Abstract
Change-detection paradigms have become a popular means of assessing the nature of visual working memory, but little research has addressed the mechanisms by which changes are actually detected. That is, how is the memory of the previous stimulus array compared with the perceptual representation of the current stimulus array? In the present study, we sought to determine whether the comparison process is limited or unlimited in capacity. To accomplish this, we made use of the fact that the detection of a change triggers a shift of attention to the changed object, as indicated by the presence of an attention-related ERP component (the N2pc wave) over the hemisphere contralateral to the changed item. If the comparison process is limited in capacity, then N2pc latency should increase as the number of items to be compared increases. However, if the comparison processes is not limited in capacity, then N2pc latency should be independent of the number of comparisons required.
Two change-detection experiments were conducted that required subjects to remember a sample array containing several oriented bars and compare it to a test array presented after a 900-ms retention interval. In the first experiment, the orientation of one bar changed on half the trials, and subjects reported the presence or absence of a change. In the second experiment, one bar always changed in orientation, and subjects reported whether this change occurred in the left or right hemifield. In both experiments, an N2pc component was elicited by the changed item in the test array, and its onset time (ca. 200 ms) was essentially constant across set sizes of one to four objects. These results indicate that the comparison between the memory representation of the sample array and the perceptual representation of the test array was rapid and unlimited in capacity.