Fig 1. Hiragana stimuli used in the probabilistic learning task in experiments 1a and 1b and experiment 3.

Each pair of stimuli was randomly presented in separate trials. During each trial participants chose one of the stimuli of the pair. Feedback following participant’s choice was determined probabilistically. Reward probability (indicated below each stimulus) differed between characters. Task version 1 and 2 differed with respect to the characters associated with more probable positive and more probable negative feedback, respectively. Specifically, in task version 2, Hiragana stimuli were switched within the AB, CD and EF pair. Half of the participants were subjected to version 1 and the other half to task version 2. In experiment 3, a task version was used for which only the Hiragana stimuli within the AB pair were switched.