Introduction

For group-living animals, primates included, the recognition of conspecifics is crucial for their survival. Humans have specialized brain areas to recognize faces[1] and whole bodies[2–5] and their expertise in face recognition is demonstrated by the ‘inversion effect’, showing that faces and whole bodies, but not objects, are recognized configurally rather than by their parts[6–8]. Importantly, their recognition is disproportionally impaired, relative to objects such as houses or cars, when they are seen inverted rather than upright[6]. Conclusive evidence has shown that this effect is primarily due to a disruption in the processing of configural, rather than featural, information in faces [e.g., [9–13]. The face inversion effect has been observed in chimpanzees too, and although not all chimpanzees show this effect at all times[14, 15], overall there is evidence that configural processing is a critical element of efficient face detection in chimpanzees as well[16, 17]. Thus, effects of inversion have been observed for faces and whole bodies, but are generally not found for individual body parts[18]. Intriguingly, previous studies included almost all body parts, except the most obvious one, which is the behind, as we will outline below.

Previous research has shown that in recognizing each other, chimpanzees do not rely on the face alone[14], but also easily recognize each other by their behinds[19]. Most non-human female primates, chimpanzees included, show a swelling and reddening of the anogenital region around the time of ovulation[20]. At some point during human evolution, these changes in size and color along the menstrual cycle have disappeared, and large quantities of ‘permanent’ adipose tissue on the behind emerged[21, 22]. Possibly, this became more adaptive when our species started to walk upright, or to hide oestrus as to be attractive for males throughout the menstrual cycle and foster pair bond formation and shared caring for offspring. To date, it is not known how behinds as compared to faces are recognized in humans and their closest relatives, but this knowledge can enhance our understanding of the evolution of face processing, as we will argue below.

Face recognition plays an incredibly important role in the survival of animals living in social groups, including humans and chimpanzees. The changeable properties of faces like expression and gaze, display emotions and intentions and are used by observers to predict behavior[23]. The more or less invariant properties of faces are used for identification and display physical characteristics, including sex, age and attractiveness[1].

The primate behind is unlike any other body part. In chimpanzee females, as in other catarrhine primate species, the anogenital region swells, becomes shiny and smooth and reddens around the time of their ovulation. Male mating interest is positively correlated with these changes, but the changes do not go unnoticed by competing females either[20]. Besides accumulating adipose tissue on their behinds, which might have been a sign of fitness in harsh savanna conditions, human females also developed relatively large breasts and are unique amongst primates to develop these already before their first pregnancy[22]. It has been suggested that breasts evolved to resemble the bottom, being more visible when walking upright[21, 24]. Also, humans, especially females, developed reddened and thicker lips and fattier faces as compared to chimpanzees, which is also a sign of beauty and these features are often accentuated with make-up[25]. (On a side note, from personal observation it seems that bonobo’s fall in between chimpanzees and humans in terms of lip color). Thus, human faces share important features with the ancient primate behind. And there are even more shared properties. Faces and behinds have a reliable structure across individuals and are ubiquitous in the environment, ensuring high levels of exposure. Furthermore, faces and behinds are both symmetrical, an important characteristic both in lower and higher level visual processing[26, 27]. Also, the correct interpretation of the conveyed information by faces and behinds, including identity, fitness and fertility, is crucial for reproductive success[20, 28, 29].

Research has suggested that colour vision in primates was selected for discriminating the spectral modulations on the skin of conspecifics, presumably for the purpose of recognizing emotional states and socio-sexual signals including threat displays and swellings[30]. Whereas humans are predominantly furless, the face and behind are some of the few hairless areas on the body of a chimpanzee. The expansion of areas on the body showing bare skin over human evolution is one potential way of utilizing this medium for signalling. Therefore, the visibility of the skin and its colouration might contribute to the signalling function of faces and behinds.

