Social status gates social attention in monkeys

Humans rapidly shift attention in the direction other individuals are looking, following gaze in a manner suggestive of an obligatory social reflex [1–4]. Monkeys’ attention also follows gaze, and the similar magnitude and time-course of gazefollowing in rhesus macaques and humans [5] is indicative of shared neural mechanisms. Here we show that low-status male rhesus macaques reflexively follow the gaze of all familiar rhesus macaques, but high-status macaques selectively follow the gaze only of other high-status monkeys. These results suggest that gaze-following in monkeys involves reflexive and voluntary components, and that the strength of these mechanisms varies according to social status. We probed the impact of social status on gaze-following in rhesus macaques performing a simple visual orienting task [5] (Figure 1A). Each monkey (four highstatus, three low-status) fixated a central target which was replaced by an image of a familiar monkey’s face looking left or right. After 100, 200, 400, 600 or 800 ms, the face disappeared and a peripheral target appeared randomly to the left or right; monkeys then shifted gaze to the target to receive a juice reward. Crucially, the face’s gaze direction did not predict the target location. We demonstrated previously that saccade reaction times for monkeys and humans are faster on those trials in which observed gaze is congruent with target location; reaction time savings thus served as our operational definition of gazefollowing. We hypothesized that monkeys would generally follow gaze, but that the strength and timing of gaze-following would be modulated by social status. Figure 1. Observed gaze and social status influence saccade reaction times. (A) Visual orienting task. The impact of social status on gaze-following was examined by comparing reaction times for saccades made to a peripheral target after viewing an image of a familiar monkey looking left or right. Each monkey first fixated a central yellow square (±3°) for 200–500 ms. The yellow square was then extinguished and a monkey face, shown here in inset, was illuminated centrally for a variable duration. If the monkey maintained fixation, the face was extinguished and a peripheral yellow square simultaneously illuminated at one of two fixed positions located symmetrically 10–15° to the left or right. Gaze shifts to the peripheral target within 350 ms were rewarded with a small squirt of juice. (B) Gaze-following dynamics vary with social status in male rhesus macaques. Average (±S.E.M.) saccade reaction times plotted as a function of face viewing duration on congruent and incongruent trials, for high-status (left) and low-status (right) subjects. For high-status monkeys, gaze cues evoked significant reaction time savings only at the 200 ms viewing duration. For low-status monkeys, both early reaction time savings and later inhibition of return (IOR) were observed. **post-hoc test p < 0.00005; *post-hoc test p < 0.005; all other contrasts p > 0.05. C on gr ue nt In co ng ru en t