Both Savage et al.'s and Mehr et al.'s articles draw on evidence from infancy and early childhood to strengthen their arguments, yet in our view they do not adequately acknowledge the critical importance of developmental timing especially during middle childhood. Both papers take a somewhat static approach to describing musical abilities during childhood, essentially equating the very presence of any sensitivity to a musical structure during early childhood with the adult form of that capacity. Coordinated group behaviors are at the center of both papers, and both assume that because the newborn brain responds to missing events in a drum pattern, the capacity for drumming and dance are present at birth. Although they acknowledge the slower developmental trajectory of precise entrainment, they nevertheless conclude that “the capacity to perceive and move to a beat/dance is a mode of participation that is accessible to large numbers of individuals regardless of age, familiarity with the music, or instrumental/singing virtuosity.” We argue that these abilities are not necessarily accessible regardless of age or familiarity, and that a more nuanced account of when and how musical abilities and behaviors develop may hold the key to understanding the co-evolution of musicality and music.

Framing musicality as a gradually gated developmental process rather than a static suite of capacities could reveal why some cultural variants are favored over others, and what functions are served by particular musical features and behaviors. Similar to language, musical systems are transmitted across generations, so knowing something about who is learning and when learning occurs may be important for characterizing the pressures that act upon those systems. Because musical and linguistic systems only survive if children can learn them, systems must maximize transmissibility by adapting to the capacities of the learner and the acquisition process. In language evolution, children's memory limitations and learning biases may give rise to increased grammatical complexity in languages that are learned or created primarily by children (Christiansen & Chater, Reference Christiansen and Chater2008; Senghas, Kita, & Özyürek, Reference Senghas, Kita and Özyürek2004), and reduced complexity in populations with more adult learners (Lupyan & Dale, Reference Lupyan and Dale2010). In music, a recent study suggests rhythms also evolve as they are learned, reproduced, and transmitted across multiple “generations” of adult participants, and initially random sequences become increasingly structured, learnable, and consistent with observed musical universals (Ravignani, Delgado, & Kirby, Reference Ravignani, Delgado and Kirby2016). If children learn differently than adults, then understanding these developmental constraints could explain how and why certain features exist across diverse musical systems.

Drumming and dancing are emphasized by both target papers, but if these abilities are assumed to be drivers of evolutionary processes, understanding their developmental timing is as important as systematic examination of variation across cultures. In order to dance to music, a listener must infer the beat and meter from cues in the musical surface, internally sustain that structure, and coordinate movements with the music and others in culturally appropriate ways. Although we agree that detecting changes, discriminating rhythms, and moving to music are precursors to mature adult behaviors, a growing body of evidence suggests that childhood capacities are starkly different than those observed during adulthood. Although some studies describe preschoolers performing “synchronous drumming” whenever drumming is not random (e.g., Kirschner and Tomasello, Reference Kirschner and Tomasello2009), other evidence suggests children's synchronization to music does not approach adult-like levels until late childhood (McAuley, Jones, Holub, Johnston, & Miller, Reference McAuley, Jones, Holub, Johnston and Miller2006; see Hannon, Nave-Blodgett, & Nave, Reference Hannon, Nave-Blodgett and Nave2018). Similarly, although children can tell when a metronome matches the music, they are less accurate than adults and show no sensitivity to simultaneous metrical levels until early adolescence (Einarson & Trainor, Reference Einarson and Trainor2016; Nave-Blodgett, Hannon, & Snyder, Reference Nave-Blodgett, Hannon and Snyder2020). These findings parallel recent evidence that language learning may have a more protracted developmental time course than previously assumed (Hartshorne, Tenenbaum, & Pinker, Reference Hartshorne, Tenenbaum and Pinker2018; McMurray, Danelz, Rigler, & Seedorff, Reference McMurray, Danelz, Rigler and Seedorff2018).

Although we agree that musical behaviors play an important role during infancy and early childhood, middle childhood is also a critical period of development. During this time, children exhibit an increase in the ability to learn and transmit knowledge across myriad domains (e.g., cognitive, behavioral, and social) as they begin to physically and hormonally mature (Del Giudice, Reference Del Giudice2009; Lancy & Grove, Reference Lancy and Grove2011) and better understand their role within a larger social network (House et al., Reference House, Kanngiesser, Barrett, Broesch, Cebioglu, Crittenden and Yilmaz2020). The transition to middle childhood is characterized by increased motor control and complex reasoning skills (Weisner, Reference Weisner, Sameroff and Haith1996) that emerge alongside increases in prosociality (Crittenden & Zes, Reference Crittenden and Zes2015; Fehr, Glätzle-Rützler, & Sutter, Reference Fehr, Glätzle-Rützler and Sutter2013). Evolutionary models of childhood have proposed that middle childhood is a “developmental switch point” where genetic and environmental inputs converge to act on phenotypes (Del Giudice, Reference Del Giudice2009). This stage is characterized by heightened sensitivity to the environment, potentially mediated by the expression of new genetic or epigenetic factors (Del Giudice, Reference Del Giudice, Halfon, Forrest, Lerner and Faustman2018), and is accompanied by the acquisition of large amounts of knowledge through imitation, teaching, and other forms of cultural learning (Hewlett, Fouts, Boyette, & Hewlett, Reference Hewlett, Fouts, Boyette and Hewlett2011; Kline, Boyd, & Henrich, Reference Kline, Boyd and Henrich2013).

In music, this learning may entail increased awareness of metrical hierarchies and a stronger grasp of the relationship between the musical surface, meter, and rhythmic movements in one's culture. For example, in Western music, events are statistically more likely to occur at strong metrical positions, but recent study suggests that this is not the case for some non-Western corpora (Holzapfel, Reference Holzapfel2015; London, Polak, & Jacoby, Reference London, Polak and Jacoby2016). If the ability to perceive meter does not emerge until middle childhood or adolescence (Nave-Blodgett et al., Reference Nave-Blodgett, Hannon and Snyder2020), this means that the capacity to participate in more complex forms of musical coordination (such as group or partner dances) is only just emerging at a time when individuals are refining their social skills and paying greater attention to social models. Importantly, this suggests that capacity to dance varies with age and familiarity, and it undermines the notion that displays of group synchrony are perceptible to all on-lookers. We advocate for a more thorough treatment of existing developmental evidence, including incorporation of middle childhood as a key developmental period, and further research addressing how and when individuals learn and master specific musical structures and behaviors across cultures.