Innate sensitivity to face-to-face biological motion

Summary Sensitivity to face-to-face stimuli configurations, which likely indicates interaction, seems to appear early in infants’ development, and recently a preference for face-to-face (vs. other spatial configurations) has been shown to occur in macaque monkeys. It is unknown, however, whether such a preference is acquired through experience or as an evolutionary-given biological predisposition. Here, we exploited a precocial social animal, the domestic chick, as a model system to address this question. Visually naive chicks were tested for their spontaneous preferences for face-to-face vs. back-to-back hen dyads of point-light displays depicting biological motion. We found that female chicks have a spontaneous preference for the facing interactive configuration. Males showed no preference, as expected due to the well-known low social motivation of males in this highly polygynous species. These findings support the idea of an innate and sex-dependent predisposition toward social and interacting stimuli in a vertebrate brain such as that of chicks.


INTRODUCTION
The brains of young vertebrate animals are equipped with animacy detectors, mechanisms to recognize the presence of other animals in the environment.][12][13][14][15] This sensitivity to animacy signals not only helps organisms to spot the presence of other living beings in their vicinity but also lays the foundation for the development of social interactions. 16,17Stimuli involving multiple agents that interact with each other may provide information about social events.One example is when two bodies are positioned face-to-face and therefore suggest an interaction compared to a back-to-back positioning. 18,19n humans, visual sensitivity to the difference between face-to-face and back-to-back bodies appears early in development 18 and evolves into a preference for face-to-face bodies, which is shared with monkeys. 20In those studies, the looking times of six-month-old infants were computed while presenting simultaneously two dyads showing face-to-face and back-to-back bodies, respectively, finding a significant difference in looking times toward the two types of dyads. 18The same looking-time paradigm was adopted in adult humans and macaques, confirming the different attention driven by the two types of stimuli, and thus, the relevance of such configurations. 20Those results are compatible with the evidence of perceptual adaptation for efficient processing of seemingly interacting (e.g., face-to-face) bodies.In fact, it has been shown that, in conditions of visual noise, facing bodies are more likely to be detected and recognized than the same two bodies facing away from each other. 21,22Other studies showed that the face-to-face positioning of bodies impacts the very early, preattentive stages of visual perception, 19,23 up to visual memory. 24,25The behavioral advantage in processing facing people has a counterpart in neuroimaging results showing that, in humans, body-selective regions of the visual cortex respond more strongly to multiple bodies that appear to be interacting (i.e., face-to-face), relative to unrelated bodies. 26,27owever, whether the visual preference for face-to-face bodies is acquired through experience or rather represents an example of an evolutionarily given predisposition is unknown, and it is the issue we aim to address here.
We tested newly hatched visually naive chicks for their spontaneous preferences for a pair of biological motion patterns representing two point-light hen silhouettes walking either face-to-face or back-to-back (Figure 1A).We measured the first approach and time spent near the two stimuli in both male and female newly hatched chicks.

RESULTS
To test whether the chicks were more likely to first approach the face-to-face over the back-to-back stimuli, we tracked the animals' first choice (Figure 1B).
After the first approach, choice proved to stabilize over time.Indeed, we performed a longitudinal analysis with time and sex as factors, to investigate whether response changes across the 30 min of test.This did not reveal any significant main effect (time: 1.5, df = 3.6, p = 0.2; sex 2.5, df = 1, p = 0.1; interaction: 0.8, df = 3.6, p = 0.5).Still, a trend diverging from random choice after the first minutes could be observed in female (Figure 1C).
To test whether the chicks were more likely to spend most of the overall time close to the face-to-face over the back-to-back stimuli, we computed the proportion of time animals spent close to the face-to-face configuration during the overall 30 min of the experiment (Figure 1D).][30][31] A permutation test on the proportion of time spent close to the stimuli revealed a sex difference (permutation p = 0.047, effect size Cohen's d = 0.2).Females spontaneously chose to spend more time near the face-to-face configuration (Z = À2.16,p = 0.03, effect size Cohen's d = 0.2; post-hoc achieved power 60%), while males did not exhibit any preference (Z = À0.18,p = 0.9, effect size Cohen's d = 0.01).

