Social Reinforcement Learning Parameters Change Following Web-based Interpretation Bias Modification: Secondary Analysis of a Digital Mental Health Intervention (Preprint)

Abstract

Biases in social reinforcement learning, or the process of learning to predict and optimize behavior based on rewards and punishments in the social environment, may underlie and maintain some of the negative cognitive biases that are characteristic of social anxiety. However, little is known about how cognitive and behavioral interventions may change social reinforcement learning in anxious individuals.

This study assessed whether a scalable, web-based cognitive bias modification for interpretations (CBM-I) intervention changed social reinforcement learning biases in participants high in social anxiety symptoms. This study focused on two types of social reinforcement learning relevant to social anxiety: learning about other people and learning about one’s own social performance.

Participants (N=106) completed 2 laboratory sessions, separated by 5 weeks of ecological momentary assessment tracking emotion regulation strategy use and affect. Approximately half (n=51) also completed up to 6 brief daily sessions of CBM-I in week 3. Participants completed a task that assessed social reinforcement learning about other people at both laboratory sessions, and a task that assessed social reinforcement learning about one’s own social performance at the second session. Behavioral data from these tasks were computationally modeled with Q-learning and analyzed with mixed effects models.

After CBM-I, participants updated their beliefs about others more slowly (p=.04, d=-0.29) but used what they learned to make more accurate decisions (p=.005, d=0.20), choosing rewarding faces more frequently, effects not observed among participants who did not complete CBM-I. They also showed less biased updating about their social performance compared to participants who did not complete CBM-I, learning similarly from positive and negative feedback, and from feedback on items related to poor (vs. good) social performance. Regardless of intervention condition, participants at session 2 (vs. 1) updated their expectancies about others more from rewarding (p=.003, d=0.43) and less from punishing outcomes (p=.001, d=-0.47), and they became more accurate at learning to avoid punishing faces (p=.001, d=0.20).

Both CBM-I and tracking emotion regulation strategy use and affect may have beneficial effects on social reinforcement learning for socially anxious individuals.