Increased ventromedial prefrontal cortex activity in adolescence benefits prosocial reinforcement learning

Abstract

Learning which of our behaviors benefit others contributes to social bonding and being liked by others. An important period for the development of (pro)social behavior is adolescence, in which peers become more salient and relationships intensify. It is, however, unknown how learning to benefit others develops across adolescence and what the underlying cognitive and neural mechanisms are. In this functional neuroimaging study, we assessed learning for self and others (i.e., prosocial learning) and the concurring neural tracking of prediction errors across adolescence (ages 9-21, N=74). Participants performed a two-choice probabilistic reinforcement learning task in which outcomes resulted in monetary consequences for themselves, an unknown other, or no one. Participants from all ages were able to learn for themselves and others, but learning for others showed a more protracted developmental trajectory. Prediction errors for self were observed in the ventral striatum and showed no age-related differences. However, prediction error coding for others was specifically observed in the ventromedial prefrontal cortex and showed age-related increases. These results reveal insights into the computational mechanisms of learning for others across adolescence, and highlight that learning for self and others show different age-related patterns.

A Domain-General Developmental “Do-GooD” Network Model of Prosocial Cognition in Adolescence: A Systematic Review

“…TPJ activation accounted for computations regarding the value of giving to others whereas the VS had greater activation during selfish decisions, and the VMPFC responded to both selfish and prosocial decisions ( Hutcherson et al, 2015 ). Other neurocomputational work on prosocial learning suggests that the VS and VMPFC activations were also related to prediction error (PE) signals in adolescents, where the VS represents PE while learning for oneself while the VMPFC represents PE while learning for others ( Westhoff et al, 2021 ). However, in a previous adult study using the same experiment, PE in the VS was impartial to self-versus other, and the subgenual ACC contributed to prosocial PE ( Lockwood et al, 2016 ).…”

Social learning across adolescence: A Bayesian neurocognitive perspective

“…Thus, normative influence is particularly pertinent in situations where behavior is observed by peers ( Deutsch and Gerard, 1955 ) and social reward or punishment are more imminent ( Liu et al, 2017 ). We note that in this review, we focus on how adhering to group norms or pleasing others is a means to maximize subsequent social reward for the self, rather than on how maximizing benefits for others such as in prosocial learning ( Lockwood et al, 2016 , Westhoff et al, 2021 ) can be a goal in and of itself, which is outside the scope of the current review (but see Theriault et al, 2021 ).…”

l-DOPA and oxytocin influence the neurocomputational mechanisms of self-benefitting and prosocial reinforcement learning

“…Importantly, to behave in a socially adaptive manner we do not only need to learn how to obtain rewards for ourselves but also how to benefit others, since actions that intend to benefit others, also referred to as prosocial behavior, are crucial for social functioning and the formation and maintenance of reciprocal social relationships (Carlo, 2013). Prior research suggests that this so-called "prosocial learning" relies on the same reinforcement algorithms as individual or self-benefitting learning (Cutler et al, 2021b;Lockwood, Apps, Valton, Viding, & Roiser, 2016;Martins, Lockwood, Cutler, Moran, & Paloyelis, 2022;Westhoff, Blankenstein, Schreuders, Crone, & van Duijvenvoorde, 2021). However, whereas the encoding of PEs during both types of learning involved the VS, only prosocial PEs were found to be encoded in the subgenual anterior cingulate cortex (sgACC), with stronger responses in this latter brain region for those with higher levels of empathy, suggesting social specificity on the implementation level (Lockwood et al, 2016;Martins, Lockwood, et al, 2022 but see; Westhoff et al, 2021).…”