Whether humans and chimpanzees process behinds the way they process faces is thus far unknown. If so, they should not only be able to recognize individuals by these body parts, but also process them configurally, and show a behind inversion effect. This would show which body parts or information sources different species rely on when recognizing individuals, including potential mates or competitors. In the current study, we therefore investigated whether behinds, like faces, are processed configurally, and whether the putative behind inversion effect is enhanced in chimpanzees, where, in contrast to humans, the female genital region changes in size and colour over the menstrual cycle. Given that human and chimpanzee faces and behinds are furless, we predicted that the presence of colour (red) would foster the recognition process. These questions were addressed in four different experiments, two in humans and two in chimpanzees, where both species matched images of faces, behinds and, as a control condition feet (Body Part). The images showed humans or chimpanzees (Stimulus Species) in upright or inverted position (Body Parts Orientation; see Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

A. Stimulus examples shown in Experiment 1 and 3. What is shown are upright examples on the left panel and inverted examples on the right. The different stimulus categories are shown in each row, with human faces in the first row, chimpanzee behinds in the second, and human feet in the third. The correct response in each row, starting at the top, is left, right, and right (the same response is correct for both the upright and inverted examples). B. Chimpanzee behinds (depicted on the left), contain specific features like human faces (depicted on the right). The black lines highlight those features. C. An example of a chimpanzee participant conducting the task with desaturated images. Experiments 2 and 4 used desaturated images.

https://doi.org/10.1371/journal.pone.0165357.g001

In Experiment 1, human participants were requested to match faces, behinds and feet of both humans and chimpanzees. We tested our hypothesis that they would show a face inversion effect[6–8] and explored the possibility of a behind inversion effect. On the one hand, humans show a body inversion effect[2, 8], and thus, a behind inversion effect seems plausible. But on the other hand, there is not much literature on how humans process behinds, and the behinds that we see in our daily life are usually covered, making it equally likely that humans would not show a behind inversion effect. Experiment 2 tests the same hypotheses but with the additional factor of color. Orientation has been proven to be dominant over color when it comes to face processing[31]. That said, activity in brain areas involved in face processing is boosted when faces are presented in color[32, 33]. We predict that humans will still show a face inversion effect when faces are presented in greyscale, but that the effect is somewhat blunted as compared to Experiment 1 where faces are shown in color. In Experiment 3 we test our hypothesis based on previous observations that chimpanzees will show a face inversion effect[14–17]. Given the high socio-sexual relevance of behinds, we predict that they will show a behind inversion effect too. In Experiment 4 we investigate whether this putative behind inversion effect disappears when turning the pictures into greyscale, which presumably to a large extent will remove the sexual signal[30, 34].

Method

Results

Experiment 1 Human participants, coloured stimuli

In human participants, an interaction between Body Part (face, behind, foot) * Body Part Orientation (upright, inverted) * Stimulus Species (human, chimpanzee), F(2, 6.834) = 4.319, p = .013, demonstrated a specific inversion effect for human faces, t(6.834) = 2.927, p = .003, Fig 2A, but not for human and chimpanzee behinds and feet or chimpanzee faces (ps > .173). See S2 Table.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Reaction times for human and chimpanzee participants to conspecific body parts.

Reaction time is presented as a difference score (Diff), i.e., Reaction times for Inverted minus Upright stimuli. A. Experiment 1, human participants, stimuli in color. B. Experiment 2, human participants, desaturated stimuli. C. Experiment 3, chimpanzee participants, stimuli in color. D. Experiment 4, chimpanzee participants, desaturated stimuli. x p < .1; * p < .05; ** p < .01. Error bars represent the standard error of the mean.

https://doi.org/10.1371/journal.pone.0165357.g002

Discussion

The current study shows chimpanzee’s expertise in recognizing behinds and suggests they process the bright pink sex swellings of female chimpanzees configurally and in a similar way as humans process faces. The female chimpanzee’s behind has a very high socio-sexual signaling function and the changes in size and color over the menstrual cycle reflect fertility. For that reason, it is important for conspecifics to be able to quickly detect this signal in the environment, but at the same time, it is vital to know who the behind belongs to[19]. For male chimpanzees this is relevant to prevent inbreeding. In turn, for female chimpanzees it is relevant to be aware of competing females to protect their own mating success.

