Original Article
Is cooperativeness readable in static facial features? An inter-cultural approach

https://doi.org/10.1016/j.evolhumbehav.2013.08.002Get rights and content

Abstract

There is evidence in the literature that non-verbal physical features are used as cues for a propensity to cooperate. However, further studies of the human ability to visually detect cooperativeness are required. In particular, the existence of static facial cues of altruism remains questionable. Moreover, an investigation of both sex differences and cross-cultural applicability with respect to altruism detection skills is crucial in the context of the evolution of human cooperation. In this study, we used both a public good game and a charitable contribution to assess the cooperativeness of 156 men and 172 women in rural Senegal and took facial photographs of these individuals. The second portion of the study was conducted in France. In total, 194 men and 171 women were asked to distinguish the most and least selfish individual from a series of 80 pairs of Senegalese facial photographs, each pair consisting of the highest and the lowest contributor from a group in the public good game. Using mixed modeling techniques, we controlled for facial masculinity, age and socio-economic status. For male pairs, both male and female French raters were able to identify more often than by chance which individual made the smallest contribution to the public good in each group; however, detection was not successful with female faces. These results suggest that sex-specific traits are involved and that only male facial traits indicating cooperative skills are, at least inter-culturally, readable. The specific facial traits involved are investigated. However, the charitable contribution was not correlated with the contribution to the public good, and further work is necessary to identify which specific altruistic traits are detectable and to assess the generality of these results.

Introduction

A large body of experimental studies demonstrates that a substantial fraction of people behave cooperatively1 with non-kin individuals even if they are given the opportunity to behave selfishly in anonymous interactions. This finding, observed in both industrialized (Andreoni and Miller, 1993, Cadsby et al., 2007, Cadsby and Maynes, 1998, Keser, 1996, Keser and Van Winden, 2000, Willinger and Ziegelmeyer, 1999) and small-scale societies (Fehr and Leibbrandt, 2011, Henrich, 2000, Henrich, 2004, Henrich et al., 2005, Lamba and Mace, 2011), remains a puzzle for evolutionary biologists (Alexander, 2006, Boyd and Richerson, 2009). Indeed, natural selection should, a priori, select selfish behaviors rather than cooperative ones because these selfish behaviors should be more competitive. Altruistic behavior with non-kin could be positively selected if cooperators receive direct benefits enhancing their fitness (West et al., 2007). Direct benefits can be gained by direct or indirect reciprocity when a cooperative behavior elicits a cooperative response by other individuals who observed the interaction, either directly or indirectly (Alexander, 1987, Fehr, 2004, Nowak and Sigmund, 1998, Trivers, 1971; for a review see: Nowak, 2006, West et al., 2007). If there is a choice to reciprocate toward more cooperative individuals because of the benefits they confer, then competition to be more cooperative than others can result (competitive altruism hypothesis: Noe and Hammerstein, 1995, Roberts, 1998). In addition, it has been suggested that more cooperative individuals are preferentially chosen as sexual partners, and thus cooperation can be sexually selected (Miller, 2007, Tognetti et al., 2012, Van Vugt and Iredale, 2013). Cooperation can therefore be used as a cue that other individuals can use to make social decisions, either to select partners to cooperate with or to select sexual partners. In addition, if cooperative behavior is socially and/or sexually selected, cues for the propensity to cooperate can evolve into signals. Cues are non-selected ways of transmitting and obtaining information, whereas a signal is defined as “any act or structure which alters the behavior of other organisms, which evolved because of that effect, and which is effective because the receiver's response has also evolved” (Maynard-Smith & Harper, 2003). One of the main mechanisms hypothesized to ensure the honesty of a signal is condition-dependence; signals have to be costly so that only high-quality individuals can support the cost of exhibiting them (Grafen, 1990, Zahavi, 1977). Altruistic acts represent costly and risky behavior (Clutton-Brock et al., 1998, Heinsohn and Legge, 1999). They could therefore signal the propensity to cooperate and also the individual's quality (theory of costly signaling: Gintis et al., 2001, Roberts, 1998, Zahavi, 1995); such a signal can be used, for example, in mate choice.

