Reliance on head versus eyes in the gaze following of great apes and human infants: the cooperative eye hypothesis

https://doi.org/10.1016/j.jhevol.2006.10.001Get rights and content

Abstract

As compared with other primates, humans have especially visible eyes (e.g., white sclera). One hypothesis is that this feature of human eyes evolved to make it easier for conspecifics to follow an individual's gaze direction in close-range joint attentional and communicative interactions, which would seem to imply especially cooperative (mututalistic) conspecifics. In the current study, we tested one aspect of this cooperative eye hypothesis by comparing the gaze following behavior of great apes to that of human infants. A human experimenter “looked” to the ceiling either with his eyes only, head only (eyes closed), both head and eyes, or neither. Great apes followed gaze to the ceiling based mainly on the human's head direction (although eye direction played some role as well). In contrast, human infants relied almost exclusively on eye direction in these same situations. These results demonstrate that humans are especially reliant on eyes in gaze following situations, and thus, suggest that eyes evolved a new social function in human evolution, most likely to support cooperative (mututalistic) social interactions.

Introduction

One of the central puzzles of human evolution is when and how humans became so cooperative. Humans engage in frequent, large-scale, complex, even institutionalized cooperation with non-kin to a degree unprecedented among the primates, if not all animal species (Richerson and Boyd, 2005).

Humans also cooperate in some unique ways in close-range social interactions involving two or a few individuals acting together toward a concrete goal, possibly communicating as they do so. Similarly, chimpanzees hunt monkeys together in small groups (Boesch and Boesch, 1989, Mitani and Watts, 1999), and indeed, recent experimental research has established that in cooperation of this kind they (i) know when they do and do not need a partner, and (ii) can distinguish helpful from unhelpful partners (Melis et al., 2006). However, there are still differences with human cooperation in terms of both cognitive and motivational bases. Warneken et al. (2006) found that young human children were much more motivated than young chimpanzees to engage in cooperation for its own sake (not just for an instrumental goal), and they communicated during the cooperation in much more complex ways than did their ape cousins.

Perhaps of special importance, humans seem especially inclined, as compared with other primates, to engage with one another in collaborative activities around objects—so-called joint attentional interactions (Bard and Vauclair, 1984, Tomasello and Carpenter, 2005). In these triadic interactions, each participant typically monitors what the other is attending to visually—including the other's monitoring of their own activities and visual attention—so as to coordinate actions more effectively (Tomasello et al., 2005). As one important example, human mothers and infants regularly engage in joint attentional interactions around objects, and these form the referential context within which skills with language develop (Bruner, 1983, Tomasello, 2003). Quantitatively, when human infants interact with caregivers, they engage in joint attentional interactions that average twice as long in duration as those of human-raised great apes with their caregivers, and importantly, during these interactions the duration of infants' looks to the face/eyes of the caregiver average twice as long as those of the apes (Carpenter et al., 1995). In joint attentional interactions among adult humans, as well as adult-child pairs, participants make frequent use of the visually-based pointing gesture. This gesture requires the participants to monitor the gaze direction of one another fairly closely, and pointing is not used by other primate species in their natural communication (Call and Tomasello, in press). In general, it would seem to be an advantage in initiating and maintaining collaborative/joint attentional/communicative interactions of the human kind that one's eyes be easily visible to others in order to facilitate a shared activity—assuming that the other is a cooperative partner not overly inclined toward exploitation.

