Abstract
Purpose: Neural microstimulation experiments in the superior colliculus (SC) of primates enhanced performance in a spatially selective manner while they maintained fixation, suggesting a role for covert visual attention in SC (e.g., Mü;ller et al, PNAS 102, 524, 2004; Cavenaugh, et al., J Neurosci 26, 11347, 2006). Previous fMRI experiments have shown a rough retinotopic organization in human SC corresponding to direct visual stimulus (Shneider & Kastner, J Neurophysiol 94, 2491, 2005). We performed experiments to determine if retinotopically organized signals corresponding to covert visual attention were present in human SC. Methods: We measured the retinotopic organization of SC to direct visual stimulation using a broad wedge of flickering dots (eccentricity 1–9°, azimuthal width 144°) that slowly rotated (24-s period) around the fixation mark. To measure the retinotopy of covert attention, we used a full-field flickering (4-Hz) grating (1 cpd) stimulus at 90% contrast. Subjects were cued to perform a difficult orientation-discrimination task within a region (eccentricity 1-9°, azimuthal width 90°) that slowly rotated (24-s period) in azimuth around fixation.. High-resolution fMRI (1.3 mm voxels) was acquired (3 s/volume) in 8 slices within SC using a 3 shot spiral sequence and a TE of 45 ms. Results: We obtained detailed retinotopic maps of covert attention in the superficial layers of SC. These maps were in good registration with those corresponding to direct visual stimulation. There was evidence for weaker retinotopic responses to both direct visual stimulation and covert attention in the deep layers, but little (if any) response in the intermediate layers. Conclusions: Signals corresponding to covert visual attention are present in the superficial layers of SC. These signals are retinotopically organized in register with the signals produced by direct visual stimulation. Similarly organized signals may be evident in the deep layers.