All of the current theories that attempt to explain the evolution of altruism among non-kin, e.g., indirect reciprocity, costly signaling, competitive altruism and sexual selection, are therefore based on the assumption that altruists can be distinguished from egoists by potential partners, thereby allowing the assortment of individuals by cooperative altruistic behavior (Wilson & Dugatkin, 1997). Preferences for a similarly cooperative partner can lead to the evolution of cooperation altruism if the advantages of selfish individuals are out-competed by the benefits of mutual cooperation between altruists. Thus, there is likely to be selection for an ability to use visual cues to quickly detect whether an individual is altruistic.

Several studies have suggested that humans are able to estimate another individual's propensity to cooperate, thereby allowing assortment based on cooperativeness. For instance, a study performed in secondary school classes revealed that students were able to predict the decisions of their classmates in a dictator game more accurately than chance alone would have predicted and that altruists were friends with more altruistic students than were egoists (Pradel, Euler, & Fetchenhauer, 2009). At the university level, altruist students tend to choose other altruist students to play at an economic game (Sheldon, Sheldon, & Osbaldiston, 2000).

Nevertheless, altruistic acts may not be the only traits conveying information about the individual's cooperativeness. Every behavioral or physical trait correlated with the individual's cooperativeness could also be used by potential partners as cues to distinguish altruists from egoists. In humans, several experimental studies demonstrate that facial traits could be involved in this detection. For example, several studies used facial photographs obtained while individuals were playing an economic game (Verplaetse, Vanneste, & Braeckman, 2007) or silent video clips of individuals' faces while they were talking about themselves (Brown et al., 2003, Fetchenhauer et al., 2010, Oda, Naganawa, Yamauchi, Yamagata and Matsumoto-Oda, 2009, Oda, Yamagata, Yabiku and Matsumoto-Oda, 2009) to demonstrate that non-verbal facial cues are implicated in the detection of altruism; the participants in these studies were able to either differentiate cooperators from non-cooperators or to predict the cooperativeness of a target individual more accurately than chance alone would dictate.

Facial emotional expressivity is likely to be among the non-verbal traits that serve as cues of cooperativeness. Previously published studies have provided evidence that cooperators exhibit greater numbers of emotional expressions (positive or negative) than do non-cooperators (Mehu, Grammer and Dunbar, 2007, Mehu, Little and Dunbar, 2007, Schug et al., 2010). Moreover, cooperators exhibit more expressions that are under involuntary control, such as the felt smile, which is characterized by the activity of the orbicularis oculi muscle (Brown et al., 2003, Oda, Yamagata, Yabiku and Matsumoto-Oda, 2009).

However, little empirical research has investigated the role of traits other than expressions and movements (i.e., static traits) in the detection of cooperativeness, although there is evidence that individuals' decisions regarding whom to trust are biased by static traits, such as facial attractiveness (Wilson & Eckel, 2006) and similarity to kin (DeBruine, 2002). Moreover, an experimental study based on economic games demonstrated that the facial width-to-height ratio of males, a sexually dimorphic and testosterone-linked facial trait, could be used as a cue of trustworthiness; men with greater facial widths were more likely to exploit the trust of others and that other players, particularly females, were less likely to trust males with wide faces (Stirrat & Perrett, 2010). These results emphasize the existence of static traits that indicate altruism, at least in males; the previously established findings also suggest that the ability to detect cooperativeness may be dependent on the sex of the rater and the sex of the target. Therefore, the first aim of this study was to evaluate the existence of static facial cues of cooperativeness and to examine whether their detection was sex-specific.

The results of this previous study also provide an indirect indication that men with low testosterone levels (i.e., less masculinized men) behave more cooperatively and are considered to be more cooperative; this conjecture is supported by the finding that based on a self-report personality scale, cooperativeness is negatively correlated with testosterone level (Harris, Rushton, Hampson, & Jackson, 1996). In addition, men with high testosterone levels are more likely to reject low-ultimatum game offers (Burnham, 2007). Male facial width, which is linked to testosterone level (Verdonck, Gaethofs, Carels, & de Zegher, 1999), could be one of several traits that are involved in the indication of cooperativeness. Every testosterone-linked facial trait (e.g., larger jawbones, more prominent cheekbones, and thinner cheeks) grouped under the concept of facial masculinity is a likely candidate with respect to the indication of cooperativeness. The second aim of this study was thus to examine whether facial masculinity is linked with an individual's propensity to cooperate and whether it is used as a cue of cooperativeness, at least in men.

