Purchase this article with an account.
Peter Tse, Florian Baumgartner, Mark Greenlee; fMRI BOLD signal reveals neural correlates of microsaccades. Journal of Vision 2007;7(9):318. doi: 10.1167/7.9.318.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Microsaccades are small, conjugate, involuntary eye-movements made while voluntarily fixating, which play a role in minimizing perceptual fading. The goal of this research was to determine the neural correlates of microsaccade occurrence in early visual areas of cortex using fMRI. This is an important issue for fMRI researchers because so far no laboratory has controlled for the possibility that microsaccade events, rates or magnitudes are correlated with experimental conditions. If microsaccades are found to generate BOLD signal changes in the cortex, many past fMRI results may have arisen as a result of this confound. Methods: We recorded both involuntary microsaccades, and voluntary saccades from one eye in the spatiotemporal domain of microsaccades using a Limbus infrared eyetracker (1000Hz sampling), while collecting fMRI data in a mixed, event-related/block design (3T Siemens Allegra scanner, TR=402ms, 11 slices along calcarine, TE=30ms, FA=35°, 800 volumes, n=6, 6–8 runs per subject). In two blocks per run subjects executed small voluntary saccades (0.16°, 12 pseudorandom events per block), maintained fixation on a small point jumping left and right. A long fixation-only period separated these blocks. The fixation point was centered on a 1° wide horizontal white band laid over a polar grating. Retinotopic mapping used standard methods (TR=2000ms, TE=30ms, FA=90°, 30 slices, 152 volumes). Results: BOLD signal is greater in early visual areas for very small voluntary saccades, relative to fixation epochs, which are as small as true microsaccades. An event-related deconvolution analysis of true microsaccades that occurred during fixation-only epochs revealed upward modulation of the BOLD signal in V1 and V2 after the occurrence of a microsaccade. We conclude that, to the extent that microsaccades may be correlated with experimental conditions, the results of fMRI studies may arise because of microsaccades, and not the experimental variables under consideration, forcing a reevaluation of many past fMRI results.
This PDF is available to Subscribers Only