Both target articles acknowledge that human musicality evolved to serve more than one adaptive function. Although Mehr, Krasnow, Bryant, and Hagen exalt credible signaling to a unitary mechanism, they eventually promote two ultimate-level explanations in terms of credible signaling of coalition strength and parental attention, manifesting in divergent forms of music (i.e., loud drumming/chanting vs. lullabies). In this way, one of their arguments opposing Savage et al.'s music and social bonding (MSB) hypothesis – namely that (some types of) musics are poorly designed to coordinate groups – can be turned against themselves in that lullabies are poorly designed to signal coalition strength and war chants are poorly designed to signal parental attention. This functional multiplicity leads Savage and colleagues to search for an overarching super-function through synthesis and extension of previous proposals.

Savage et al.'s MSB account is convincing in many aspects. Yet, their overview of candidate neurobiological mechanisms exposes the somewhat rudimentary state of research in this area. For example, their Figure 3 treats the oxytocinergic and endogenous opioid systems under the same umbrella despite comprising distinct – albeit interconnected (Kovatsi & Nikolaou, Reference Kovatsi and Nikolaou2019) – systems. In light of oxytocin's well-established role in social learning and memory (Chini, Leonzino, Braida, & Sala, Reference Chini, Leonzino, Braida and Sala2014), direct links of oxytocin to learning/knowledge (e.g., through laryngeal motor cortex, ventral tegmental area, or Broca's area; Theofanopoulou, Boeckx, & Jarvis, Reference Theofanopoulou, Boeckx and Jarvis2017) could also be considered in addition to indirect links via the dopaminergic system (Baskerville & Douglas, Reference Baskerville and Douglas2010; Love, Reference Love2014). Although the detailed interactions of these three systems lie beyond the scope of a brief commentary, it suffices to say that the overarching super-function envisioned by the target authors indeed requires a unifying neurobiological framework.

Given its manifold roles in a multitude of peripheral and central nervous processes with direct relevance for the proposed adaptive functions of music (as depicted in Fig. 1), we suggest that the nonapeptide oxytocin may contribute crucially to such a unifying neurobiological framework. Specifically, oxytocin-mediated increases in generosity (Zak, Stanton, & Ahmadi, Reference Zak, Stanton and Ahmadi2007), emotional face recognition (Shahrestani, Kemp, & Guastella, Reference Shahrestani, Kemp and Guastella2013), and (potentially) trust (Kosfeld, Heinrichs, Zak, Fischbacher, & Fehr, Reference Kosfeld, Heinrichs, Zak, Fischbacher and Fehr2005; but see Declerck, Boone, Pauwels, Vogt, & Fehr, Reference Declerck, Boone, Pauwels, Vogt and Fehr2020) may increase music's effects on group cohesion resulting in social coping benefits. Changes in eye gaze (Eckstein et al., Reference Eckstein, Bamert, Stephens, Wallen, Young, Ehlert and Ditzen2019), selective sociality (Carter, Reference Carter2017), and sexual arousal and orgasm (Alley & Diamond, Reference Alley and Diamond2020) mediated by oxytocin may enhance musical effects on courtship behavior and exclusionary pair formation. Oxytocin-mediated empathy (Wu, Li, & Su, Reference Wu, Li and Su2012), playful behavior (Szymanska, Schneider, Chateau-Smith, Nezelof, & Vulliez-Coady, Reference Szymanska, Schneider, Chateau-Smith, Nezelof and Vulliez-Coady2017), and peripheral effects on parturition and lactation (Carter, Reference Carter2014) could underlie musical infant–caregiver bonding. Moreover, oxytocin-related increases in synchrony (Gebauer et al., Reference Gebauer, Witek, Hansen, Thomas, Konvalinka and Vuust2016; Josef, Goldstein, Mayseless, Ayalon, & Shamay-Tsoory, Reference Josef, Goldstein, Mayseless, Ayalon and Shamay-Tsoory2019) along with out-group gloating (Shamay-Tsoory et al., Reference Shamay-Tsoory, Fischer, Dvash, Harari, Perach-Bloom and Levkovitz2009), derogation (De Dreu, Greer, Van Kleef, Shalvi, & Handgraaf, Reference De Dreu, Greer, Van Kleef, Shalvi and Handgraaf2011), and threat perception (Egito, Nevat, Shamay-Tsoory, & Osório, Reference Egito, Nevat, Shamay-Tsoory and Osório2020) may promote competitive musical coalition signaling. Oxytocin's analgesic (González-Hernández, Rojas-Piloni, & Condés-Lara, Reference González-Hernández, Rojas-Piloni and Condés-Lara2014) and anxiolytic (Lancaster et al., Reference Lancaster, Goldbeck, Pournajafi-Nazarloo, Connelly, Carter and Morris2018) effects may, in turn, enhance music's facilitation of coordinated physical labor. Finally, oxytocin-enhanced social learning and memory (Graustella & MacLeod, Reference Graustella and MacLeod2012) could increase music's capacity for preserving civilizational knowledge and forming shared cultural identities.

