Risk attitude and choice consistency identify robust behavioral profiles in decision-making under risk.

Individual differences in risky decision-making may reflect both risk attitude and the consistency with which preferences are expressed. We examined whether complementary behavioral descriptors could identify reproducible participant profiles in 67 healthy adults performing binary choices between safe options and gain-domain risky alternatives with equiprobable outcomes, varying in expected value and outcome variance. After robust trial-level response-time (RT) outlier screening, psychometric choice functions yielded two risk-attitude descriptors-overall risk attitude and context-dependent variability-and two choice-consistency descriptors-psychometric slope consistency and lapse-related stochasticity. RT outliers represented 8.7% of analyzed trials and decreased across blocks. Overall risk attitude and contextual variability were weakly associated, whereas greater contextual variability was consistently related to lower choice consistency. Controlled model-based clustering analyses supported a reproducible two-profile representation. Profile 1 captured participants with greater risk aversion, greater context-dependent variability, lower choice consistency, and higher lapse rates. Profile 2 captured participants with near-neutral risk attitude, low contextual variability, high choice consistency, and low lapse rates. These findings reveal a reproducible organization of individual behavior in a multidimensional risk-consistency space and provide a foundation for subsequent systems-level analyses of profile-dependent neural dynamics.