Abstract
In an attempt to determine the size of the spatial window of attention we required subjects to identify 2 target letters from a display of 24 letters, expecting performance to improve the closer the targets were to each other. The results were opposite to this expectation.
The 2 target letters were chosen randomly from a circular display (radius 4 deg) containing 24 letters. The display (duration 100 to 300 ms) was preceded and followed by masks. Attention was directed to the target locations by means of either a colour cue or a numeral cue. Identification of target letters improved as the directional separation between targets was increased from 15° to 180°. Performance was better when targets were in different hemifields but also improved with increasing separation within a hemifield.
The results demonstrate interference at an attentional level analogous to lateral sensory masking. Attending to one item does not enhance recognition of nearby items, as might be expected with an expandable attentional beam or zoom lens, but rather interferes with the perception of nearby items. This outcome is consistent with the idea that nearby locations share limited processing resources and with neurophysiological findings on attentionally-mediated changes in receptive fields.
Get full access to this article
View all access options for this article.