Stroop Interference in a Delayed Match-to-Sample Task: Evidence for Semantic Competition

Discussions of the source of Stroop interference continue to pervade the literature. Semantic competition posits that interference results from competing semantic activation of word and color dimensions of the stimulus prior to response selection. Response competition posits that interference results from competing responses for articulating the word dimension versus the color dimension at the time of response selection. To induce semantic competition in memory while eliminating response competition, we embedded a Stroop-like task into a delayed match-to-sample task. We probed each dimension independently by presenting either color targets (i.e., two colors) or word targets (i.e., two words). Participants viewed a real color word or a nonsense control word in black or colored fonts congruent or incongruent with the real color word. After a 5s delay, participants viewed color targets or word targets (i.e., a match and a foil), and were instructed to select the correct match (sample color if color targets; sample word if word targets). We manipulated whether the foil conflicted with the sample along its other dimension (e.g., word sample “red” in blue font with the word “red” as the match and the word “blue” as the foil) and show Stroop-like interference such that performance deteriorated for real but not nonsense words during the critical conflict trial. A follow-up experiment with 10 and 25s delays between sample and targets replicated the results while showing a decrement in performance at 25s. Results provide evidence for semantic competition in memory while eliminating response competition