Inhibition of return to color: A replication and nonextension of Law, Pratt, and Abrams (1995)

We presented subjects with an unpredictive cue that was followed after a 15 0- to 900-msec interstimulus interval (ISI) by a detection target Cue and t arget were blue or red squares that appeared at fixation and in an otherwis e uniform black field. In a filler condition, a task-irrelevant filler stim ulus (magenta square) was presented during the ISI; in a no-filler conditio n, no stimulus appeared during the IST. Using only a 900-msec IST, Law, Pra tt and Abrams (1995) reported slower reaction times (RTs) when cue and targ et were the same color, but only when the task-irrelevant filler was presen ted during the ISI. They argued that attention is first drawn to the cue co lor and that inhibition of return (IOR; see Posner & Cohen, 1984) is establ ished when attention is drawn away from that cue color representation by th e task-irrelevant filler Critical to their view is the assumption that IOR occurs only after attention is drawn away from the cue color by the filler. Assuming a time course for the withdrawal of attention from the cue color representation, Law et al.'s view predicts growth of the inhibitory effect as a function of ISI in the filler condition as well as facilitation at ear ly ISIs in the no-filler condition (because there is no filler to withdraw attention fr om the cue color). Contrary to these predictions, we found tha t the inhibitory effect observed by Law et al. at the 900-msec ISI was pres ent at-and did not vary in magnitude across-the range of ISIs tested. And t here was never facilitation in the no filler condition. These results chall enge Law et al.'s inference that IOR for foveally presented colors was oper ating in their paradigm.