The time course of attentional zooming: a comparison of voluntary and involuntary allocation of attention to the levels of compound stimuli.

The time course of attentional zooming between the levels of hierarchically structured compound stimuli (level-specific covert orienting of attention) is explored experimentally. The experiment compares the RT-SOA functions of voluntarily and involuntarily initiated level-specific reorienting within a cost-benefit experiment using level-specific cues. With respect to the two modes of initiation, the results reveal no functional differences between attention shift and attentional zooming. Both can be initiated either voluntarily or involuntarily, the latter mode dominating the former; for both, involuntary initiation produces faster reorienting of attention than does voluntary initiation, and for both, involuntary initiation is more effective than voluntary initiation. However, the time needed to complete attentional orienting is about twice as long for zooming than for shifts. This quantitative difference suggests that there is a functional difference between level-specific and horizontal covert orienting. The difference is explained by postulating that zooming and attention shifts differ in the number of parameter sets that have to be adjusted in the reorienting of attention. The experiment also reveals that attentional zooming to the local level needs more time than does zooming to the global level. This result gives some support to the hypothesis that the RT difference between global and local identifications (Navon's global-dominance phenomenon) is due to an additional step in the course of reorienting attention away from the global level (which is usually attended to first) to the local level, when this level is to be identified.