Programming shifts of spatial attention

Abstract

Two experiments are described, which were aimed at studying attention allocation at different times following directional cognitive cue presentation. In Experiment 1, the stimulus display consisted of a fixation cross and four boxes for stimulus presentation. The boxes were arranged horizontally above fixation. In Experiment 2, there were two rows of boxes, one above and one below fixation. In both experiments, the subjects were asked to shift attention from fixation to the cued box and to respond manually as fast as possible to the occurrence of a visual imperative stimulus, regardless of its location. The imperative stimulus was presented in the cued box or in one of the non-cued positions. The time interval between cue and imperative stimulus onset (SOA) varied from 100 to 700 msec. The results showed a striking difference in the pattern of attention allocation following short (100-160 msec) and long (300-700 msec) SOAs. Following short SOAs, attention spanned a relatively large portion of space, with the locations closer to fixation producing the fastest RTs. With long SOAs (300, 500, or 700 msec), attention was focused on the cued box. We propose that direction and amplitude of an attention movement are initially computed in such a way as to form a broad, attentional sector within which the cued location is included. All locations that lie within this sector initially get attention benefits. Then, attention is tuned more narrowly. After a delay of about 300 msec, the computation of attention direction and amplitude is completed and attention is restricted to the cued location.