Fig. 1.

Sequential learning task, risk-probability formulation and hierarchical drift–diffusion model(HDDM) (a) 2-step sequential learning task (b) Evolution of all stage 2 choice reward probabilities across trials (c) risk/uncertainty as a function of second stage reward probability (Eq. 1) which follows an inverted U-curve (d) An example plot with reward probabilities of two stage2 choices, their corresponding risk(Eq. 1) and trial-wise risky choice among the two choices (e) pictorial representation of the drift–diffusion model with its parameters (threshold—a, drift rate-v and response bias-z) and (f) pictorial representation of the HDDM structure with input variables and estimates.