The dawn of social bonds: what is the role of shared experiences in non-human animals?

Abstract

Group-living animals can develop social bonds. Social bonds can be considered a type of social relationship characterized by frequent and consistent affiliative (non-reproductive) interactions. Social bonds with conspecifics bring many advantages, also in terms of direct fitness. A characteristic of social bonds is that they need time to develop. Several studies on humans have emphasized the fact that sharing experiences can affect the strength of social bonds. A similar trend can be spotted in non-human species. For example, a recent experiment showed that if chimpanzees watched a video together with a conspecific, they spent more time in proximity compared to conspecifics with whom they did not actively watch a video. Another experiment on fish showed that individuals who experienced a situation of high predation risk together, showed preference for each other compared to those who did not. As the link between shared experiences and social bonds is not explicitly recognized in non-human animals, the main goal of this work is to propose the exploration of this novel research path. This exploration would contribute to shed light on the evolutionary mechanisms of social bond (or friendship) development and maintenance between individuals in different vertebrate species, from fish to non-human primates.

1. Introduction

Many animals live in groups [1]. Within group-living animals, social bonds (or friendship [2–5]) can exist when individuals exchange frequent and consistent affiliative (nonreproductive) interactions [5]. Social bonds with group members bring multiple benefits: alliance formation [6], cooperative defense of territories [7] or against predators [8,9], and also enhanced direct fitness. Individuals able to develop strong social bonds with conspecifics, gain benefits in terms of health [10]—including stress resistance [11,12]—longevity [13], reproductive output [14] and even offspring survival [15]. Social bonds and their benefits are well described in different animal classes, from fishes to mammals (e.g. fishes: [8]; birds: [16]; mammals: [17], including primates: [2]), including humans [3].

A social relationship can be defined as the unique history of interactions between two individuals [18,19]. As each dyad can exchange an exclusive set of social interactions over time, social relationships among dyads tend to be differentiated (e.g. fishes: [20]; birds: [21]; mammals: [17], including primates: [22]). Although many variables are known to affect (non-sexual) social preference (e.g. familiarity: [23–25]; phenotype similarity: [26,27]; habitat and diet similarity: [28]; social status: [29]), the existence of differential social relationships may also elicit a preference towards one individual over another [30–32]. According to the definition of social relationship itself [18,19], social bonds need time to develop [33,34]. However, it is not only a matter of time per se, but how it is spent together (i.e. going through positive or negative shared experiences), which can affect the strength of social bonds.

The main goal of this work is to propose the exploration of the role that shared experiences play in the development of social bonds in non-human animals. In the literature on the human species, there are complex debates on what ‘shared' means, especially in terms of awareness [35,36]. This opinion piece focuses on non-human species (from fishes to non-human primates). Although behavioural synchrony might possibly be thought of as a prototypical expression of awareness, individual awareness may be too difficult to measure and demonstrate in other organisms besides humans. Thus, the expression ‘shared experience’ is used in a conservative way (i.e. two individuals simultaneously facing the same experience, regardless of their awareness that the other is facing it as well) throughout the manuscript.

2. Shared experiences in humans

In humans, it has been shown that both positive and negative shared experiences can affect the strength of social bonds (see [37] for measures of the strength of social bonds). Positive experiences can include, but are not limited to, sports [38–40], music-related activities [41,42], laughter [43,44] and (modest) alcohol consumption [45]. Also one of the most negative human experiences, warfare, is widely know to promote social bonds among soldiers [46]. A special case of negative experiences concerns the presence of pain (e.g. human rituals [47,48]—even though speculations over the psychological mechanisms involved in human rituals are beyond the scope of this work). The reasons why such shared experiences affect the strength of social bonds can be traced back to an increase of cooperation and activity coordination [38,40,46]. Some of the underlying mechanisms that promote and maintain the association between these benefits and the strengthening of social bonds are stress reduction/endorphine release [40,43].

Not only active experience sharing (e.g. singing or dancing together [40,42]), but also cases in which individuals more passively experience together an externally presented stimulus can influence social bonding. For example, both performing a response time task together, or watching a video together can promote social closeness, even since childhood [49,50]. Also the passive sharing of a painful experience appears to promote social bonds. For example, Bastian et al. [51] performed experiments in which a first group of people experienced a painful situation together (e.g. eating a hot chili pepper), while the control group experienced a neutral situation together (e.g. eating a hard candy). People sharing a negative (painful) experience were more likely to perceive a social bond and were more keen to cooperate with each other compared to people sharing a neutral experience. The reasons why such passively shared experiences affect the strength of social bonds can be due to joint attention [49]. Few studies have emphasized the fact that the type of stimulus may also play a role, with social closeness occurring in cases of belief-affirmation (but not in cases of belief-disaffirmation messages [52]), and in cases of high arousal/negative valence stimuli (rather than in the three other states of the Russell's circumplex model of affect [53]: [54]).