The current study replicates previous research on the face inversion effect in humans, demonstrating that they process faces configurally[2]. In line with our hypothesis, the face inversion effect was dampened when faces were turned into greyscale, but still strongly significant, which is in line with previous research in humans showing that orientation is more important than color when it comes to processing human faces[31]. Also without color, the human face contains many high contrasting features such as eye whites, a prominent nose and lips and eyebrows. Although facial color can provide important social information, such as about emotions and health, there are also minor alterations over the menstrual cycle [40]. However, these small changes are beyond any comparison with the rich coloration of the chimpanzee behind where the alterations are much more obvious. In chimpanzees, the relevance of color for processing behinds is reflected in the absence of the behind inversion effect when pictures of behinds were presented in greyscale. In real life, the size and color of the swelling change in synchrony over the menstrual cycle. Thus, a full swelling around estrus is always redder than the female behind half a cycle later. It is therefore possible that due to the un-naturalistic mismatch between color (grey) and size (full swelling), these behinds were processed as objects, i.e., identified by the parts rather than as a whole.

Like humans, great apes are optimally equipped to process color and the spectral sensitivity of the cones in their retinas is ideal for discriminating both density of hemoglobin and oxygen saturation of the blood[30]. Also, the brain areas specialized in processing faces and bodies possess unique neural wiring to effectively process color[32, 33]. Once developed over the course of evolution, color vision (and especially trichromatic color perception) proceeded to impose a selective pressure on certain external traits such as the pink female sexual swelling in chimpanzees and the red lips and cheeks in humans.

A limitation of this study is the low number of individuals in the chimpanzee sample. Although this is common in most primate research and is largely compensated for by the large number of trials per individual, it is possible that effects would have been stronger had we been able to test a larger sample. Moreover, the chimpanzees in our sample were adolescents and adults and we can therefore only speculate about whether this specialization in processing behinds is inborn or related to expertise and emerged sometime during the developmental trajectory. In humans, the specialization for faces occurs already in the first couple of months of life[41]. In fact, already from birth, infants are interested in other people’s faces and eyes and make eye-contact[42]. The making of eye-contact is also facilitated in our species as walking upright freed the hands of parents, allowing them to carry their babies in their arms more often[43]. In contrast, chimpanzees are knuckle-walkers and carry their infants on their belly or back. For them, the swellings become particularly relevant only around puberty. The swellings also appear around that time, i.e., around the age of 10, and at that age, the color of the face changes from pink to a permanent black tint, reducing the contrast with the rest of the body[20]. The swellings stand out enormously in terms of color, size, smoothness and shininess and have a much stronger socio-sexual signaling function in the chimpanzee than the face. Future experiments with larger sample sizes are needed to test for sex differences and could also benefit from including male behinds as a control condition. In addition, it might be valuable to repeat this experiment in the bonobo (Pan Paniscus), as this species is as closely related to us as the chimpanzee but uses sex as a way to prevent and solve conflicts, has an alpha female rather than an alpha male[44] and is known to be highly attentive towards pictures showing genitals and even pay more attention to this category than to images showing threat displays[45].

In sum, applying well-established psychological paradigms to our closest relatives represents a promising approach to providing insight into the evolution of behavior. For primates, being able to recognize each other is necessary for detecting mates. Yin’s(1969) landmark article about the ‘face inversion effect’ turned the face-literature upside-down and hundreds of articles since describe that humans process faces unlike objects. But how faces compare to another body part similar in shape, size, color and attractiveness was thus far unknown. The present study demonstrates that chimpanzees, unlike humans, show a ‘behind inversion effect’ and suggests that identity recognition ‘moved up’ from the bottom to the face in our uprightly walking species. The findings of our study suggest that over human evolution the face took over important properties shared with the primate behind and largely replaced its socio-sexual signaling function, making our species attuned to faces.

Supporting Information

Acknowledgments

We thank Willemijn van Woerkom for her assistance in the data collection, and Drs. T. Matsuzawa, I. Adachi and M. Hayashi, and the members of KUPRI for their generous support throughout the chimpanzee study.