Relative, not absolute, stimulus size produces a correspondence effect between stimulus size and response position.

Recent studies demonstrated a novel compatibility (or correspondence) effect between physical stimulus size and horizontally aligned responses: Left-hand responses are shorter and more accurate to a small stimulus, compared to a large stimulus, whereas the opposite is true for right-hand responses. The present study investigated whether relative or absolute size is responsible for the effect. If relative size was important, a particular stimulus would elicit faster left-hand responses if the other stimuli in the set were larger, but the same stimulus would elicit a faster right-hand response if the other stimuli in the set were smaller. In terms of two-visual-systems theory, our study explores whether ‘vision for perception’ (i.e., the ventral system) or ‘vision for action’ (i.e., the dorsal system) dominates the processing of stimulus size in our task. In two experiments, participants performed a discrimination task in which they responded to stimulus color (Experiment 1) or to stimulus shape (Experiment 2) with their left/right hand. Stimulus size varied as an irrelevant stimulus feature, thus leading to corresponding (small-left; large-right) and non-corresponding (small-right; large-left) conditions. Moreover, a set of smaller stimuli and a set of larger stimuli, with both sets sharing an intermediately-sized stimulus, were used in different conditions. The consistently significant two-way interaction between stimulus size and response location demonstrated the presence of the correspondence effect. The three-way interaction between stimulus size, response location and stimulus set, however, was never significant. The results suggest that participants are inadvertently classifying stimuli according to relative size in a context-specific manner.