Savage et al. argue for musicality as having evolved for the overarching purpose of social bonding. By way of contrast, we highlight contemporary predictive processing models of human cognitive functioning in which the production and enjoyment of music follows directly from the principle of prediction error minimization.

Studying a complex cultural phenomenon like music requires many kinds of expertise. Savage et al. adopt a pluralistic approach, considering multiple forms of evidence and perspectives from multiple fields. This commentary argues that a similar scholarly ecumenicism should be embraced by more studies of music and other cultural phenomena.

Savage et al. propose that music filled a hypothetical “bonding gap” in human sociality by Baldwinian gene-culture coevolution (or protracted cognitive niche construction). Both these stepping stones to an evolutionary account of the function and origin of music are problematic. They are scrutinized in this commentary, and an alternative is proposed.

The music learning environment is a context in which fundamental forces and values underlying human musicality may be evident. Social bonding within music-making groups is characterized by a high degree of complexity whereas issues of clarity, accuracy, and coordination remain the focus of learning. Physical and cognitive impairments that compromise music learning opportunities offer a critical test of music's link to social bonding.

Collective, synchronous music-making is far from ubiquitous across traditional, small-scale societies. We describe societies that lack collective music and offer hypotheses to help explain this cultural variation. Without identifying the factors that explain variation in collective music-making across these societies, theories of music evolution based on social bonding (Savage et al.) or coalition signaling (Mehr et al.) remain incomplete.

Both of the companion target articles place considerable performance on music performance ability, with specific attention paid to singing in harmony for the music and social bonding (MSB) hypothesis proposed by Savage and colleagues. In this commentary, I evaluate results from recent research on singing accuracy in light of their implications for the MSB hypothesis.

Savage et al. make a compelling case, Mehr et al. less so, for social bonding and credible signalling, respectively, as the main adaptive function of human musicality. We express general advocacy for the former thesis, highlighting: (1) overlap between the two; (2) direct versus derived biological functions, and (3) aspects of music embedded in cultural evolution, for example, departures from tonality.

A cross-species perspective can extend and provide testable predictions for Savage et al.'s framework. Rhythm and melody, I argue, could bootstrap each other in the evolution of musicality. Isochrony may function as a temporal grid to support rehearsing and learning modulated, pitched vocalizations. Once this melodic plasticity is acquired, focus can shift back to refining rhythm processing and beat induction.

Focus on the evolutionary origins of musicality has been neglected relative to attention on language, so these new proposals are welcome stimulants. We argue for a broad comparative approach to understanding how the elements of musicality evolved, and against the use of overly simplistic evolutionary accounts.

To corroborate the music and social bonding hypothesis, we propose that future investigations isolate specific components of social bonding and consider the influence of context. We deconstruct and operationalize social bonding through the lens of social psychology and provide examples of specific measures that can be used to assess how the link between music and sociality varies by context.

Music is an artistic cultural innovation, and therefore it may be considered as intuitive thought expressed in symbols, which can efficiently convey multiple meanings in learning, thinking, and transmission, selected for and passed on through cultural evolution. The symbolic system has personal adaptive benefits besides social ones, which should not be overlooked even if music may tend more to the latter.

Based on their social bonding hypothesis, Savage et al. predict a relation between “musical” behaviors and social complexity across species. However, our qualitative comparative review suggests that, although learned contact calls are positively associated with complex social dynamics across species, songs are not. Yet, in contrast to songs, and arguably consistent with their functions, contact calls are not particularly music-like.

Although it can be straightforward to define the features of physical traits, complex cultural categories tend to elude widely accepted definitions that transcend cultural and historical context. Addressing papers by Mehr et al. and Savage et al., which both aim to explain music as an evolved trait, we discuss fundamental problems that arise from their conceptualizations of music.

By focusing on the contributions of subcortical structures, our commentary suggests that the functions of the hippocampus underlying “displacement,” a feature enabling humans to communicate things and situations that are remote in space and time, make language more effective at social bonding. Based on the functions of the basal ganglia and hippocampus, evolutionary trajectory of the subcomponents of music and language in different species will also be discussed.

We extend Savage et al.'s music and social bonding hypothesis by examining it in the context of Chinese music. First, top-down functions such as music as political instrument should receive more attention. Second, solo performance can serve as important cues for social identity. Third, a right match between the tones in lyrics and music contributes also to social bonding.

Music is part of the cultural practice and, at the same time, is interwoven with biology through its effects on the brain and its likely evolutionary origin. Studies on music, however, are traditionally based on the humanities and often carried out in a purely historical context, without much input from neuroscience and biology. Here, we argue that lullabies are a particularly suited test case to study the biological versus cultural aspects of music.

Mehr et al.'s hypothesis that the origins of music lie in credible signaling emerges here as a strong contender to explain early adaptive functions of music. Its integration with evolutionary biology and its specificity mark important contributions. However, much of the paper is dedicated to the exclusion of popular alternative hypotheses, which we argue is unjustified and premature.

We propose an approach reconciling the ultimate-level explanations proposed by Savage et al. and Mehr et al. as to why music evolved. We also question the current adaptationist view of culture, which too often fails to disentangle distinct fitness benefits.

By fostering bonding (Mehr et al.; Savage et al.), music illustrates marvelously its ability to induce emotional experience. But, music can induce emotion more generally as well. To help explain how music fosters bonding and induces other emotions, I propose that music derives this power from the evolution of what I term “gestural messaging.”

Credible signaling may have provided a selection pressure for producing and discriminating increasingly elaborate proto-musical signals. But, why evolve them to have hierarchical structure? We argue that the hierarchality of tonality and meter is a byproduct of domain-general mechanisms evolved for reasons other than credible signaling.