Living in groups involves a balance of costs and benefits and, especially for primates, this inevitably entails competition and cooperation with the other group members. If food resources are scarce and all individuals are competitive, splitting into smaller groups will be favored. However, if group members cooperate with each other, for example by hunting together, larger groups become advantageous. Despite the high costs of investing in bonding relationships with non-kin, one prominent characteristic of primate society is the large and complex group memberships including non-kin group members (Byrne & Whiten, Reference Byrne and Whiten1988; Harcourt & de Waal, Reference Harcourt and de Waal1992), and conducive social skills and bonding mechanisms have, therefore, evolved among primates (Van Schaik, Reference Van Schaik2016).

Social play, for example, generally follows a bell-shaped curve through ontogeny with a peak in the juvenile phase (Fagen, Reference Fagen, Pereira and Fairbanks1993); however, primates in particular continue to play as adults (Palagi, Reference Palagi2018). Some species such as great apes play frequently even when they become adults, and researchers have proposed that social play functions to share positive emotions (e.g., laughing) and reduce stress or tension (Palagi et al., Reference Palagi, Burghardt, Smuts, Cordoni, Dall'Olio, Fouts, Řeháková-Petrů, Siviy and Pellis2006a, Reference Palagi, Cordoni, Demuru and Bekoff2006b), effects that have parallels with music and dance. It is reported that the relative striatal volume is correlated with the rate of social (but not non-social) play behavior (Graham, Reference Graham2011). The striatum structure is intimately connected to dopaminergic pathways, and thus to the experience and anticipation of pleasure and reward, as well as to sequential behavior, motor control, and cognitive flexibility (Ashby, Turner, & Horvitz, Reference Ashby, Turner and Horvitz2010; Elliott, Newman, Longe, & Deakin, Reference Elliott, Newman, Longe and Deakin2003; Erikson et al., Reference Erikson, Boot, Basak, Neider, Prakesh, Voss, Graybiel, Simons, Fabiani, Gratton and Kramer2010; Vink et al., Reference Vink, Kahn, Raemaekers, van den Heuvel, Boersma and Ramsey2005). Given that social play, along with grooming, plays a crucial role in social bonding, which in turn may have fitness benefits (e.g., in chimpanzees, Muller & Mitani, Reference Muller and Mitani2005), selection may have favored a link between play and an enhanced ability to make sequential behaviors for rhythmic and repetitive joint actions with others, and experiencing pleasure and reward through those activities.

However, among primates, humans are extremely highly motivated to share psychological states with others (shared intentionality, Tomasello & Carpenter, Reference Tomasello and Carpenter2007). As “togetherness” or “jointness” is what researchers consider to distinguish cooperative or collaborative interactions between humans and nonhuman primates (Tomasello & Moll, Reference Tomasello and Moll2010), a bonding system such as music might have been selected for such purposes. Some research has led to the suggestion that coalition signaling did not evolve in most nonhuman primate societies because their bonding relationships are not strong enough, and so they succumb to the collective action problem (CAP) or social dilemma, when all individuals would benefit by cooperating but fail to do so because of conflicting interests that discourage joint action (Willems & Van Schaik, Reference Willems and Van Schaik2015). This suggests that credible signaling of coalition strength, size, and cooperation ability mentioned in Mehr et al. can evolve after group members have strongly bonded relationships.

Concerning the cross-species approach, the perception/production dissociation mentioned by Savage et al. should be possible to explore in nonhuman primates. Indeed, although imitation is relatively rare in nonhuman primates, they recognize when their actions are being imitated (Haun & Call, Reference Haun and Call2008; Paukner, Anderson, Borelli, Visalberghi, & Ferrari, Reference Paukner, Anderson, Borelli, VOsalberghi and Ferrari2005) and show more affiliative responses to those who recently imitated them (Paukner, Suomi, Visalberghi, & Ferrari, Reference Paukner, Suomi, Visalberghi and Ferrari2009). In vocal communication, some species such as Japanese macaques match acoustic features (i.e., frequency range) when they call back to other group members (Sugiura, Reference Sugiura1998). However, whether acoustic similarity increases bonding in relationships has not been rigorously tested. Additionally, I suggest that abilities or predispositions related to music-making but which are not components of the current communication system, should also be considered for understanding the biological roots of musicality. For example, chimpanzees, but not monkeys, seem to be intrinsically motivated to match their body movement (i.e., tapping) to auditory rhythms (Hattori, Tomonaga & Matsuzawa, Reference Hattori, Tomonaga and Matsuzawa2013; Reference Hattori, Tomonaga and Matsuzawa2015; Zarco, Merchant, Prado, & Mendez, Reference Zarco, Merchant, Prado and Mendez2009), although it is a rudimentary capacity and entrainment occurs only when the auditory tempo is close to the individual's spontaneous motor tempo. We also recently reported that rhythmic body movements such as swaying can be induced in chimpanzees (Hattori, Reference Hattori, Anderson and Kuroshimain press; Hattori & Tomonaga, Reference Hattori and Tomonaga2020a, Reference Hattori and Tomonagain press). A similar effect has been reported in the wild, in the form of chimpanzees' rhythmic displays and vigorous charges in reaction to rain or waterfalls (Goodall, Reference Goodall1986). Although neither the rain dance nor waterfall dance appears to be a joint activity involving multiple individuals, components of musicality may have arisen based on those predispositions in chimpanzees' and also humans' ancestors. As nonhuman animals do not themselves make music, any proto-musical behaviors have their own species-specific variations. Understanding such variations in relation to bonding systems is a laudable aim of further research in the comparative approach to the biology of music.