On ignore trials, the to-be-ignored color was always the majority color. Additionally, the target was never in the to-be-ignored color; rather, it was always one of the heterogeneously colored letters (see Fig. 4). Therefore,participants could benefit from the to-be-ignored information by knowing that the target was not the to-beignored color, thus they could attempt to ignore 6 out of 12 items. Critically, in Experiment 2a, the to-be-ignored color varied from trial to trial; thus, the participant could not learn anything consistent about the to-be-ignored information. However, in Experiment 2b, the to-beignored information was held constant for the entire experiment (e.g., a participant could ignore red on allignore trials). The specific to-be-ignored color was randomly assigned for each participant. As on neutral trials, the spatial location of all colored letters was randomly selected so that stimuli would not be grouped by color.