Testosterone Administration Increases the Computational Impact of Social Evaluation on the Updating of State Self-Esteem

Background: High self-esteem promotes well-being and buffers against anxiety. However, state self-esteem (SSE) is not stable but rather is dynamically updated based on evaluations received from others. Particularly in men, decreased SSE is related to aberrant behaviors and clinical symptoms. A critical physiological mechanism that underlies these associations may involve a sex hormone, testosterone. However, the causal relationship between testosterone and the process of updating SSE in men remains unknown. Methods: The study had a double-blind, placebo-controlled, between-participants design. First, we administered a single dose (150 mg) of testosterone or placebo gel to healthy young men (N = 120). Subsequently, the participants completed a social evaluation task in which they adjusted their prediction of potential evaluation by others and dynamically reported their SSE based on the social feedback they received. Meanwhile, we applied a computational modeling approach to investigate the dynamic changes in their SSE. Results: Exogenous testosterone significantly influenced the participants’ expectation of receiving positive social feedback from raters with different approval rates and separately amplified the changes in average SSE when the participants received positive or negative feedback from the raters. Even more importantly, computational modeling showed that the participants who received testosterone (vs. the placebo) assigned a higher weight to expected social feedback and social prediction errors when updating their SSE. Conclusions: The findings provide potential clinical implications for combining exogenous testosterone with interventions aimed at enhancing SSE through positive social feedback as a preclinical treatment for aberrant behaviors and clinical symptoms.