Motor decisions in "chicken game" tasks-where aiming closer to a threshold increases reward but overshooting results in zero or even a penalty-consistently exhibit a robust risk-seeking bias relative to the Bayesian optimal solution. This study investigated the sources of this bias by examining the relationship between individuals' subjective risk strategies and their objective risk attitudes assessed through a computational model. Participants performed motor tasks under chicken game conditions with explicit instructions to adopt a risk-seeking, risk-neutral, or risk-averse strategy or under free-choice conditions. In the free-choice condition, participants explicitly engaged in risk-seeking behaviors, confirming an objective risk-seeking bias and suggesting that an explicit preference for risk is a general source of bias. Under the explicit instruction condition, the objective behavior closely matched the instructed strategy, indicating an accurate internal computation of risk attitudes. Critically, individual differences in free-choice bias were explained by posttrial subjective risk reports and the magnitude of deviation from the Bayesian optimal solution under risk-neutral conditions. Together, these findings provide evidence for largely independent contributions of explicit and implicit risk preferences to sensorimotor risk-seeking biases. (PsycInfo Database Record (c) 2026 APA, all rights reserved).
Onagawa et al. (2026) studied this question.