I use the news stylesheet because color-naming times are longer for lists that contain conflict items such as the word green printed in red than they are for lists that consist of control items such as red squares or red swastikas. Stroop attributed color-word interference to the difference in the strength of association between stimuli and responses: The association between a word stimulus and a reading response is stronger than the association between a color stimulus and a naming response. Stroop tested the strength of association interpretation by training the color-naming response to conflict stimuli over 8 days. There was a 34% gain (from 49.6 s to 32.8 s) in color-naming speed over 8 days, although the time the participants took to name the colors of conflict stimuli (24.7 s) was still not as fast as the time they took to name colors of control items (i.e., swastiskas). After the participants had been trained in color naming, Stroop switched the task to word reading. Reading time for conflict stimuli increased by 79% relative to the time taken before training on color naming (i.e., from 19.4 s to 34.8 s).
MacLeod (1998) also reported that color-word interference might be reduced but not eliminated by training on color naming. One should note, however, that MacLeod included congruent displays (e.g., the word red printed in red), which might have altered the way in which participants prepared for incongruent displays (Monahan, 2001). The color-word interference reflects the cost of resolving the conflict that arises from the color-name responses that are activated by each dimension of a conflict display. Although the color-naming task demands a response on the basis of the color of print dimension, a response on the basis of the name of the word is activated unintentionally. (The degree of activation is affected by an attentional--allocation strategy; see MacLeod, 1991, for a review of data and theory on color-word interference.)