Attentional selection and attentional gradients: an alternative method for studying transient visual-spatial attention

Event-related potential (ERP) effects of transient and sustained non-spatial visual attention were investigated in an experiment where subjects were instructed to attend to the color or form of visual stimuli in order to detect infrequently presented targets with the relevant feature. The to-be-attended feature was either varied in a trial-by-trial fashion (transient attention) or was kept constant for an entire experimental block (sustained attention). Both transient and sustained attention resulted in a larger negativity for attended as compared to unattended stimuli between 200 and 300 ms post-stimulus. This effect was more pronounced for sustained attention than for transient attention and larger for attention to color than for attention directed to stimulus form. In the sustained, but not in the transient attention condition, color attention resulted in larger positivities for attended stimuli between 150 and 200 ms and in the P3 time range. These effects are interpreted as evidence for the existence of non-spatial attentional selection processes that are more effective under sustained than under transient attention conditions and are different from processes of visual-spatial attention. Moreover, the results indicate a special status for sustained attention to stimulus color.