Automaticity and word perception: Evidence from Stroop and Stroop dilution effects.

Abstract

The Stroop effect is cut in half by adding a neutral word to the display. D. Kahneman and D. Chajczyk's (1983) "attention capture" account of "Stroop dilution" holds word recognition to be involuntary but strictly serial. The authors compared attention capture to 3 alternatives involving parallel rather than serial processing: In the lexicon, activation is divided among multiple words; postlexically, multiple words race for access to response processes; or prelexically, feature processing is degraded by multiple patterns whether or not they are words. Results support the latter. Multiple patterns are processed in parallel. If any are color words, Stroop effects occur but are reduced because any color word's input to lexical memory is lower in quality than if a single color word were the only pattern. Thus, lexical encoding is involuntary but can operate on several input representations in parallel, with effectiveness determined by input quality.

Dilution of compatibility effects in Simon-type tasks depends on categorical similarity between distractors and diluters

“…The neutral word diluter eliminated the compatibility effect originating in the distractor words LEFT or RIGHT but had no effect on the influence of a left or right distractor arrow [F(1,94) 11.69, MS e 699, p .001]. Previous studies showed that some amount of dilution occurs regardless of the nature of the diluter (Brown et al, 1995;Kahneman & Chajczyk, 1983;Mitterer et al, 2003), but the present results indicate that the success of word diluters in reducing the compatibility effect is strongly dependent on the nature of the distractor that is the source of the compatibility effect. Alternatively, the symbol diluter produced only a small and nonsignificant reduction in the compatibility effect, regardless of whether this effect originated in a word or a symbol distractor was run for the factors of compatibility (compatible vs. incompatible) and diluter condition (no diluter vs. symbol diluter vs. word diluter).…”

“…As in Experiment 1, the symbolic diluter led to a small (5-msec) reduction in the compatibility effect that also did not reach significance [t (48) influence the effect of diluters in the Simon task. First, Brown et al (1995) concluded that the magnitude of the Stroop dilution effect is at least partially determined by the visual complexity of the diluters, as compared with the distractors, with relatively more complex diluters leading to greater visual interference. Applying this logic to the present case, the complexity of the diluters relative to the distractors would have been less in Experiment 1 than in Experiment 2, since the distractors were words versus a single repeating symbol.…”

“…On the basis of this result, Brown et al (1995) suggested a second account, that Stroop dilution is a function of visual complexity and occurs at an early visual processing stage, before lexical encoding. According to Brown et al's early visual interference account, all items in a visual display are processed in parallel.…”

“…When a neutral word (e.g., talk) is presented simultaneously with the target and distractor, the Stroop effect is diminished, and Stroop dilution has occurred. Brown, Roos-Gilbert, and Carr (1995) summarized three accounts of Stroop dilution. First, dilution may occur at the level of lexical processing.…”

Process dissociation, cognitive architecture, and response time: Comments on Lindsay and Jacoby (1994).

“…For that reason, despite a fair amount of research searching for an appropriate baseline, researchers have not settled on one. As a result, many researchers (e.g., Brown, Roos-Gilbert, & Carr, 1995;Kahneman & Chajczyk, 1983;Logan & Zbrodoff, 1979) have compared performance for congruent stimuli with performance for incongruent stimuli rather than comparing performance for both with performance for neutral stimuli, just as Lindsay and Jacoby did.…”