel social partner in both sexes, via sex-specific behavioral expression. This leads us to infer that naturally occurring NAc pruning serves to reduce social behaviors primarily directed toward a familiar conspecific in both sexes, but in sex-specific ways.

Social feedback enhances learning in Williams syndrome

Williams syndrome (WS) is a rare genetic condition characterized by high social interest and approach motivation as well as intellectual disability and anxiety. Despite the fact that social stimuli are believed to have an increased intrinsic reward value in WS, it is not known whether this translates to learning and decision making. Genes homozygously deleted in WS are linked to sociability in the general population, making it a potential model condition for understanding the social brain. Probabilistic reinforcement learning was studied with either social or non-social rewards for correct choices. Social feedback improved learning in individuals with Williams syndrome but not in typically developing controls or individuals with other intellectual disabilities. Computational modeling indicated that these effects on social feedback were mediated by a shift towards higher weight given to rewards relative to punishments and increased choice consistency. We conclude that reward learning in WS is characterized by high volatility and a tendency to learn how to avoid punishment rather than how to gain rewards. Social feedback can partly normalize this pattern and promote adaptive reward learning.

Computational markers of experience- but not description-based decision-making are associated with future depressive symptoms in young adults

BACKGROUND

Early prediction of high depressive symptoms is crucial for selective intervention and the minimization of functional impairment. Recent cross-sectional studies indicated decision-making deficits in depression, which may be an important contributor to the disorder. Our goal was to test whether description- and experience-based decision making, two major neuroeconomic paradigms of decision-making under uncertainty, predict future depressive symptoms in young adults.

METHODS

One hundred young adults performed two decision-making tasks, one description-based, in which subjects chose between two gambling options given explicitly stated rewards and their probabilities, and the other experience-based, in which subjects were shown rewards but had to learn the probability of those rewards (or cue-outcome contingencies) via trial-and-error experience. We evaluated subjects' depressive symptoms with BDI-II at baseline (T1) and half a year later (T2).

RESULTS

Comparing subjects with low versus high levels of depressive symptoms at T2 showed that the latter performed worse on the experience- but not description-based task at T1. Computational modeling of the decision-making process suggested that subjects with high levels of depressive symptoms had a more concave utility function, indicating enhanced risk aversion. Furthermore, a more concave utility function at T1 increased the odds of high depressive symptoms at T2, even after controlling depressive symptoms at T1, perceived stress at T2, and several covariates (OR = 0.251, 95% CI [0.085, 0.741]).

CONCLUSIONS

This is the first study to demonstrate a prospective link between experience-based decision-making and depressive symptoms. Our results suggest that enhanced risk aversion in experience-based decision-making may be an important contributor to the development of depressive symptoms.

Reinforcement Learning in Patients With Mood and Anxiety Disorders vs Control Individuals: A Systematic Review and Meta-analysis

IMPORTANCE

Computational psychiatry studies have investigated how reinforcement learning may be different in individuals with mood and anxiety disorders compared with control individuals, but results are inconsistent.

OBJECTIVE

To assess whether there are consistent differences in reinforcement-learning parameters between patients with depression or anxiety and control individuals.

DATA SOURCES

Web of Knowledge, PubMed, Embase, and Google Scholar searches were performed between November 15, 2019, and December 6, 2019, and repeated on December 3, 2020, and February 23, 2021, with keywords (reinforcement learning) AND (computational OR model) AND (depression OR anxiety OR mood).

STUDY SELECTION

Studies were included if they fit reinforcement-learning models to human choice data from a cognitive task with rewards or punishments, had a case-control design including participants with mood and/or anxiety disorders and healthy control individuals, and included sufficient information about all parameters in the models.

DATA EXTRACTION AND SYNTHESIS

Articles were assessed for inclusion according to MOOSE guidelines. Participant-level parameters were extracted from included articles, and a conventional meta-analysis was performed using a random-effects model. Subsequently, these parameters were used to simulate choice performance for each participant on benchmarking tasks in a simulation meta-analysis. Models were fitted, parameters were extracted using bayesian model averaging, and differences between patients and control individuals were examined. Overall effect sizes across analytic strategies were inspected.

MAIN OUTCOMES AND MEASURES

The primary outcomes were estimated reinforcement-learning parameters (learning rate, inverse temperature, reward learning rate, and punishment learning rate).

RESULTS

A total of 27 articles were included (3085 participants, 1242 of whom had depression and/or anxiety). In the conventional meta-analysis, patients showed lower inverse temperature than control individuals (standardized mean difference [SMD], -0.215; 95% CI, -0.354 to -0.077), although no parameters were common across all studies, limiting the ability to infer differences. In the simulation meta-analysis, patients showed greater punishment learning rates (SMD, 0.107; 95% CI, 0.107 to 0.108) and slightly lower reward learning rates (SMD, -0.021; 95% CI, -0.022 to -0.020) relative to control individuals. The simulation meta-analysis showed no meaningful difference in inverse temperature between patients and control individuals (SMD, 0.003; 95% CI, 0.002 to 0.004).

CONCLUSIONS AND RELEVANCE

The simulation meta-analytic approach introduced in this article for inferring meta-group differences from heterogeneous computational psychiatry studies indicated elevated punishment learning rates in patients compared with control individuals. This difference may promote and uphold negative affective bias symptoms and hence constitute a potential mechanistic treatment target for mood and anxiety disorders.

Computational Psychiatry Needs Time and Context

Why has computational psychiatry yet to influence routine clinical practice? One reason may be that it has neglected context and temporal dynamics in the models of certain mental health problems. We develop three heuristics for estimating whether time and context are important to a mental health problem: Is it characterized by a core neurobiological mechanism? Does it follow a straightforward natural trajectory? And is intentional mental content peripheral to the problem? For many problems the answers are no, suggesting that modeling time and context is critical. We review computational psychiatry advances toward this end, including modeling state variation, using domain-specific stimuli, and interpreting differences in context. We discuss complementary network and complex systems approaches. Novel methods and unification with adjacent fields may inspire a new generation of computational psychiatry. Expected final online publication date for the Annual Review of Psychology, Volume 73 is January 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

I, robot: depression plays different roles in human-human and human-robot interactions

Socially engaging robots have been increasingly applied to alleviate depressive symptoms and to improve the quality of social life among different populations. Seeing that depression negatively influences social reward processing in everyday interaction, we investigate this influence during simulated interactions with humans or robots. In this study, 35 participants with mild depression and 35 controls (all from nonclinical populations) finished the social incentive delay task with event-related potential recording, in which they received performance feedback from other persons or from a robot. Compared to the controls, the mild depressive symptom (MDS) group represented abnormalities of social reward processing in the human feedback condition: first, the MDS group showed a lower hit rate and a smaller contingent-negative variation (correlated with each other) during reward anticipation; second, depression level modulated both the early phase (indexed by the feedback-related negativity (FRN)) and the late phase (indexed by the P3) of reward consumption. In contrast, the effect of depression was evident only on FRN amplitude in the robot feedback condition. We suggest that compared to human-human interaction, the rewarding properties of human-robot interaction are less likely to be affected by depression. These findings have implications for the utilization of robot-assisted intervention in clinical practice.