Abstract
Purpose: We examined the neuronal mechanisms of cross-modal and uni-modal attentional selection by measuring fMRI responses to ignored visual stimuli when attention was directed toward another stimulus either in the same modality or a different modality. Methods: Subjects viewed 3 stimuli presented simultaneously: a drifting sinusoidal grating to the left of fixation, one to the right of fixation and a binaural auditory tone. On any given trial, subjects were cued to attend to one of the three stimuli and judged which interval contained either the faster speed or the higher frequency, depending on the cue. Our design allowed us to associate fMRI responses to the location of attention rather than changes in the stimulus or task difficulty. The cue alternated between 2 of the 3 stimuli in 20-sec blocks (6 blocks per scan, 8 trials per block) providing relative fMRI responses to an attended visual stimulus, and to an ignored visual stimulus when attention is either directed toward another visual stimulus, or toward an auditory stimulus. Results: In early retinotopic visual areas (V1, V2, and V3) we found smaller fMRI responses to an ignored visual stimulus when attention was directed toward the other visual stimulus. However, in area MT+, fMRI responses to an ignored visual stimulus were smaller when attention was directed to the auditory stimulus. We further quantified the magnitude of these effects using an event-related design. Our results are consistent with a model of attention in which within-modality suppression is greater than cross-modal suppression only in early visual areas.
Supported by NIH-NEI 12925