Abstract
It has been shown in both animal and human studies that attentional selection of simple features (e.g. color) enhances processing of this feature across the entire visual field, i.e. it is spatially global. However, it remains unclear whether such global facilitation of features is mandatory or whether it can remain spatially local when the task requires it. To address this question, we presented two completely overlapping fields of red and blue dots, one on each side of a central fixation cross. Participants performed a divided attention task in which they had to attend to the dots of one color or the other in both visual fields concurrently. Trials in which participants attended to the same color on both sides ('attend same') were characterized by good behavioral performance and clear enhancement of steady-state visual evoked potentials (SSVEPs) elicited by attended stimuli. However, when the attended color on one side was to be ignored on the other side ('attend opposite'), there was a dramatic drop in behavioral performance and attentional enhancement of SSVEP-amplitudes vanished. A control experiment confirmed that this was due to global spread of feature attention leading to equal enhancement of attended and unattended colors in 'attend opposite' trials rather than a failure to concurrently attend to different colors in the left and right fields. Our results thus demonstrate that spatially global selection of features is a fundamental process which even occurs when it explicitly conflicts with task demands.
Meeting abstract presented at VSS 2012