For this purpose, we initially assessed cooperativeness in a sample of men and women in rural Senegal and obtained facial photographs of the sampled individuals. We used two measures. The first measure is a public good game, which is more similar to natural situations (such as food sharing, collective hunting and collective building) than dyadic games (used in Stirrat and Perrett, 2010, Verplaetse et al., 2007) or self-reports (used in Fetchenhauer et al., 2010, Oda, Naganawa, Yamauchi, Yamagata and Matsumoto-Oda, 2009, Oda, Yamagata, Yabiku and Matsumoto-Oda, 2009). The second measure involves a charitable contribution; at the end of the public good game, participants could donate part of their gain from the game to the village school. The second part of the study was conducted in France; for each combination of raters' and players' sex, raters were asked to distinguish the most and the least selfish from a series of paired photographs of Senegalese players, while controlling for players' age, socio-economic status and facial masculinity.

Section snippets

A measurement of cooperativeness

The experimental measurements of cooperativeness were obtained in five rural villages that are located in the Sine Saloum area of Senegal (which is close to the west coast of Africa). The participants were recruited on a voluntary basis. The protocols that were used to recruit the study's participants and collect data were approved by the French National Commission on Informatics and Liberties (CNIL declaration #1321739). Written informed consent was obtained from all of the adult subjects and

Cooperativeness in the public good game and the donation

Globally, Senegalese participants allocated slightly more than half of their initial endowment to the public good both during the initial period (mean ± s.d.: 104 ± 58 grams of rice) and as an average over the course of all five periods of the game (mean ± s.d.: 109 ± 52 grams). From their final gain, they donated 111 ± 122 grams of rice, with 1/3 of participants not making any donation (108 out of 318 participants, after removing the 10 participants whose data were missing). Wilcoxon non-parametric

Discussion

The main aim of this study was to examine whether individuals are able to discriminate cooperators from non-cooperators on the basis of static facial cues. Facial traits, including static characteristics, are known to represent a component of attractiveness (through averageness, symmetry and secondary sexual traits); indicate phenotypic quality, genetic quality and developmental health (Little et al., 2007, Little and Hancock, 2002, Scheib et al., 1999, Thornhill and Gangestad, 1993, Thornhill

Acknowledgments

We are grateful to all of the participants in our study. We thank D. Sarr and E. Corr for their work as local guides, Mareme Niang for helpful advice and insights regarding the Senegalese culture, and P.Y. Biba-Nkouka for his help with the experiment that was conducted in France. This study was supported by the CNRS of France (www.cnrs.fr) and the Region Languedoc-Roussillon ‘Chercheur(se)s d'Avenir’ (n°: DGA3/DESR/2012/Q159). The funders had no role in the design, data collection, statistical

References (75)

  • C. Keser

    Voluntary contributions to a public good when partial contribution is a dominant strategy

    Economics Letters

    (1996)
  • J.T. Manning et al.

    Asymmetry and the menstrual cycle in women

    Ethology and Sociobiology

    (1996)
  • M. Mehu et al.

    Smiles when sharing

    Evolution and Human Behavior

    (2007)
  • R. Noe et al.

    Biological markets

    Trends in Ecology & Evolution

    (1995)
  • J. Pradel et al.

    Spotting altuistic dictator game players and mingling with them: the elective assortation of classmates

    Evolution and Human Behavior

    (2009)
  • J. Schug et al.

    Emotional expressivity as a signal of cooperation

    Evolution and Human Behavior

    (2010)
  • R. Thornhill et al.

    Facial attractiveness

    Trends in Cognitive Sciences

    (1999)
  • J. Verplaetse et al.

    You can judge a book by its cover: The sequel. A kernel of truth in predictive cheating detection

    Evolution and Human Behavior

    (2007)
  • S.A. West et al.

    Sixteen common misconceptions about the evolution of cooperation in humans

    Evolution and Human Behavior

    (2011)
  • M. Willinger et al.

    Framing and cooperation in public good games: an experiment with an interior solution

    Economics Letters

    (1999)
  • A. Zahavi

    The cost of honesty

    Journal of Theoretical Biology

    (1977)
  • R.D. Alexander

    The biology of moral systems

    (1987)
  • R.D. Alexander

    The challenge of human social behavior

    Evolutionary Psychology

    (2006)
  • D.R. Anderson et al.

