Estimating the proportions and latencies of reaction time outliers: A pooling method and case study of lexical decision tasks.

Following the exclusion of the unrealistically fast RTs, a preliminary ANOVA on individual-participant mean correct RTs showed highly significant effects of word versus nonword trial, response repetition versus response alternation (i.e., whether the preceding trial’s response had been the same or different), and the interaction of these two factors ( p < 0 . 001 and η p 2 > 0 . 34).

also in 132,142 other papers