3. Shared experiences in non-human animals

Taking the development of human social bonds through shared experiences as the starting point, the main goal of this work is to stimulate further studies to better understand whether, and how, the same overall mechanisms apply in other vertebrate species, from fishes to non-human primates. Our main question concerns the role played by shared experiences in the development of a social bond/a partner preference when such shared experiences do not involve opportunities to cooperate or to exchange mutually beneficial actions (e.g. between-group interactions: [55–57]; predator inspection: [58]). Evidence that this could occur in non-human animals came from an experiment on guppies. Individuals who experienced high predation risk together showed preference for each other compared to those who did not [20]. Contrary to findings by Dugatkin and Alfieri [58] who showed that individuals cooperated for predator inspection, the simulation of a predation event (i.e. parts of a dead fish were put in the tank) may have not required any cooperation between individuals [20]. Still, fishes developed a preference toward each other. One could argue that partner preference was developed as a result of the increased spatial cohesion between the two individuals under the condition of high perception of predation risk. However, Heathcote et al. [20] controlled for the effect of spatial cohesion between the two individuals and they did not find differences between the two conditions (high perception of predation risk versus no perception of predation risk). Thus, what was the rationale for developing a social bond/ a partner preference? The same question is raised by a recent experiment showing that if chimpanzees watched a video together with a conspecific, they spent more time in proximity compared with conspecifics with whom, although in proximity, they did not actively watch a video together [59]. Similarly, if chimpanzees watched a video together with a human, they approached the human faster compared with those with whom they did not actively watch a video together [59].

4. Future directions

Apart from testing the role of shared experiences for the development of social bonds between individuals of different non-human species, three other questions arise. First, what is the difference between sharing positive or negative experiences for the development of social bonds? We could hypothesize that the difference has to do with the time needed to develop a social bond. Sharing a positive experience can trigger positive predisposition towards another individual (e.g. dancing in synchrony: [40]). A series of positive experiences may need to be repeated for a social bond to develop. On the contrary, a shared negative experience may instantaneously trigger a social bond [54] due to the urgent need for cooperation [58]. One of the proximate mechanisms promoting the instantaneous creation of a social bond may be the need for stress release [57].

Second, does the phase in the development of a social relationship play a role in the effect that positive or negative experiences have on social bonds? We would expect a stronger effect at the beginning of the social bond formation, compared to when the bond is already formed. According to Parkinson et al. [33], (human) friends have an extremely similar way of perceiving reality and an extremely similar neural response to stimuli. However, it is not clear whether individuals initially choose to associate with similar people (neural homophily), or whether similarities are the result of friends influencing each other. By sharing an experience at the onset of social bond formation, it is possible that a similar neural response is exerted on both individuals, thereby creating a shared perception from the beginning (see also [60]). The effect of a shared experience for individuals with a pre-existing social bond is maybe diluted by previous shared experiences and/or previous interactions.

Third, does individual heterogeneity [61] play a role in the effects that shared experiences may have on social bonds? Within-group individual variability is evolutionarily advantageous in a variety of contexts [62]. Focusing solely on variability in individual personality (only one aspect of individual heterogeneity), for example, we would predict a negative relationship between boldness levels and the strength of social bonds [63], a positive relationship between boldness levels and the time needed to develop a social bond [64] and between homophily in personality and strength in social bonds [65], regardless of the type of experience. As bold individuals are less risk averse [66] and they show an overall more consistent behaviour [67,68], we would also predict fewer differences in the effect of positive or negative experiences for the development of social bonds compared to shy individuals. We would also hypothesize differences in the effect of positive or negative experiences for the development of social bonds depending not only on individual boldness level but also on partner personality. Although boldness level is consistent over time [69], there may be short-term behavioural variations due to partner personality (i.e. shy individuals become bolder when associated with bold individuals: [67], but see also [70]) or previous social experiences (i.e. bold individuals become bolder if they where previously associated with shy individuals: [67]). Thus, we would expect a smaller effect of partner personality, but also a stronger effect of prior social experiences, in the case of bold compared to shy individuals. In all the predictions stated above, it is also worth considering that boldness increases as a response to fright stimuli (e.g. a simulated predation event), especially for shy individuals [68].

A series of laboratory experiments could help shed light on these unanswered questions. Such experiments would create positive, negative and neutral shared experiences for different dyads present in experimental groups. For example, these shared experiences could differ in the degrees of predation risk [20]. Effects of predation risk on individual and group behaviour are well known [71], but verifying whether and how it may affect the strength of social bonds would be a step forward in our knowledge. It is important to specify that individuals modify their behaviour based on their perception of predation risk [72], so experiments could include the playback of species-specific alarm calls and/or the use of model predators. Other types of experience include, but are not limited to, differences in environmental conditions (e.g. optimal temperature and oxygen concentration versus lower or higher temperature and suboptimal oxygen concentration) or in food quality (e.g. abundant highly preferred food versus abundant unpalatable food). Potential differences in the strength of social bonds between dyads who shared different experiences (e.g. positive versus negative of the same experience type) will be based on individual preferences that can be tested with a well-established procedure [27].

5. Conclusion

Although most of the literature on shared experiences and social bonds concern humans, we cannot exclude the presence of similar mechanisms in non-human species. Indeed, recent studies on guppies [20] and chimpanzees [59] found clear results with potentially encouraging interpretations. We think this may be a promising research path to follow with non-human animals as study subjects. Indeed, it may contribute to understand whether the psychological mechanisms behind human social behaviour are also subtly present in other organisms. A better understanding of how social bonds develop can give insight into both social cognition and on the ecological and social contexts under which sociality is promoted in different vertebrate species.

Data accessibility

This article has no data.

Authors' contributions

L.B. proposed and developed the idea, and wrote the first draft of the manuscript; M.G. developed the idea and contributed to the writing of the manuscript.

Conflict of interests

We declare we have no competing interests.

Funding

We received no funding for this study.