Fig. 4: Experiment 2 task design.
From: Reinforcement learning increasingly relates to memory specificity from childhood to adulthood
A Each block of the reinforcement-learning task included nine unique stimuli, which comprised three exemplars each drawn from three broader categories. Each stimulus set also included an additional stimulus category with five novel stimuli, as well as two additional novel exemplars per sampled category. B In the category-predictive condition, rewards on every trial were sampled from Bernoulli distributions with win probabilities determined by the stimulus categories. In the exemplar-predictive condition, rewards on every trial were sampled from Bernoulli distributions with win probabilities determined by the individual exemplars. The optimal action (depicted by the shaded color) resulted in wins on 90% of trials and losses on 10% of trials. The two other actions resulted in wins on 10% of trials and losses on 90% of trials. C On every trial of the reinforcement-learning task, participants saw a stimulus and three choice options. After selecting an option, they viewed the outcome of their choice: either a win (+1 point) or a loss (ā1 point). The images in the figure are illustrative; actual task stimuli differed slightly. Image credit: iStock/GlobalP (https://www.istockphoto.com/portfolio/GlobalP), Life on White (https://www.lifeonwhite.com/).
