Academic journal article Psychonomic Bulletin & Review

The Time It Takes to Switch Attention

Academic journal article Psychonomic Bulletin & Review

The Time It Takes to Switch Attention

Article excerpt

An experiment is reported that measured the time it takes to switch attention from one set of locations to another in response to a cue that indicates the relevant locations. The experiment compared sequences of trials in which the same locations were cued in succession with sequences in which different locations were cued in succession in order to separate cue-encoding time from attention-switching time. Same-location sequences require cue encoding but not attention switching. They were substantially faster than different-location sequences, which require both cue encoding and attention switching. Formal models were fitted to time-course functions generated by presenting the cues 0,100, 200, 300, or 400 msec before the target displays. The model fits suggest that cue encoding took 67-74 msec and attention switching took 76-101 msec.

The time it takes to switch attention has been measured in many ways, and the measurement is controversial. In visual search experiments, it takes 30-50 msec to reject a distractor (Wolfe, 1998). That interval measures switching time only if distractors are examined serially and individually, which are controversial assumptions (Townsend, 1990; Treisman, 1982). In attentional blink experiments, it takes 500-600 msec to finish processing one target and to begin processing another (Ward, Duncan, & Shapiro, 1996). That interval confounds switching time with target processing time and masking (Moore, Egeth, Berglan, & Luck, 1996). This article examines a procedure popularized by Eriksen and colleagues, in which a cue indicates the position of a target in a multielement display (Colegate, Hoffman, & Eriksen, 1973; Eriksen & Collins, 1969; Eriksen & Hoffman, 1972). The interval between the onset of the cue and the onset of the target (stimulus onset asynchrony, or SOA) is manipulated to measure the time-course of switching attention. Performance improves as SOA increases and reaches asymptote at long SOAs. Researchers often interpret the SOA at which performance reaches asymptote as a measure of the time it takes to switch attention (Colegate et al., 1973; Tsal, 1983). Typical values range from 150 to 300 msec.

There are two problems with using the asymptotic SOA as an estimate of attention switching time. First, it is difficult to estimate the asymptote without fitting an explicit function to the curve. The asymptotic SOA is often defined as the point at which differences between successive SOAs first become significant (Tsal, 1983). This is questionable statistically (Sperling & Weichselgartner, 1995). second, switching time is probably a random variable, so the asymptotic SOA reflects the upper limit of the distribution of switching times rather than the mean (Sperling & Weichselgartner, 1995). This article addresses these problems by adapting methods for analyzing time-course functions from Logan and Bundesen (2003). These methods treat the time-course function as the cumulative distribution of finishing times for switching attention, which reflects the probability that the subject has switched attention to the cued location at a given SOA. Explicit functions are fitted to the time-course function and parameters of the functions are interpreted as measures of switching time.

There is another problem: The time-course function reflects the time required to encode the cue as well as the time to switch attention. Cue-encoding time must be a random variable, so the time-course function reflects the probability that the cue was encoded at a given SOA as well as the probability that attention was switched. I address this problem by adapting procedures and models from the explicit task-cuing procedure in the taskswitching literature. In this procedure, a cue indicates which task to perform on a subsequent target. To estimate task-switching time, researchers compare trials on which tasks repeat with trials on which tasks change. Cue encoding occurs whether tasks repeat or change, but task switching occurs only when tasks change. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.