Rhesus macaques recognize unique multimodal face-voice relations of familiar individuals and not of unfamiliar ones.

Communication signals in nonhuman primates are inherently multimodal. However, for laboratory-housed monkeys, there is relatively little evidence in support of the use of multimodal communication signals in individual recognition. Here, we used a preferential-looking paradigm to test whether laboratory-housed rhesus monkeys could 'spontaneously' (i.e., in the absence of operant training) use multimodal communication stimuli to discriminate between known conspecifics. The multimodal stimulus was a silent movie of 2 monkeys vocalizing and an audio file of the vocalization from one of the monkeys in the movie. We found that the gaze patterns of the monkeys that knew the individuals in the movie were reliably biased toward the individual that did not produce the vocalization. In contrast, there was not a systematic gaze pattern for the monkeys that did not know the individuals in the movie. These data are consistent with the hypothesis that laboratory-housed rhesus macaques can recognize and distinguish between conspecifics based on auditory and visual communication signals.