Opposite effects of capacity load and resolution load on distractor processing

Abstract

Previous studies suggest that an increased perceptual load decreases distractor processing whereas an increased working memory load increases distractor processing. According to the load theory of attention, an increased perceptual load allows the target to out-compete the distractors at a relatively early stage of processing. In contrast, an increased working memory load interferes with cognitive control, increasing interference at later stages. Here, we suggest that the critical distinction is not between perceptual load and working memory load, but rather between an emphasis on resolution and an emphasis on capacity. That is, perceptual load manipulations typically emphasize resolution (fine-grained discriminations), whereas working memory load manipulations typically emphasize capacity (simultaneous processing of multiple relevant stimuli). To test this hypothesis, we examined the effects of a working memory load that emphasized either the number of items to be stored (capacity load) or the precision of the representations (resolution load). To compare high and low capacity loads, subjects were required to store either 4 colors or 2 colors in working memory to perform a change detection task; the resolution load was minimized by the use of large change magnitudes on change trials. To compare high and low resolution loads, set size was held constant at 2, and subjects were required to detect either small or large color changes. An Eriksen flanker task was presented during the retention interval of the change detection task to assess distractor processing. We found that an increased capacity load led to increased flanker interference (a measure of distractor processing), whereas an increased resolution load led to reduced flanker interference. The latter finding was further replicated with a different manipulation of resolution load. The opposite effects of capacity load and resolution load on distractor processing suggest that working memory load can affect both early and late stages of processing. Meeting abstract presented at VSS 2012