ARE ATTENTIONAL DWELL TIMES INCONSISTENT WITH SERIAL VISUAL-SEARCH

Duncan, Ward, and Shapiro (1994) estimated that attention must remain focused on an object for several hundred milliseconds before being shi fted to another object, and they referred to this period as the attent ion at dwell time. An important implication of these long estimates of the dwell time for models of visual search is that the search process must not involve an item-by-item serial scanning mechanism. If it did , then searching through an array of items would require enormous amou nts of time, which-based on data from visual search experiments-it doe s not, The present report, however, provides evidence that the long es timates of attentional dwell time were caused, at least in part, by th e use of masked targets. Implications of these variable estimates of t he attentional dwell time for models of visual search are discussed.