Original Article
Exploring the evolutionary foundations of empathy: consolation in monkeys

https://doi.org/10.1016/j.evolhumbehav.2014.04.002Get rights and content

Abstract

Unsolicited third-party affiliation occurs when victims of aggression receive a spontaneous affinitive contact from a bystander. Consolation is a specific type of unsolicited third-party affiliation showing two key components: 1) it alleviates distress in the victims and 2) is preferentially directed towards friends. Consolation was thought to be present only in humans, chimpanzees, and bonobos linked to their higher cognitive and empathic abilities. Previous investigations in monkeys found unsolicited third-party affiliation in only two species with no evidence of consolation. In the research presented here we set out to test a number of hypotheses concerning third-party affiliation in Macaca fuscata and M. tonkeana, two species that differ remarkably for social style. M. fuscata is despotic, while M. tonkeana is one of the most tolerant species of macaques. We found no evidence of unsolicited third-party affiliation in M. fuscata, but it was present in M. tonkeana. In this species we found that unsolicited third-party affiliation reduced anxiety (measured by scratching) in the victims and was directed towards friends and especially towards females who experienced higher levels of anxiety compared to males. Third-party affiliation also occurred more frequently in the absence of reconciliation. All the key features used to recognize consolation in humans and great apes are present in M. tonkeana making it difficult not to conclude that consolation exists in this species. Since consolation is most often considered to be driven by empathy, our results suggest that Tonkean macaques are capable of empathetically reacting to the victim’s state of anxiety. Our results support the Social Constraints Hypothesis showing that the degree of tolerance is a key factor in the expression of consolation. Investigating behavioral patterns driven by even the most basic forms of empathy requires the choice of an appropriate species and Macaca tonkeana is a good model to investigate the full phylogenetic range, evolutionary depth, and origin of empathy in primates.

Introduction

For social species, including humans, aggression can have dramatic consequences not only for the individuals involved but also for the whole group (Aureli and de Waal, 2000, Esteban et al., 2012). Conflict can generate further conflicts and/or can affect affiliative interactions in the group (Barash and Lipton, 2011, de Waal, 2000, Ellemers, 2012, Palagi and Cordoni, 2009, Watts et al., 2000).

Reconciliation, defined as an affinitive interaction between former opponents in the first minutes after a fight, is an effective conflict resolution mechanism (de Waal & Roosmalen, 1979). Reconciliation reduces the probability of further attacks, limits anxiety in the victim, and restores the benefits associated with a good relationship (Aureli, Cords, & Van Schaik, 2002).

After a conflict, victims can also receive a friendly, spontaneous contact from a bystander not involved in the aggression, a phenomenon labeled “unsolicited bystander affiliation” or “unsolicited third-party affiliation” (Fraser et al., 2009, Romero et al., 2010). This first spontaneous post-conflict affiliative contact from a third-party to the victim was called consolation and was reported for great apes and humans (chimpanzees: Fraser and Aureli, 2008, Koski and Sterck, 2007, Kutsukake and Castles, 2004, Palagi et al., 2006, Romero and de Waal, 2010, Wittig and Boesch, 2003, Wittig and Boesch, 2010; gorillas, Cordoni & Palagi, 2007; bonobos, Clay and de Waal, 2013a, Clay and de Waal, 2013b, Palagi and Norscia, 2013, Palagi et al., 2004; humans, Eisenberg, 1992, Fujisawa et al., 2006). Consolation reduces the victim’s anxiety, is provided by friends and is apparently based on empathy. Some researchers skeptical of the empathic basis of consolation still prefer to use the more neutral term, “unsolicited bystander affiliation” (Call, Aureli, & de Waal, 2002).

Early studies using the same protocol applied to great apes failed to find evidence for consolation in four macaque species, Macaca fascicularis, M. fuscata, M. sylvanus, and M. nemestrina, (de Waal & Aureli, 1996). These data were reinforced when other investigators failed to reveal even unsolicited bystander affiliation in catarrhine monkeys (Schino et al., 2004, Watts et al., 2000). Up to now unsolicited bystander affiliation was found in only two monkey species, but it lacked important attributes of consolation (Call et al., 2002, Schino and Marini, 2012). Call et al. (2002) suggested that consolation was absent in stump-tailed macaques because post-conflict third-party affiliation only limited further aggression by the victim. Schino and Marini (2012) concluded that that there was no consolation in mandrills, because bystander affiliation did not reduce the victims' anxiety and was not received primarily from kin/friends.

