To reduce the influence of a few outlying data points (<1% of trials) and to eliminate fast “guessing” responses, analysis of reaction times (RTs) was limited to correct responses made between 100 ms and 1200 ms after target onset. Correct responses made after the 500-ms response deadline (but before 1200 ms) were retained for analyses of RTs, and all responses made between 100 and 1200 ms were coded for categorization accuracy and included in analyses of accuracy rates. Preliminary analyses indicated that participants’ sex was unrelated to any of the effects of interest and therefore was not included in the main analyses. ERP analyses were limited to a core set of 15 electrodes representing five midline scalp locations from frontal to parietal areas (Fz, FCz, Cz, CPz, and Pz), five lateral locations left of midline (F3, FC3, C3, CP3, and P3), and their right-hemisphere homologues. Greenhouse-Geisser-adjusted p values are given for all effects involving repeated measures with more than two levels.
