Fig. 1

Trial structure for the (A) single-cue probabilistic learning task; (B) multi-cue probabilistic learning task; and (C) reinforcement learning task. In the single-cue probabilistic learning task (A), participants are first shown a stimulus for 1.5s. Then a response screen depicting two art periods (e.g., “Londs” and “Grakes”) is shown, during which participants have to respond within 2.5s. Feedback is then provided for 1s, followed by a fixation cross to signal the next trial. In the multi-cue probabilistic learning task (B), participants are first shown a stimulus, and they are asked to predict if the stimulus will receive a high or low rating within 5s. Then a red fixation cross to signal incoming feedback was presented for 1s, and the feedback was presented for 1.5s. The trial ends with a black fixation cross for a randomly jittered inter-trial interval (ITI) between 1.5s and 4.5s. In the reinforcement learning task (C), two stimuli are presented, and participants have to decide which has a higher value for a given year within 2.5s. This is followed by feedback for 1s, and then a fixation cross to signal the next trial