Third-party affiliation towards victims was also observed in rooks (Seed, Clayton, & Emery, 2007), ravens (Fraser & Bugnyar, 2010), dogs (Cools, Van Hout, & Nelissen, 2008), and wolves (Palagi & Cordoni, 2009). Although these authors hypothesized that consolation might be present, it was never demonstrated. Up to now, consolation defined as third party affiliation biased towards friends and reducing anxiety was only demonstrated in chimpanzees, bonobos, and humans (Clay and de Waal, 2013a, Clay and de Waal, 2013b, Fraser et al., 2008, Palagi and Norscia, 2013, Romero et al., 2010, Zahn-Waxler and Radke-Yarrow, 1990).

Consolation is considered an empathy-based phenomenon, but it does not necessarily require the most complex form of empathy (i.e. cognitive perspective taking, sensu de Waal, 2008). Empathy is a multilayered phenomenon, whose shared building blocks may provide a framework for more complex emotional and cognitive processes (Hecht et al., 2012, Panksepp and Panksepp, 2013). Different levels of empathy are recognized according to neurobiological, psychological, and behavioral complexity (Hecht et al., 2012, Iacoboni, 2009). The ability to react to the feelings of others appears early in life (Clay and de Waal, 2013b, Hatfield et al., 1994) and its neurobiological substrates were documented in human and non-human primates (Decety and Jackson, 2006, Rizzolatti et al., 2007). A subconscious, reflexive appreciation of others' emotional states does not necessarily imply the attribution of mental states (intentions). However, sensitivity to others' emotions is a building block of even the most complex forms of Theory of Mind (Seyfarth & Cheney, 2013). It may have evolved because natural selection has favored individuals that are motivated to "empathize" with others and take care of their social interactions (De Marco et al., 2011, Seyfarth and Cheney, 2013). Both neurobiological and ethological findings (Mancini et al., 2013, Palagi et al., 2009, Paukner et al., 2009) amply confirm that many non-human animals are able to emotionally connect to each other through mimicry and behavioral contagion, which represent the most basal building blocks of empathy (see the Russian Doll Model by Preston & de Waal, 2002). Only cross-species investigations, such as the research we present here, can shed light on how higher cognitive forms of empathy have evolved in humans.

We investigated this phenomenon by focusing on unsolicited third-party affiliation in Macaca tonkeana and Macaca fuscata. The 20 macaque species are organized in multi-male, multi-female social groups that vary on a gradient ranging from more intolerant (Grade 1) to more tolerant (Grade 4) social systems (Matsumura, 1999, Thierry, 2000). According to the Social Constraints Hypothesis (de Waal & Aureli, 1996) these differences in social styles (de Waal & Luttrell, 1989), already present in infancy (Thierry, 1985a), influence a wide range of behaviors including aggression, affiliation, dominance, and nepotism (Aureli et al., 1997, Balasubramaniam et al., 2012, Petit et al., 1997, Thierry, 1985b, Thierry, 1990). Despotic species like Japanese macaques (Macaca fuscata) have a strong, kin-centric power asymmetry between dominants and subordinates, marked submission behaviors, unidirectional conflicts, and low levels of social tolerance (Aureli et al., 1997, Kutsukake and Castles, 2001). In contrast, more tolerant species, such as Tonkean macaques (Macaca tonkeana), have relationships, which are minimally influenced by rank and kinship. No formal indicators of subordination are present and the proportion of friendly interactions, measured by grooming rates, among non-kin is relatively high (Butovskaya, 2004, Butovskaya and Kozintsev, 1996). It was well known that conflict resolution mechanisms in Tonkean macaques differ from those of Japanese macaques (Schino et al., 2004). Reconciliation, quadratic affiliation, and peaceful interventions are more common in Tonkean macaques (Ciani et al., 2012, De Marco et al., 2010, Demaria and Thierry, 2001, Petit and Thierry, 1994, Thierry, 1985a, Thierry, 1985b). For these reasons we thought that M. tonkeana might be a good model species to study the basis of unsolicited bystander affiliation in a monkey. According to the Social Constraints Hypothesis (de Waal & Aureli, 1996), we expected to find unsolicited bystander affiliation in Macaca tonkeana and to confirm its absence in Macaca fuscata (Prediction 1).

We also wanted to test a number of hypotheses concerning third party affiliation. According to the Substitute for Reconciliation Hypothesis, third-party affiliation restores relationships between former opponents (Fraser and Aureli, 2008, Palagi et al., 2004, Schino and Marini, 2012). If, in Macaca tonkeana, the Substitute for Reconciliation Hypothesis (Fraser and Aureli, 2008, Palagi and Cordoni, 2009, Palagi et al., 2004, Romero and de Waal, 2010, Wittig and Boesch, 2003, Wittig and Boesch, 2010) explains the distribution of unsolicited bystander affiliation towards the victim, it is expected that this post-conflict mechanism would function as a substitute of reconciliation when post-conflict affiliation between the victim and the aggressor fails to occur (i.e. higher levels of unsolicited bystander affiliation in absence of reconciliation; Prediction 2).

