Separate coding of different gaze directions in the superior temporal sulcus and inferior parietal lobule

Abstract

Electrophysiological recording in the anterior superior temporal sulcus (STS) of monkeys has demonstrated separate cell populations responsive to direct and averted gaze. Human functional imaging has demonstrated posterior STS activation in gaze processing, particularly in coding the intentions conveyed by gaze, but to date has provided no evidence of dissociable coding of different gaze directions. Because the spatial resolution typical of group-based fMRI studies (approximately 6-10 mm) exceeds the size of cellular patches sensitive to different facial characteristics (1-4 mm in monkeys), a more sensitive technique may be required. We therefore used fMRI adaptation, which is considered to offer superior resolution, to investigate whether the human anterior STS contains representations of different gaze directions, as suggested by non-human primate research. Subjects viewed probe faces gazing left, directly ahead, or right. Adapting to leftward gaze produced a reduction in BOLD response to left relative to right (and direct) gaze probes in the anterior STS and inferior parietal cortex; rightward gaze adaptation produced a corresponding reduction to right gaze probes. Consistent with these findings, averted gaze in the adapted direction was misidentified as direct. Our study provides the first human evidence of dissociable neural systems for left and right gaze.

Figure 1

Experimental Design and Example Stimuli The adaptation experiment comprised five sections—an initial preadaptation phase to familiarize the subjects with probe stimuli and task and four adaptation phases comprising two sections each. The format of the trials in each section is illustrated in Figure 1. The preadaptation phase comprised four presentations of the 12 models (six male and six female) posing three gaze directions (10° left, 0° (direct), and 10° right; 144 stimuli in total). Trials consisted of a 1500 ms. presentation of a probe face and then a blank intertrial interval (ITI) of 1000 ms. Presentation order was randomized, and subjects pressed one of three keys with their right hand to indicate gaze direction. There were four adaptation phases, ordered either as LRRL or RLLR (where L = left adaptation and R = right adaptation) and counterbalanced across subjects. The second and third phases were separated by short breaks. Each phase had the same basic structure and comprised two sections. Section 1 contained two 4000 ms presentations of each of the 12 models gazing in one consistent direction—25° left or 25° right). Subjects were instructed to stare at the eye region of each face, and no response was required. Trials in section 2 consisted of a “top-up” adaptation face (4000 ms) gazing 25° in the adapted direction (i.e., same direction as in section 1) and, immediately after this, a probe face (1500 ms), and then a blank ITI (1000 ms). Probes were identical to those from the Pre-adaptation phase (12 models × three gaze directions [10° left, 0°, and 10° right] × two presentations; 72 stimuli in total), and subjects categorized their gaze as left, direct, or right. The top-up adaptation and probe faces were of similar size and were shown in the same central position, but they never had the same identity. In addition, vertical eye position and interocular distance were deliberately not standardized across identities to ensure that switching between the top-up and probe faces did not induce perception of apparent gaze motion . Probe faces were always identified with a bold outline as illustrated. All images were 256 grayscale. The whole experiment lasted just over 1 hr.

Figure 2

Behavioral Data Mean percentage of correct gaze responses to probe faces (10° left, direct, and 10° right) as a function of the direction of gaze adaptation (left and right). Performance for the same probe faces in the preadaptation phase is also shown for comparison. Error bars show standard errors. For the adaptation graph, LL = left adaptation-left gaze probe, LD = left adaptation-direct gaze probe, and so on. For the preadaptation data (right graph), L = left probe, D = direct probe, and R = right probe.

Figure 3

Sagittal, Coronal, and Transverse Slices through the Anterior STS and IPL (A) The right anterior superior temporal sulcus (57, 9, −27): Sagittal and transverse sections on the mean across subjects of their normalised mean EPI image, and a coronal section of a canonical T1-weighted image (both in MNI space). (B) The right inferior parietal lobule (60, −54, 30): Sagittal and transverse sections on a canonical T1-weighted image in MNI space and a coronal section of the mean, across subjects, of their normalized mean EPI image in MNI space. Both are thresholded at p < .005 (5 contiguous voxels) for purposes of illustration.

Figure 4

Neuroimaging Data Mean event-related response to each of the three types of probe faces (10° left, direct, and 10° right) as a function of the direction of gaze adaptation (left and right) for (A) the maximally activated voxel in the right anterior STS (RSTS; 57, 9, −27) and (B) right inferior parietal lobule (RIP; 60, −54, 30). Also shown for the same voxels is the mean event-related response to the same probe faces in the preadaptation phase. The y axis represents estimated peak percent signal change relative to the average over all voxels and scans; error bars show standard error of the mean, between-subject differences having been removed. For the adaptation data (left graphs), LL = left adaptation-left gaze probe; LD = left adaptation-direct gaze probe, and so on. For the preadaptation data (right graphs), L = left probe, D = direct probe, and R = right probe.