It turns out, as is well-known, that humans indeed do have especially visible eyes (Kobayashi and Kohshima, 1997). Human eyes are colored in a way that helps advertise both their presence and their gaze direction much more saliently than in other primates. Kobayashi and Kohshima (2001) examined 92 primate species (including humans) and found that 85 had exposed sclera that were uniformly brown or dark brown. Microscopic analysis showed that the brown coloration of the exposed sclera was created by pigmentation deposition in the epithelium cornea, conjunctiva, and sclera. In addition, when eye coloration was compared to facial skin coloration in a subset of 81 species (including humans), 80 species were found to have low contrast in eye and facial skin coloration (i.e., the outline of the eyes and the position of the iris were difficult to distinguish due to the similarity in color of the facial skin, sclera, and iris). The only species with a transparent conjunctiva and white sclera without any pigmentation was humans. In addition, humans were the only species in which the eye outline and the position of the iris were clearly visible, since the exposed sclera was paler than the lightest colored iris or surrounding skin. Finally, the human eye and its visible regions were found to be disproportionately large and horizontally elongated for body size (i.e., the visible regions of human eyes were bigger than that of the much larger gorilla). In a quantitative comparison, Kaplan and Rogers (2002) found that the amount of visible sclera was three times greater in humans than in orangutans (when looking straight ahead—twice as large when looking to the side).

Thus, one hypothesis is that human-type eyes evolved in the context of pressures for enhanced cooperative-communicative abilities of the kind needed in mutualistic social interactions involving joint attention and visually based communication such as pointing. At present, we know of no data directly relevant to this hypothesis. But at the same time, we know of no systematic data supporting any other hypotheses explaining the uniqueness of the human eye in terms of function (although see Kobayashi and Kohshima, 2001, for some interesting related data). One approach is to compare the way that humans and their closest primate relatives follow the gaze direction of others, specifically, in the extent to which they use the eyes versus the head in gaze following. There is relevant information in previous studies of gaze following in both nonhuman primates and human infants, but in none of these has the distinction between following the direction of the head alone versus following the direction of the eyes alone been systematically tested. It is thus unclear if humans make special use of the eyes as compared with other primate species including great apes.

In experimental paradigms with great apes in which the subject must determine whether a human can see her or not—for example, before gesturing—the results are mixed. In some studies, chimpanzees and other apes care whether the human is bodily oriented toward them (e.g., Povinelli and Eddy, 1996a, Liebal et al., 2004), but they do not seem to care whether the human's eyes are open or closed (Povinelli and Eddy, 1996a, Kaminski et al., 2004). Other studies, however, have found that some apes were sensitive to whether the human's eyes were open or closed in such situations (Call and Tomasello, 1994, Gómez, 1996), and other studies have indicated that monkeys can take into account the state of the human's eyes in competitive situations (Vick and Anderson, 2003, Flombaum and Santos, 2005) and in some gaze following situations (Deaner and Platt, 2003). In experimental paradigms in which subjects have to use human-given cues to locate a hidden reward—the so-called object choice paradigm—the evidence is again mixed. Numerous studies have shown that apes and monkeys find it difficult to use the head or eye orientation of a human experimenter to locate a hidden reward (see Call and Tomasello, 2003, for a review). Other studies, however, have found positive results both with face and eye direction in apes (Povinelli and Eddy, 1996b, Itakura and Tanaka, 1998).1 What is currently unknown is the relative strength of the head versus the eyes in eliciting gaze following behavior in any nonhuman primate species.

Human infants follow gaze direction to near targets during the first year of life (D'Entremont et al., 1997), and to more distant targets at around the first birthday (Carpenter et al., 1998). Infants seem to follow glancing with the eyes (no head movement) from around 18 months of age (Corkum and Moore, 1995). In the only study with a direct comparison of head and eyes, Brooks and Meltzoff (2002) found that 14-month-old infants follow an adult's gaze direction more often when her eyes are open than when they are closed—and also more often when her eyes are unobstructed versus when they are covered with a blindfold. Even in human infants, however, there has been no direct comparison of the relative effectiveness of head versus eyes in eliciting gaze following behavior.

In the current study, we systematically tested the role of the head versus the eyes in the gaze following of great apes and human infants. Based on humans' greater propensity for object-centered cooperative/communicative interactions (joint attentional interactions) and their especially visible eyes, our hypothesis was that human infants would be more influenced by the eyes than the head, whereas great apes would be more influenced by the head than the eyes. We tested this “cooperative eye hypothesis” in a simple 2 × 2 design with the two factors Head (turned or not) and Eyes (open or closed). This design enabled a direct comparison between the role of head and eyes in the gaze following of humans versus great apes. As an additional probe, we also presented these same subjects with a human with his back turned, who then looked up. This enabled us to differentiate the role of the head versus the face (front of the head) more specifically. Because of various practical considerations (e.g., ape housing), the methods used with the great apes and human infants were slightly different, and so the two studies are reported separately.

