Fig. 1.

(a) Deterministic transition matrix presented to participants during the training phase to aid learning. (b) Deterministic reward matrix. Note that this was never presented to participants; they must instead learn the reward structure through trial and error. (c) Task as presented to participants. The white box denotes the state that the participant is currently in. Symbols below each state denote the deterministic reward achieved by transitioning away from that state; ‘++’ = +140 points; ‘+’ = +20 points; ‘-’ = −20 points; ‘--’ = −140 points.