Material-specific capacity limitations in visual word recognition: Evidence from the Stroop paradigm

The first set of experiments explores a variant of Stroop's paradigm in which a string of characters is coloured and appears at fixation, and a colour word distractor, in white, flanks the target string. Colour word distractors produce a Stroop effect when the elements at fixation are nonlinguistic (Experiments 5–8), regardless of whether one or all of the display elements at fixation are coloured. In contrast, colour word distractors fail to produce a Stroop effect when the material at fixation is linguistic (Experiments 1–4), regardless of whether one or all of the display elements are coloured. These experiments replicate and extend some (but not all) of the effects reported by Besner and Stolz (1999a,b). The second set of experiments explores another variant of the Stroop paradigm in which a colour patch target appears at fixation and a colour word distractor appears as a flanker. The Stroop dilution effect refers to the reduction in the size of the Stroop effect when a neutral word distractor is added to the display as a second flanker. Experiment 9 replicates the standard Stroop dilution effect. Experiments 10, 11, and 12 show that the Stroop effect is not observed if a word or a nonword is coloured and appears at fixation instead of the flanker position. Experiment 13 shows a full-blown Stroop effect if the coloured string at fixation is a colour word. Experiments 14, 15, and 16 show that a full-blown Stroop effect is observed if the coloured string at fixation is “non-linguistic.” Taken together, these results implicate a role for domain specific limited capacity processing in the context of the Stroop paradigm.