Testosterone eliminates strategic prosocial behavior through impacting choice consistency in healthy males

Abstract

Humans are strategically more prosocial when their actions are being watched than when they act alone. Using a psychopharmacogenetic approach, we investigated the computational and endocrinological mechanisms of such audience-driven prosociality. 187 participants received either a single dose of testosterone or a placebo and performed a prosocial and self-oriented reinforcement learning task. Crucially, the task was performed either in private or when being watched. Rival theories of testosterone’s role in status-seeking suggest that the hormone might either strengthen or diminish audience-depended generosity. We show that exogenous testosterone strongly decreases submission to audience expectations, full eliminating strategic i.e., feigned generosity. We next performed reinforcement-learning drift-diffusion modeling to elucidate which latent aspects of decision-making testosterone acted on. Computational modeling revealed that testosterone compared to placebo did not deteriorate reinforcement learning per se, rather, in presence of the audience, the hormone impacted the expression of the learned information into behavioral choice. These results indicate that instead of deceptively increasing socially desirable behavior, testosterone boosts honest forms of status-seeking, arguably by impacting the motivational link between learned values and behavior.

A Hitchhiker’s Guide to Bayesian Hierarchical Drift-Diffusion Modeling with dockerHDDM

“…We recommend the following resources for quickly catching up and avoid misuse or abuse of HDDM. Table S1 The drift-diffusion model (DDM) is one of the most widely used computational models (Ratcliff et al, 2016) to quantify processes of decision-making in neuroscience (Cavanagh et al, 2011;Herz et al, 2017;Herz, Zavala, Bogacz, & Brown, 2016;Shadlen & Shohamy, 2016), psychology (Hu, Lan, Macrae, & Sui, 2020;Johnson, Hopwood, Cesario, & Pleskac, 2017;Kutlikova, Zhang, Eisenegger, Honk, & Lamm, 2022), behavioral economics (Desai & Krajbich, 2022;Sheng et al, 2020), and psychiatry (Ging-Jehli, Ratcliff, & Arnold, 2021;Pedersen et al, 2021). DDM describes how a process of evidence accumulation to a decision boundary gives rise to experimentally observed reaction time and behavioral choices (e.g., Voss et al, 2013; Figure 1).…”