Section snippets

Methods

Participants. Participants were 19 great apes housed at the Wolfgang Köhler Primate Research Center: 11 chimpanzees (Pan troglodytes) ranging in age from 4 to 27 years, 4 gorillas (Gorilla gorilla) ranging in age from 5 to 25 years, and 4 bonobos (Pan paniscus) ranging in age from 6 to 20 years. Three additional chimpanzees and five orangutans were tested but did not pay attention to the gaze cues sufficiently for their skills to be reliably assessed; specifically, subjects who failed to attend

Methods

Participants. Participants were 20 human infants at 12 months-of-age (within two weeks on either side) and 20 human infants at 18 months-of-age (within two weeks on either side). All children were from a middle-sized German city. There were approximately equal numbers of boys and girls at each age. Infants were recruited from a child database consisting of families who had previously volunteered for developmental research. Five additional infants (three 18-month-olds and two 12-month-olds) were

General discussion

The results of the current studies provide strong support for the cooperative eye hypothesis. Both great apes and human infants followed the gaze direction of a human reliably. In the condition in which E's head and eyes both were oriented upward, both apes and infants looked up seven to nine times more often than when E's head and eyes both were oriented downward (see Table 1, Table 2). However, in the conditions in which head and eye orientation were incongruent, the species showed very

References (55)

  • M. Tomasello et al.

    The ontogeny of gaze following in chimpanzees and rhesus macaques

    Anim. Behav.

    (2001)
  • R. Adolphs

    Recognizing emotion from facial expressions: psychological and neurological mechanisms

    Behav. Cogn. Neurosci. Rev.

    (2002)
  • J.R. Anderson et al.

    Macaques but not lemurs co-orient visually with humans

    Folia Primatol.

    (1999)
  • R. Bakeman et al.

    Observing Interaction: An Introduction to Sequential Analysis

    (1986)
  • S. Baron-Cohen et al.

    Are children with autism blind to the mentalistic significance of the eyes?

    Br. J. Dev. Psychol.

    (1995)
  • S. Baron-Cohen et al.

    The ‘reading the mind in the eyes’ test revised version: a study with normal adults, and adults with Asperger syndrome or high-functioning autism

    J. Child Psychol. Psychiat.

    (2001)
  • S. Baron-Cohen et al.

    Is there a “language of the eyes”? Evidence from normal adults, and adults with autism or Asperger syndrome

    Vis. Cogn.

    (1997)
  • C. Boesch et al.

    Hunting behavior of wild chimpanzees in the Tai Forest National Park

    Am. J. Phys. Anthropol.

    (1989)
  • R. Brooks et al.

    The importance of eyes: how infants interpret adult looking behavior

    Dev. Psychol.

    (2002)
  • J. Bruner

    Child's Talk

    (1983)
  • T. Bugnyar et al.

    Ravens, Corvus corax, follow gaze direction of humans around obstacles

    Proc. R. Soc. Lond., B, Biol. Sci.

    (2004)
  • Burnham, T., Hare, B. Engineering human cooperation: does involuntary neural activation increase public goods...
  • J. Call et al.

    The production and comprehension of referential pointing by orangutans

    J. Comp. Psychol.

    (1994)
  • J. Call et al.

    Social cognition

  • Call, J., Tomasello, M. The Gestural Communication of Monkeys and Apes. Lawrence Erlbaum Publishers, in...
  • M. Carpenter et al.

    Social cognition, joint attention, and communicative competence from 9 to 15 months of age

    Monogr. Soc. Res. Child Dev.

    (1998)
  • M. Carpenter et al.

    Joint attention and imitative learning in children, chimpanzees, and enculturated chimpanzees

    Soc. Dev.

    (1995)
  • Cited by (355)

    View all citing articles on Scopus
    View full text