The Self-Protection Hypothesis (direct benefits for the third-party) predicts that post-conflict third party affiliation reduces the redirection of aggression (Call et al., 2002, Koski and Sterck, 2009, Schino and Marini, 2012, Wittig and Boesch, 2010). In this view, redirection should be frequent and affiliation should be received primarily from individuals that are frequently the target of redirection. Apparently, Tonkean macaques show low levels of redirection (Thierry, 1985b) suggesting that the Self-Protection Hypothesis would not explain the potential presence of spontaneous bystander affiliation in this species. In order to test this conclusion we measured the frequency of redirected aggression. If third-party affiliation functions to protect the bystander from redirection, we should expect high levels of redirection (Prediction 3). We also would expect that bystanders affiliating with victims are individuals ranking below the victim and/or those individuals receiving the highest levels of redirection (Prediction 4).

The effect of third-party affiliation on reducing renewed aggression has been tested in several studies both to verify the Tension Reduction (Palagi et al., 2006) and Self-Protection Hypotheses (Schino and Marini, 2012, Wittig and Boesch, 2010). A benefit of third-party affiliation might be to lower the probability that the victim is subjected to further aggression (Victim Protection Hypothesis) (Palagi & Norscia, 2013). We tested the Victim Protection Hypothesis by checking if third-party affiliation protected the victim against further conflicts. If third-party affiliation protects the victim (Victim Protection Hypothesis), we expect it to significantly reduce the probability of renewed attacks on him/her (Prediction 5).

The presence of an agonistic event within a group can also increase anxiety and social tension in subjects not directly involved in the conflict (De Marco et al., 2010, de Waal, 2000). We tested the Tension Reduction Hypothesis by verifying whether it reduced the diffusion of aggression. If third-party affiliation reduces tension at the group level (Tension Reduction Hypothesis) and limits the risk of bystanders' involvement in subsequent conflicts, we expect it to significantly reduce the probability of aggression among all group members (Prediction 6), with an indirect benefit for the bystander. We underline that the Victim Protection and Tension Reduction Hypotheses are not mutually exclusive.

Finally, we tested the consolatory function of the third-party affiliation (Consolation Hypothesis, Clay and de Waal, 2013a, Clay and de Waal, 2013b, Palagi and Norscia, 2013, Fraser et al., 2009, Romero et al., 2010). Consoling others requires that bystanders perceive the emotional state of victims and provide the appropriate response to reduce their anxiety (Fraser et al., 2008, Palagi and Norscia, 2013, Romero et al., 2010). The capacity to perceive others' emotions and operate to improve their emotional state is considered a basal form of empathy, defined as sympathetic concern (Preston & de Waal, 2002). Consolation in humans and apes is promoted by the social affiliation of the bystander to the victim in terms of kinship or relationship quality (Fraser et al., 2008, Romero et al., 2010).

If the Consolation hypothesis explains the presence of third-party affiliation in Tonkean macaques, we expect that third-party affiliation reduces the victims' anxiety (measured by self-scratching levels, Aureli and de Waal, 1997, Mastripieri et al., 1992, Palagi and Norscia, 2011, Sclafani et al., 2012, Troisi, 2002) (Prediction 7) and it should be primarily received from friends (Prediction 8).

Section snippets

Subjects and housing

Behavioral data were collected on a group of Macaca tonkeana (Parc Zoologique de Thoiry, France) and on a group of Macaca fuscata (Olomouc Zoo, Czech Republic). The colony of Tonkean macaques had 30 adult males, 31 adult females, and nine immature subjects (1–4 years of age). Births were regulated by contraceptive implants. The animals’ enclosure included both indoor and outdoor facilities (182 m2 and 3900 m2, respectively). The outdoor grass area was equipped with pools, rope structures, trees

Results

To test the hypotheses on third-party affiliation, we focused on adults as victims, collecting 876 PC-MC pairs for Tonkean macaques and 148 PC-MC pairs for Japanese macaques. We considered only subjects with at least five PC-MCs during which reconciliation did not take place (N = 43 for Tonkean and N = 10 for Japanese macaques). In Macaca tonkeana, attracted pairs were significantly more frequent than dispersed pairs (Paired samples t-test: t = 6.266, N = 43, df = 42, p = 0.0001; mean TCT = 21.36% ± 3.6 SE).