Figure 1. Overview of how the socio-allostatic effects of oxytocin may contribute to a unifying neurobiological framework for music's adaptive functions. Specifically, music making – which often occurs in social contexts – may lead to oxytocin release that promotes internal adjustment of sensing and response set-points as well as assists learning and prediction processes with reference to the external physical and social environment. These allostatic optimizations may, in turn, increase the efficacy of music's adaptive functions as a facilitator of social cohesion, pair formation, parental bonding, coalition signaling, group coordination, and preservation of communal knowledge and identity. These adaptive functions of music are each associated with specific musical forms employing distinct acoustic features. Such function-to-form associations are, however, statistical (rather than deterministic) as well as highly interconnected in that dance, for example, can contribute to multiple adaptive functions such as cohesion, courtship, coalition, and coordination, depending on the context in which it appears.

Importantly, the two target theories propose opposite causal relationships between oxytocin and musical synchronization. Although MSB predicts that joint music making would increase oxytocin levels, the credible signaling theory predicts that higher oxytocin levels achieved through pre-existing social cohesion would increase musical synchronization. Burgeoning evidence for both of these effects (former: Grape, Sandgren, Hansson, Ericson, & Theorell, Reference Grape, Sandgren, Hansson, Ericson and Theorell2002; Keeler et al., Reference Keeler, Roth, Neuser, Spitsbergen, Waters and Vianney2015; latter: Gebauer et al., Reference Gebauer, Witek, Hansen, Thomas, Konvalinka and Vuust2016; Josef et al., Reference Josef, Goldstein, Mayseless, Ayalon and Shamay-Tsoory2019) suggests that both hypotheses may ultimately be viable. Yet, findings are sometimes contradictory (e.g., Schladt et al., Reference Schladt, Nordmann, Emilius, Kudielka, de Jong and Neumann2017), sample sizes are typically low (Walum, Waldman, & Young, Reference Walum, Waldman and Young2016), and intranasal administration and salivary assays of oxytocin have faced methodological criticisms (Leng & Ludwig, Reference Leng and Ludwig2016; McCullough, Churchland, & Mendez, Reference McCullough, Churchland and Mendez2013). Theorizing, therefore, converges on oxytocin effects being highly individualistic/context-dependent (Bartz, Zaki, Bolger, & Ochsner, Reference Bartz, Zaki, Bolger and Ochsner2011) and modulatory/interactive rather than primary (Marsh, Marsh, Lee, & Hurlemann, Reference Marsh, Marsh, Lee and Hurlemann2020).

Consistent with this view, an allostatic theory of oxytocin has recently gained traction (Quintana & Guastella, Reference Quintana and Guastella2020). This theory posits that oxytocin facilitates the adjustment of sensing and response set-points and assists learning and prediction in ways that proactively optimize systemic adaptation with reference to a constantly changing environment. The resulting criticality in brain states and behavioral flexibility are beneficial under many circumstances – but perhaps especially so in complex social interactions such as music and dance where ongoing anticipation of environmental consequences is essential (D'Ausilio, Novembre, Fadiga, & Keller, Reference D'Ausilio, Novembre, Fadiga and Keller2015). Beyond the individual level, oxytocin's allostatic effects may further extend to social allostasis in that behavioral flexibility can resolve vicious circles of rigid interaction patterns and inflexible group dynamics (Saxbe, Beckes, Stoycos, & Coan, Reference Saxbe, Beckes, Stoycos and Coan2020; Schulkin, Reference Schulkin2011). A socio-allostatic view of oxytocin, moreover, connects well with intellectual currents such as enactive cognition and biocultural evolution where biological and cultural dimensions constitute factors in the same evolving system of living agents embedded in an environment (van der Schyff & Schiavio, Reference van der Schyff and Schiavio2017).

Savage et al. (2020) propose that music and musicality coevolved through an iterated Baldwin process whereby musical behaviors were culturally invented to tackle the environmental challenges of social living which gradually manifested as adaptive musicality traits (Podlipniak, Reference Podlipniak2017). The neurochemical feedback loop depicted in Figure 1 is consistent with such a process of gene-culture coevolution in that genetic mutations leading to tighter connections between music making and oxytocin release would accelerate the extent to which the adaptive functions of music can capitalize upon oxytocin-mediated sensing, learning, prediction, and response processes. The proposed feedback mechanism whereby oxytocin in itself stimulates more oxytocin release (Grippo et al., Reference Grippo, Pournajafi-Nazarloo, Sanzenbacher, Trahanas, McNeal, Clarke and Sue Carter2012) may further accelerate this process.

When oxytocin-mediated sensing, learning, prediction, and response capacities are sharpened in the encounter with the in-group, phenomena such as social cohesion arise, which in turn promote coordinated labor as well as civilizational mnemonics and identity formation (Fig. 1). When faced with an out-group, heightened social allostasis may manifest in coalition signaling. Finally, when presented with offspring or potential mates, infant–caregiver bonding and dyadic courtship result, respectively. These effects could be mediated by the manifold empirically substantiated effects of oxytocin on human behavior summarized above.