Visual search is something we do routinely every day. We select a coin for the parking meter from a handful of coins, or we single out an icon from among a crowd of other icons on our computer desktop. Laboratory studies have been aimed at discovering the rules that govern the efficiency of visual search. One such rule, known as inhibition of return (IOR;Posner & Cohen, 1984), is held to be effective in preventing observers from reexamining locations that have already been searched. A rule such as this would promote the examination of novel locations, thereby improving search efficiency.In a typical IOR experiment, the initial display consists of a central fixation point and two placeholders, one on either side of fixation. A cue stimulus appears briefly in one of the peripheral locations and then reappears briefly at fixation. A target stimulus is then presented, and the observer is required to respond to it as quickly as possible. IOR describes the finding that the observer's reaction time (RT) to a target presented at the previously cued location is slower than it is to a target presented at an uncued location (Posner & Cohen, 1984).In the predominant view, this increased latency is attributed to some form of inhibitory mechanism acting at either the perceptual (see, e.g., Abrams & Dobkin, 1994) or the motor (see, e.g., Taylor & Klein, 1998) level. However, a determination of whether the locus of IOR is perceptual or motor (or both) has been hampered by the common practice of using RT as the dependent measure. Because motor factors are inherently involved in RT, an uncontaminated estimate of the involvement of perceptual factors has been difficult to obtain (see, e.g., Müller & von Mühlenen, 1996).The perceptual and motor components of IOR were decoupled in a study by Handy, Jha, and Mangun (1999), who used identification of a masked target, rather than RT, as the primary response measure. By employing signaldetection methodology, a measure of perceptual sensitivity (d ) was obtained that was independent of motor factors. Handy et al.'s results revealed both classes of effects typically found in exogenous cuing studies: facilitation (larger d at the cued relative to the uncued location) at a short cue-target onset asynchrony (CTOA) and IOR (relatively smaller d at the cued location) at a long CTOA. On the basis of these results, Handy et al. concluded that the mechanism underlying IOR affects the perceptual quality of a target stimulus independently of motoric factors.Although the signal-detection methodology (d measure) employed by Handy et al. (1999) did provide evidence for a perceptual component of IOR and for reduced sensitivity at the cued location, it did not provide evidence regarding the period of time required for the perceptual processing of the target at the cued and uncued locations. The present need is for estimates of the speed of the perceptual system at the cued relative to the uncued target locations in the absence of motoric factors. If it is the case that IOR reflects a modulation of perce...