Discussion

In this paper, we showed that after an aggressive event Tonkean macaques may spontaneously contact the victim and that the two key criteria used to recognize consolation in great apes and humans were present: 1) the levels of anxiety were significantly reduced after receiving an affiliative contact from a bystander and 2) bystander affiliation was mainly directed towards friends. Our findings strongly suggest that third-party affiliation in M. tonkeana may be considered as “consolation” and

Acknowledgments

We thank C. Scopa and F. Ciani for data collection, the staff of the Parc Zoologique de Thoiry (France) and the Olomouc Zoo (Czech Republic) and W. Otello for discussions. We are grateful to Frans de Waal for his suggestions. The present work was supported by personal fundings and it complies with current laws of Italy, France, and Czech Republic.

References (91)

  • A.M. Seed et al.

    Postconflict third-party affiliation in rooks, Corvus frugilegus

    Current Biology

    (2007)
  • B. Thierry

    Social development in three species of macaque (Macaca mulatta, M. fascicularis, M. tonkeana): A preliminary report on the first ten weeks of life

    Behavioural Processes

    (1985)
  • B. Thierry

    Feedback loop between kinship and dominance: The macaque model

    Journal of Theoretical Biology

    (1990)
  • J. Altman

    Observational study of behaviour sampling methods

    Behaviour

    (1974)
  • F. Aureli et al.

    Differential kinship effect on reconciliation in three species of macaques (Macaca fascicularis, M. fuscata, and M. sylvanus)

    Journal of Comparative Psychology

    (1997)
  • F. Aureli et al.

    Inhibition of social behavior in chimpanzees under high-density conditions

    American Journal of Primatology

    (1997)
  • F. Aureli et al.

    Natural conflict resolution

    (2000)
  • F. Aureli et al.

    Reconciliation, consolation, and redirection in Japanese macaques (Macaca fuscata)

    Behaviour

    (1993)
  • F. Aureli et al.

    Distress prevention by grooming others in crested black macaques

    Biology Letters

    (2010)
  • K.N. Balasubramaniam et al.

    Hierarchical steepness, counter-aggression, and macaque social style scale

    American Journal of Primatology

    (2012)
  • D.P. Barash et al.

    Payback: why we retaliate, redirect aggression, and take revenge

    (2011)
  • M. Butovskaya

    Kinship and different dominant styles in groups of three species of the genus Macaca (M. arctoides, M. mulatta, M. fascicularis)

    Folia Primatologica

    (1993)
  • M. Butovskaya

    Social space and degrees of freedom

  • M.L. Butovskaya et al.

    Gender-related factors affecting primate social behavior: grooming, rank, age, and kinship in heterosexual and all-male groups of stumptail macaques

    American Journal of Physical Anthropology

    (1996)
  • Z. Clay et al.

    Bonobos respond to distress in others: consolation across the age spectrum

    PloS One

    (2013)
  • Z. Clay et al.

    Development of socio-emotional competence in bonobos

    Proceedings of National Academy of Science USA

    (2013)
  • A.K.A. Cools et al.

    Canine reconciliation and third-party-initiated postconflict affiliation: Do peacemaking social mechanisms in dogs rival those of higher primates?

    Ethology

    (2008)
  • G. Cordoni et al.

    Response of captive lowland gorillas (Gorilla gorilla gorilla) to different housing conditions: testing the aggression-density and coping models

    Journal of Comparative Psychology

    (2007)
  • A. De Marco et al.

    Collective arousal when reuniting after temporary separation in Tonkean macaques

    American Journal of Physical Anthropology

    (2011)
  • F.B.M. de Waal

    Primates–a natural heritage of conflict resolution

    Science

    (2000)
  • F.B.M. de Waal

    Putting the altruism back into altruism: the evolution of empathy

    Annual Review of Psychology

    (2008)
  • F.B.M. de Waal et al.

    Consolation, reconciliation and a possible cognitive difference between macaques and chimpanzees

  • F.B.M. de Waal et al.

    Toward a comparative socioecology of the genus Macaca – Different dominance styles in rhesus and stumptail monkeys

    American Journal of Primatology

    (1989)
  • F.B.M. de Waal et al.

    Reconciliation and consolation among chimpanzees

    Behavioral Ecology and Sociobiology

    (1979)
  • F.B.M. de Waal et al.

    Prosocial primates: selfish and unselfish motivations

    Philosophical Transaction of the Royal Society of London B Biological Science

    (2010)
  • F.B.M. de Waal et al.

