Abstract
A fundamental assumption of models of selective visual attention is that attending to a multidimensional object causes all dimensions of the object to be processed, regardless of whether a given dimension is relevant to current behavioral goals (e.g., Duncan, 1984; Kahneman & Chajczyk, 1983). Evidence for this assumption comes primarily from studies showing that the response compatibility of irrelevant dimensions of attended stimuli influence responses to relevant dimensions (e.g., the Stroop effect). The present studies challenge this assumption. Subjects responded to either the identity or orientation of a red target character among three white distractor characters, one of which (foil) was a possible target. Thus, the foil could be compatible or incompatible on either dimension. Attention was drawn to one of four locations by a red singleton precue. When the foil was cued (i.e., attended) only compatibility on the task-relevant dimension influenced subsequent target responses; response compatibility on the irrelevant dimension of the attended foil had no effect on performance. In contrast, substantial interference was found for incompatible values of the irrelevant dimension of the target. This pattern generalized to stimuli defined by other stimulus dimensions (e.g., color, slant, and motion) as well as to variations in precue validity. Thus, the data suggest that attending to an object does not guarantee all dimensions access to response mechanisms. Instead, the process of selection appears to involve more than the focusing of attention, and task set plays a key role in determining which dimensions of attended stimuli have access to response mechanisms.
DuncanJ.(1984). Selective attention and the organization of visual information. Journal of Experimental Psychology: General, 113, 501–517.
KahnemanD.ChajczykD.(1983). Tests of the automaticity of reading: Dilution of Stroop effects by color-irrelevant stimuli. Journal of Experimental Psychology: Human Perception and Performance, 9, 497–509.
Supported in part by NSF BCS-9817673 to C.F.