Location and color as cuing dimensions in contingent classification

Abstract

The contingent classification task is one in which stimuli in a classification task are simultaneously cued by the addition of another dimension so that the classification is reduced to a discrimination of only two stimuli. The contingencies are arranged so that the cuing dimension itself provides no response information. Four experiments were carried out in which location and color were the cuing dimensions for contingent stimuli that were either letters or lines. These stimuli were considerably more difficult to perceive than those used in earlier experiments. L0-cation provided effective contingent classification for both letters and lines, but color was ineffective with both types of stimuli. It is suggested that in order for a cuing dimension to provide effective contingent classification, its levels must in effect provide information channels, much as modalities do. Location seems to have a special status in providing such visual channels, whereas color does not. Discriminability of the levels on the cuing dimension cannot be dismissed as a factor in the different effectiveness, since the two locations were more discriminable than were the two colors. Since color was maximally discriminable, however, location still has a special status in that it can provide greater discriminability, even possibly through its ability to act as a channel.Contingent infonnation processing occurs whenever the attributes, features, or dimensions to be processed are determined by (i.e., contingent on) the level of some other stimulus property. Any such task includes both the cuing dimension, which specifies which properties are to be processed, and the contingent dimension, whose properties are so specified. The cuing dimension may be simultaneous with the contingent dimension, as in the present experiments, or it may precede (or precue) the contingent dimension. Garner ( , 1985 has reviewed the rather limited literature on this topic, including contingent stimulus-response relations (e.g., Shaffer, 1965) and contingent identification (e.g., Biederman, 1972). In both of these contingent tasks, the cuing dimension not only specifies something about the perceptual properties to be processed, but also gives direct response information.In most such experiments, when the cuing dimension precedes (i.e., precues) the occurrence of the contingent dimension, the primary purpose is in fact to investigate partial response preparation. This has been the concern of Miller (1982Miller ( , 1983Miller ( , 1985 and Proctor (1984, 1985). The precuing technique has been used for a variety of other purposes as well (see LaBerge, Van Gelder, & Yellott, 1970;Posner, 1978). However, the intentional or coincidental direct cuing of the response in these experiments limits their ability to investigate purely perceptual relations between the cuing and the contingent stimulus dimensions. Contingent ClassificationThe contingent classification task is a form of contingent information processing in which the cuing dimension provides no direct response in...

Temporally selective attention modulates early perceptual processing: Event-related potential evidence

“…The different latencies of the earliest attention effects across studies suggest that spatial selection may affect earlier visual processing than temporal selection does. This hypothesis is consistent with the argument that spatial information has special status in the visual system and temporal information in the auditory system (Bregman, 2002;Garner, 1987;Kubovy, 1981;Sanders & Poeppel, 2007).…”

Chronometric evidence for entrained attention

“…Chronometric measures of performance for targets well above threshold provide another approach to studying attention and minimize the contribution of response biases. Although some have proposed that any information about a target acts to facilitate reaction time (RT) by lowering a response criterion (Duncan, 1980(Duncan, , 1984, this hypothesis has not received wide support (Garner, 1985;Garner, 1987;Tsal, 1983). Some types of information, such as spatial cues about likely location, improve performance speed without appreciably degrading accuracy.…”

Time course of object identification: Evidence for a global-to-local contingency.

“…Subjects were able to use such contingencies to decrease reaction time. Garner (1985Garner ( ,1987 used a generally similar paradigm, although the cuing dimensions in Garner's tasks did not eliminate possible responses. Garner found that when the relevant discrimination was difficult, processing was facilitated when location was the cuing dimension but not when color was the cuing dimension.…”