Stimuli in the Medicine Task used in Experiment 1 and 2. Participants were repeatedly presented with four different types of medicines, which could be followed by two different outcomes (i.e., the patient was healed or remained ill). The repeated presentation of different medicine-outcome contingencies allowed participants to infer the actual contingency for each medicine, which could in turn be affected by the participant’s feeling of control. The contingency table represents the percentage of trials in which a particular medicine was followed by a particular outcome.