Purchase this article with an account.
Vivian Ciaramitaro, Geoffrey Boynton; Behavioral measures of cross-modal attention are consistent with fMRI responses in V1 and not MT+. Journal of Vision 2007;7(9):178. doi: https://doi.org/10.1167/7.9.178.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: In a previous fMRI study, we showed that the response to an unattended visual stimulus depended on the modality subjects were attending to, and that this dependence varied across visual areas. In V1/V2, the response to an unattended stimulus was weakest when another visual stimulus was attended. However, in MT+, the response to an unattended stimulus was weakest when an auditory stimulus was attended. This leads to the question, which areas, higher or lower, best represent the perceptual effects of cross-modal attention? In the present study, we used a visual motion aftereffect (MAE) to measure the response to an unattended visual stimulus as a function of the attended modality. Will our psychophysical measures of attention agree with our fMRI measurements in V1 or MT+?
Methods: Two drifting gratings, to the left and right of fixation, and auditory stimuli (binaural or dichotic) were presented simultaneously. Subjects were cued to attend to one stimulus and perform either a speed discrimination task on a visual stimulus or a frequency discrimination task on an auditory stimulus. The contrast of the visual gratings varied from trial to trial. At the end of each trial, the strength of the MAE was measured using a nulling procedure in which subjects judged the direction of motion of a fixed contrast grating presented unpredictably at one of the two possible grating locations.
Results: The strength of our MAE at an unattended visual stimulus was weaker when attending to another visual stimulus compared to attending to an auditory stimulus. This is consistent with our fMRI measurements in V1/V2. Furthermore, the MAE strength increased gradually as a function of stimulus contrast, which is again consistent with contrast-response functions in V1, not MT+. We therefore conclude that our MAE results reflect responses in earlier stages of visual processing.
This PDF is available to Subscribers Only