    Reconciliation and redirected affection in rhesus monkeys

    Behaviour

    (1983)
  • J. Decety et al.

    A social-neuroscience perspective on empathy

    Current Directions in Psychological Science

    (2006)
  • C. Demaria et al.

    A comparative study of reconciliation in rhesus and Tonkean macaques

    Behaviour

    (2001)
  • R.I.M. Dunbar

    Primate social systems

    (1988)
  • R.I.M. Dunbar

    Functional significance of social grooming in primates

    Folia Primatologica

    (1991)
  • N. Eisenberg

    The caring child

    (1992)
  • N. Ellemers

    The group self

    Science

    (2012)
  • J. Esteban et al.

    Ethnicity and conflict: theory and facts

    Science

    (2012)
  • J.C. Flack et al.

    Dominance style, social power, and conflict

  • O.N. Fraser et al.

    Reconciliation, consolation and postconflict behavioral specificity in chimpanzees

    American Journal of Primatology

    (2008)
  • Cited by (49)

    • Emotion recognition in nonhuman primates: How experimental research can contribute to a better understanding of underlying mechanisms

      2021, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      Anecdotes of prosocial behaviour in apes, such as a chimpanzee rescuing an unrelated infant chimpanzee from the water (Goodall, 1986) and a gorilla rescuing a 3-year-old boy who had fallen into her zoo enclosure (de Waal, 1997) served as additional support for this proposal. However, third party post-conflict affiliation has since then also been observed in monkeys (Call et al., 2002; Palagi et al., 2014a) and there may be more similarities on a cognitive level between apes and monkeys than previously thought (Clay et al., 2018). Moreover, the capacity for cognitive empathy in apes has even been questioned (Koski and Sterck, 2010).

    • Challenges in the comparative study of empathy and related phenomena in animals

      2020, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      According to our discussion of the mimicry hypothesis of emotional contagion, it seems commendable for future research to carefully interpret collected empirical data in light of both models, and to compare them accordingly. Contagious yawning has been taken as indicative of empathy, or at the very least as evidence for emotional contagion (e.g. Palagi et al., 2014a, 2014b; Norscia and Palagi, 2011; Clay et al., 2018). In a broad sense, yawn contagion is considered as a form of mimicry (Yoon and Tennie, 2010; see Table 1 for definitions of mimicry and behavioural contagion).

    • Oxytocin administration modulates rats’ helping behavior depending on social context

      2020, Neuroscience Research
      Citation Excerpt :

      A variety of mammals, such as apes (Hare and Kwetuenda, 2010; Palagi et al., 2014), canines (Quervel-Chaumette et al., 2015) and rodents (Bartal et al., 2011; Burkett et al., 2016), exhibit prosocial behavior (e.g., helping, consolation, and food sharing), voluntary behavior through which they provide benefits to other individuals (Eisenberg and Miller, 1987; Decety and Jackson, 2004).

    • A dyad shows mutual changes during social buffering of conditioned fear responses in male rats

      2019, Behavioural Brain Research
      Citation Excerpt :

      These results suggest that a dyad shows mutual changes during social buffering of conditioned fear responses in male rats. Although we did not discuss an alternative interpretation in which the associate may have shown increased social behaviors to console the conditioned subject, as seen in chimpanzees [72] and Tonkean macaques [73], we do not exclude the possibility that rats have sufficiently complex cognitive abilities to exhibit consolation behavior. Rather, we believe that interpreting the findings without presuming higher cognitive function is, paradoxically, the most appropriate way to examine the existence of higher brain function, and to objectively evaluate empathetic ability in rats.

    • Empathy is not in our genes

      2018, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      Whenever an agent encounters a threat or exciting opportunity in the presence of conspecifics, the agent experiences an emotion alongside the opportunity to observe others experiencing the same emotion. Thus, Learned Matching is consistent with reports of emotional contagion, not only in primates (Dezecache et al., 2017; Palagi et al., 2014; Ross et al., 2008), but in birds (Osvath and Sima, 2014; Perez et al., 2015; Schwing et al., 2017; Shah et al., 2015), elephants (King et al., 2010), dogs (Huber et al., 2017; Palagi et al., 2015; Quervel-Chaumette et al., 2016, pigs (Reimert et al., 2013), rodents (Burkett, 2016; Inagaki and Ushida, 2017; Meyza et al., 2017), and ants (Hollis and Nowbahari, 2013). Also consistent with Learned Matching, many examples of emotional contagion in nonhuman animals depend on auditory rather than visual cues, on vocal rather than facial or postural expressions of emotion (Briefer, 2018).

    View all citing articles on Scopus
    View full text