Table 1. Summary of four free model parameters, the variables, the transformations used to map values to unconstrained space and their function in modelling participant behaviour.

Name	Variable	Transform	Function
Precision	β	x1 = ln β	Captures the impact of DEV, derived by forward planning, on action selection
Strategy preference	θ	x2 = θ	Heuristic preference of pursuing a parallel (θ>0) or sequential (θ<0) strategy, independent of the actual DEV
Discount rate	γ	$x_3=\ln \frac{\gamma }{1-\gamma }$	Temporal discounting of DEV by the factor γT−t, where T−t is the number of remaining trials
Reward ratio	κ	$x_4=\ln \frac{\kappa }{2-\kappa }$	Accounts for the possibility that participants may overweight (κ>1) or underweight (κ<1) the actual reward for G2-success relative to G1-success.