Reversing priming while maintaining interference

Abstract

The amount of interference and priming is examined in a Stroop color-naming task, in which distractors were either color words (conventional Stroop) or nonwords comprising the first two letters of the color words (color pseudowords). It was found that both color words and color pseudowords produced an interference effect, and that the amount of interference by these two types of stimuli was highly correlated across subjects. It was also found that color words led to a reaction time cost when distractor and subsequent target were the same, compared to trials on which no such relation existed. This was an instance of negative priming (Tipper, 1985). Conversely, color pseudowords led to positive priming and reaction time saving, under the same conditions. These data suggest that interference is related to the initial structural elements of the words, and that it may occur regardless of whether or not words can be fully analyzed, whereas priming effects may depend on whether it is possible to analyze the stimuli on a semantic level.