Academic journal article Memory & Cognition

Depth Cues Do Not Underlie Attentional Modulations of the Stroop Effect

Academic journal article Memory & Cognition

Depth Cues Do Not Underlie Attentional Modulations of the Stroop Effect

Article excerpt

The well-known Stroop effect is usually attributed to the automaticity of word reading. Recently, Wühr and Waszak (2003) had participants name the color of one of two rectangles and found that words in the relevant object produced larger Stroop effects than did words in the irrelevant object or in the background. They attributed this difference to an object-based mechanism of attentional selection that amplifies processing of all the features of an attended object. However, in the displays used by Wühr and Waszak, occlusion suggested the presence of different depth planes. Hence, the increased Stroop effect could have resulted from perceiving the words to be in the same depth plane as the relevant object and not from perceiving the words to be parts of the relevant object. Two experiments tested between these accounts by using displays without monocular depth cues. The results of both experiments replicate those of Wühr and Waszak, supporting their object-based account.

The Stroop effect (Stroop, 1935) is one of the bestknown phenomena in experimental psychology. In the usual Stroop task, participants are presented with color words in different colors (e.g., the word RED in green ink). When participants have to report the ink color, irrelevant color words produce interference in comparison with a neutral condition (e.g., the letter string XXX in green ink); congruent color words improve performance (see MacLeod, 1991, for a review). The most popular explanation for the Stroop effect holds that word reading is automatic and therefore hard to avoid (see Brown, Gore, & Carr, 2002; MacLeod & MacDonald, 2000; Posner & Snyder, 1975; but see Melara & Algom, 2003, for a different account).

However, the results of a recent study by Wühr and Waszak (2003) suggest that the automaticity of word reading is not the only reason for the robustness of the Stroop effect, at least in integrated Stroop tasks. Rather, it appears that word shape is not only processed because word reading per se is automatic, but also because the shape belongs to the attended perceptual object (see Kahneman & Henik, 1981). Participants in Wiihr and Waszak's study had to report the color of a relevant object that partially occluded an irrelevant object (see Figure 1). To provoke Stroop effects, displays contained congruent, neutral, or incongruent color words. Most importantly, these color words were presented as part of the relevant object (Figure 1A), as part of the irrelevant object (Figure 1B), or in the background (Figure 1C). Because the spatial distance between the fixation point and the color words was identical in each condition, a spotlight-like mechanism of selective attention (see Posner, 1980) would not predict any differences between these conditions. Yet Wühr and Waszak observed greater facilitation from congruent words and greater interference from incongruent words when the words were part of the relevant object than when they were part of the irrelevant object or appeared in the background. The latter two conditions did not differ. The same result was obtained when participants had to report the color of the occluded rectangle.

Wühr and Waszak (2003) explained these results in terms of an object-based mechanism of selective attention, as suggested by Duncan (1984) and Kahneman (e.g., Kahnetnan & Henik, 1981 ; Kahneman & Treisman, 1984). According to this account, preattentive processes segment the visual field into candidate objects (figures) and background, and attention selects among the candidate objects. Importantly, selection of a particular object amplifies processing of all of its features, regardless of their task relevance. The results of the Wiihr and Waszak study imply that there are not only quantitative but also qualitative differences between processing in separated and processing in integrated Stroop tasks.

Yet, there is an alternative explanation for the results of Wühr and Waszak (2003). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.