Toward a translational model of Stroop interference

Abstract

Models ofthe Stroop effect that postulate single, centralized processing stations are fundamentally incorrect. A translational model is proposed in which parallel processing systems are hypothesized that are linked by a translation mechanism. Each system processes information using its own internal code specifically linked to the type of information the system processes. Information is passed between these systems via the translation mechanism. Evidence supporting this model comes from both the literature and the experiments reported in this article. Of primary importance is the demonstration that the pattern of interference that emerges is dependent upon the cognitive system used to process the information.The Stroop paradigm was employed in the four experiments reported. In this paradigm, first reported by Stroop in 1935,1 bidimensional stimuli are created such that either dimension of a stimulus may be responded to with the same overt response. Reaction time is measured of the response to one of the stimulus dimensions, while the subject attempts to ignore the other dimension, which will conflict on some trials with the correct response. For example, a subject who is shown the word RED printed in blue ink could respond either to word meaning by saying "red" or to ink color by saying "blue."?The interesting finding is that, when subjects must read the word out loud, there is no effect of a conflicting ink color. Reaction time in the conflicting condition is as fast as in the condition in which the color word names the ink color for the trial. However, when the subject must name the ink color, reaction time in the conflicting condition is greatly increased, compared to the time required for the condition in which the color word and the ink color have the same value (e.g., RED in red).Similar findings have been reported using different stimulus sets. For example, when word meaning and spatial position are combined (e.g., the words RIGHT or LEFT appear to the right or left of fixation), reading the word out loud is not affected by a conflicting spatial position, but naming the position the word occupies is slowed if a conflicting word appears in that position (Palef & Olson, 1975;Seymour, 1974).The models proposed to account for the Stroop effect have differed in detail but have two underlying assumptions in common. The first is that, at some point in the system, all the information passes through a single, centralized decision and/or response stage. This is true regardless of the particular task employed. To account for the asymmetry in the interference effects, the second assumption is made: The two dimensions are hypothesized to arrive at the decision stage at different times.In Figure 1, a schematic of this type of model is shown for the ink color/color word form of the Stroop task. Two stimulus-analyzing stages are hypothesized: one to abstract word meaning and another to abstract ink color information (see Treisman, 1969). These stimulus analyzers work in parallel, and output from them races toward a centra...

A reverse Stroop effect without translation or reading difficulty

“…We have confirmed the Virzi and Egeth (1985) result using the present paradigm. Specifically, a version of our experiment was conducted in which the response labels were color patches rather than words.…”

“…As Coltheart noted in his review of a previous version of this article, a translation between orthography and phonology occurs in reading aloud. Thus, the assertion by Virzi and Egeth (1985) and Sugg and McDonald (1994) that reading aloud does not require a translation makes their accounts difficult to understand.…”

“…These accounts assert that the irrelevant dimension affects performance when a correct response requires a translation between memory codes (see, e.g., Glaser & Glaser, 1989;Sugg & McDonald, 1994;Virzi & Egeth, 1985). In these accounts, words and colors are processed in their own subsystems.…”

Influences of different combinations of conceptual, perceptual, and structural similarity on stimulus-response compatibility

“…Thus, perceptual similarity between the stimulus and response modes is primarily a factor when conceptual similarity is present (see also Lu & Proctor, 2001). The absence of the indicator of set-level differences due to perceptual similarity (i.e., the code-modality interaction for the congruent mapping) when S-R orientations are orthogonal and the S-R sets have structural but not conceptual similarity implies that there is not a fundamental distinction between spatial and linguistic systems, as is often hypothesized (e.g., Virzi & Egeth, 1985). Rather, the differences that are evident when the orientations are parallel and the S-R sets have conceptual similarity seem to be a function of subjects having more pre-experimental experiences with the same mode pairings of stimuli and their associated responses than with the different mode pairings (e.g., Alluisi et al, 1957).…”

Stimulus-response compatibility as a determinant of interference in a Stroop-like task

“…A number of studies have shown that Stroop interference can be significantly reduced by using non-naming responses such as buttonpressing (Flowers & Dutch, 1976;Keele, 1972;Virzi & Egeth, 1985). A possible explanation is that a printed word is "a program for a reader telling him how to pronounce it, but an object is not related to its name except arbitrarily" (Beller, 1975, p. 155).…”