Some considerations of two alleged kinds of selective attention.

Abstract

The present article deals with selective attention phenomena and elaborates on a stimulus material classification, "stimulus set" versus "response set," proposed by Broadbent (1970, 1971)9 Stimulus set is defined by some distinct and conspicuous physical properties that are inherent in the stimulus. Response set is characterized by the meaning it conveys, and thus its properties are determined by cognitive processing on the part of the organism. Broadbent's framework is related to Neisser's (1967) distinction between two perceptual-cognitive processes, namely, preattentive control and focal attention. Three experiments are reported. A before-after paradigm was employed in Experiment 1, together with a sptial arrangement manipulation of relevant versus irrelevant stimuli (being grouped or mixed). The results indicated that before-after instruction had a stronger effect under stimulus set than under response set conditions. Spatial arrangement, on the other hand, affected performances under response set but not under stimulus set conditions. These results were interpreted as supporting the idea that stimulus set material, which is handled by preattentive mechanisms, may be processed in parallel, while response set material requires focal attention that is probably serial in nature. Experiment 2 used a search task with different levels of noise elements. Although subjects were not able to avoid completely the processing of noise elements, they had much more control under stimulus set than under response set conditions. Experiment 3 dealt with memory functions and suggests differential levels of perceptual processing depending on the nature of the stimulus material. This extends the memory framework suggested by Craik and Lockhart (1972). The results of these experiments, together with evidence from other behavioral and physiological studies, lend strong support to the proposed theory. At the theoretical level, it is suggested that the distinction between stimulus and response set, and the corresponding one between preattentive mechanisms and focal attention, are on a continuum rather than being an all-or-none classification. Thus, it permits greater congnitive flexibility on the part of the organism, which is reflected through the assumption that both preattentive mechanisms and focal attention may operate simultaneously and differ only in the salience of their functioning. From a methodological point of view, the distinction between stimulus material and organismic processes is emphasized. It is argued that researchers have not given sufficient attention to the properties of the stimulus materials that they have used, and as a consequence have reached unwarranted conclusions, as exemplified by a few studies that are briefly discussed.