    Suggestions for presenting the results of data analyses

    Journal of Wildlife Manage

    (2001)
  • J. Andreoni et al.

    Rational cooperation in the finitely repeated Prisoner's Dilemma: experimental evidence

    The Economic Journal

    (1993)
  • P. Barclay et al.

    Partner choice creates competitive altruism in humans

    Proceedings of the Royal Society B

    (2007)
  • J. Bovet et al.

    Men's preference for women's facial features: testing homogamy and the paternity uncertainty hypothesis

    PLoS One

    (2012)
  • R. Boyd et al.

    Culture and the evolution of human cooperation

    Philosophical Transactions of the Royal Society B

    (2009)
  • W.M. Brown et al.

    Are there nonverbal cues to commitment? An exploratory study using the zero-acquaintance video presentation paradigm

    Evolutionary Psychology

    (2003)
  • T.C. Burnham

    High-testosterone men reject low ultimatum game offers

    Proceedings of the Royal Society B

    (2007)
  • K.P. Burnham et al.

    Evaluation of some random effects methodology applicable to bird ringing data

    Journal of Applied Statistics

    (2002)
  • D.M. Buss

    Marital assortment for personality dispositions: Assessment with three different data sources

    Behavior Genetics

    (1984)
  • T.H. Clutton-Brock et al.

    Costs of cooperative behavior in suricates (Suricata suricatta)

    Proceedings of the Royal Society B

    (1998)
  • M.F. Dal Martello et al.

    Where are kin recognition signals in the human face?

    Journal of Vision

    (2006)
  • L.M. DeBruine

    Facial resemblance enhances trust

    Proceedings of the Royal Society of London B

    (2002)
  • E. Fehr

    Don't lose your reputation

    Nature

    (2004)
  • R. Guttman et al.

    Spouse similarities in personality items: Changes over years of marriage and implications for mate selection

    Behavior Genetics

    (1987)
  • Cited by (45)

    • You cannot judge a book by its cover: Evidence from a laboratory experiment on recognizing generosity from facial information

      2022, Journal of Behavioral and Experimental Economics
      Citation Excerpt :

      Our samples of targets, namely dictators in a dictator game, and the two groups of independent evaluators, some of whom evaluated the level of generosity while others evaluated the level of attractiveness of these dictator game players were all undergraduate students from China. While there are a few pioneering studies the employed non-western samples (e.g., Oda et al., 2009; Tognetti et al., 2013; Tognetti et al., 2018), to our knowledge our study represents the first study examining recognition of generosity from facial data using an adult sample from China. Our results fail to reject the null hypothesis that our Chinese adult sample is unable to detect altruism better than random chance.

    • How head posture affects perceived cooperativeness: A cross-cultural perspective

      2022, Acta Psychologica
      Citation Excerpt :

      Yet, given that not all humans are benevolent all of the time, there is a great need for us to evaluate the behaviour of others in terms of fairness in order to differentiate altruists from egoists. Equally there is a need for humans to be tuned to (or to learn to recognise) any physical clue (including facial cues) that predicts behaviour and can therefore help identify who is a cooperator and who is not (Tognetti et al., 2013). Emotional expressivity and facial structure seem to be the important traits contributing to the detection of cooperativeness.

    • Can people detect the trustworthiness of strangers based on their facial appearance?

      2022, Evolution and Human Behavior
      Citation Excerpt :

      Bonnefon, Hopfensitz, and De Neys (2013) presented facial photographs of trustees who had either reciprocated or betrayed trust, showing that participants were more likely to transfer money to counterparts that were actually trustworthy. Other studies yielded similar results, leading various authors to conclude that people are able to detect the trustworthiness of counterparts at levels slightly above chance (ca. 55%; Bonnefon, Hopfensitz, & De Neys, 2017; De Neys, Hopfensitz, & Bonnefon, 2013, 2015, 2017; Tognetti, Berticat, Raymond, & Faurie, 2013; Verplaetse, Vanneste, & Braeckman, 2007). Yet, evidence for accurate trustworthiness detection is mixed.

    View all citing articles on Scopus
    View full text