DISCUSSION
We found that female chicks spontaneously preferred to remain close to a pair of face-to-face hen silhouettes rather than back-to-back.This finding provides evidence that the recognition of face-to-face configuration, likely indicative of interaction, which has been found in infants, adult humans, and adult macaque monkeys, is observed also in birds and does not derive from experience but it is likely to be biologically predisposed in the vertebrate brain.Moreover, given that previous studies in infants never reported a preference for face-to-face stimuli, but only a capacity for discrimination, 18,20 the present study is the first report of such an early preference.
While previous studies have demonstrated discrimination and preference for biological motion versus random motion and inverted biological stimuli (reviewed by Vallortigara 17 ), our study is also the first to reveal a preference for ''face-to-face'' interactions between agents over ''back-to-back'' interactions.Interestingly, this finding shows a different relevant level of information than all previously cited works that investigated spontaneous preferences, e.g. the tendency to approach possible instead of impossible figures, 32 to prefer consonant music, 33 or to complete over fragmented objects. 34Indeed, in all these cases, chicks were proved to show an innate sensitivity to ''natural'' configurations in the environment that meet the Gestalt rules of continuity, proximity, and order 35 as well as the expectation of consonance with the physical rules governing nature.With the present findings, instead, we showed a level of predisposition that reflects expectations about the social world, in opposition to the expectations about the physical world.This could also explain sex differences in this specific predisposition.Indeed, our experiment revealed a divergence between females and males, with the latter showing no preference.This finding is interesting in itself and aligns with the species' natural history and the existing ethological literature [28][29][30][31]36 (see also for non-human primates Brown et al. study 37 ), which indicates that males typically display lower levels of interest in social stimuli and interactions with them compared to females, as part of being a strict polygynous species. Ineed, within adult fowls, there is usually a dominant rooster and many hierarchically organized hens, 38 favoring the evolution of solitary territorial males and more socially prone females.39,40 Moreover, the two kinds of stimuli used here clearly provided perceptual cues associated with affiliative responses, which are known to be stronger in females. Thre is evidence that chicks tend to align to the apparent direction of movement of point-light displays.14 Thus, facing dyads would elicit approach responses, irrespective of which of the two point-light hens the chick aligns to; instead, non-facing dyads would elicit withdrawal responses.
Additionally, an approach is needed to perform social pecking, which is the main tool for social recognition in young chicks. 41Again, social pecking, which is more pronounced in females, 30 requires proper orienting toward conspecifics, as in facing dyads.
Of course, social interaction is a very general term and may comprise different aspects.For instance, point-light displays facing in the same direction may signal the increased likelihood of a resource between the hens (e.g., corn on the ground between them) that the chicks may be attracted to.This seems unlikely, however, because newly hatched chicks have reserves in the yolk sac for the first two days and are primarily interested in imprinting on a social partner rather than on food at this age. 41Besides, there are no obvious reasons why females should be more interested in food than males.It is more likely that chicks may prefer to congregate with groups of hens, which should motivate attraction to two facing hens rather than two hens walking in opposite directions.Whatever the underlying motivation, it is clear that the face-to-face signal to which chicks respond reflects a predisposition in their brain.

Limitations of the study
The present paper only deals with the issue of the origin (innate or learned) of the face-to-face preference and not with the particular visual cues that generate it (see for specific investigations on this topic Goupil and Papeo works 18,19,21 ).However, the fact that only females exhibit the preference supports the idea that it is the biological meaning of the face-to-face interaction that elicits interest and not some low-level aspect of the visual stimulation.

STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:

Figure 1 .
Figure 1.Stimuli and chicks' spontaneous choice (A) Example of a frame for face-to-face (left) and back-to-back (right) configurations.(B) First choice proportion for face-to-face configuration.Proportion of time spent to face-to-face configuration (preference index) as a function of time (C) and overall for the 30 min of test (D; meanGse, * p < 0.05).GIF files with the moving stimuli are available in the data repository Figshare: https://figshare.com/articles/dataset/Innate_sensitivity_to_face-to-face_biological_motion_dataset/23713299.

TABLE
d RESOURCE AVAILABILITY B Lead contact B Materials availability B Data and code availability d EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS d METHOD DETAILS d QUANTIFICATION AND STATISTICAL ANALYSIS