Do we acquire culture through other minds, or do we get access to other minds through culture?

“Culture” usually refers to the arts or to a people sharing customs and codes. In the case of music, these understandings are intertwined in two ways:

1. 1. Music is governed by customs defining the style. It almost invariably follows some system creating expectancies, implicitly known and applied by musicians and listeners.

2. 2. Musical styles typically shape neo-tribes (Bennett Reference Bennett1999) of followers who share a spectrum of social codes and customs.

Music culture, thus, is a practice as well as neo-tribes of followers. Both aspects are governed by patterned behaviour, generating anticipation.

The Austrian phenomenologist Alfred Schütz, contemplating music listening, once wrote:

Schütz puts the listener in the head of the musician. He is not proposing that listeners follow or imitate the composer, but perform the music with the composer.

Co-performance with a musician implies shared brain activity as well as a shared time dimension. Except for the parallel activation of the auditory systems, several mutual pre-motor areas are activated. The shared time dimension is induced through entrainment (synchronization to musical beats). Synchronization demands prediction. Periodical sounds produce bursts of neural oscillations on the beats. These bursts continue when the stimulus is omitted (Tal et al. Reference Tal2017). They are, thus, predictions. “Entrainment,” says anthropologist Judith Becker, “is the strongest form of interaction … It actualizes a supra-individual state” (Becker Reference Becker2001).

Synchronization also depends on familiarity with the piece and/or style. But, even when the prediction is wrong, synchronized activities occur, namely the processing of prediction errors. Deviances from musical customs elicit oscillatory reactions (the event-related potential ERAN), not only in the listener, but even in the musician (who planted the surprising event) (Koelsch et al. Reference Koelsch, Vuust and Friston2019). Surprises attract attention. In a sequence, changes between the expected and the unexpected are felt. This exemplifies listening as a participation in the musician's stream of consciousness.

Overt synchronization engenders pro-social behaviour (Repp & Su Reference Repp and Su2013). Further, the ability to synchronize is improved by oxytocin – a peptide associated with social integration (Gebauer et al. Reference Gebauer2016). These findings may be associated with bird courtship, where the male and the female of some species engage in synchronized rituals, and with human courtship, where the dance floor provides a scene for synchronization. The synchronized other, thus, can be considered an affordance – an epistemic resource, reducing uncertainty concerning the other's suitability as a partner. It confirms: same species, same fitness, and same interest.

Subjectivity involves emotions, and inter-subjectivity thus entails shared emotions. With an emotional completion, the TTOM theory would catch how we, not just think, but live through other minds.

Could emotions be computed along the lines of the free-energy principle (FEP)? Interoceptive inference entails that neural activity caused by visceral reactions is compared to predictions reflecting homoeostasis (Seth & Friston Reference Seth and Friston2016). The only ascending information is the degree of deviance. Is it possible to access the rich palette of emotions this way? It has been argued from an enactive perspective that “emotional experience simply cannot be reduced to a frame of reference” (Roesch et al. Reference Roesch, Nasuto and Bishop2012).

In the circumplex model of affect, emotions are plotted in a two-dimensional space defined by valence and arousal coordinates (Russell & Pratt Reference Russell and Pratt1980). Valence, according to FEP, could be defined as “the negative rate of change of free energy over time” (Joffily & Coricelli Reference Joffily and Coricelli2013). This means that reduction of uncertainty is rewarding. This definition offers a solution to the “dark-room problem” (Friston et al. Reference Friston, Samothrakis and Montague2012): If there is no uncertainty to reduce, the organism cannot be happy. Valence over certainty must follow an inverted U-curve, starting in boredom, ending in chaos, and peaking at an optimal level of active inference, as defined by Seth and Friston (Reference Seth and Friston2016). Neither the uneventful, nor chaos can provide the epistemic affordance needed for active inference. Arousal could be caused by the degree and precision of the prediction error. This is in accordance with the pattern of the piloerection reaction to music, which is often caused by music expressing longing and triggered by an unexpected change in the music (e.g., a change of tonality). Here, physiologically assessed arousal, which may reflect uncertainty, is followed by relaxation, which may reflect certainty (Vickhoff et al. Reference Vickhoff, Åström, Theorell, von Schéele and Nilsson2012). The experience is delightful.

Musical emotion is an excellent subject to test interoceptive inference. Music is computable, it plays with expectancies and it makes us happy without obvious rewards.

A short answer to the initial question: Music exemplifies that we do not just acquire culture through other minds. It is a reciprocal dependency. Music, written, or sounding, is a script guiding action. So are, to varying degrees, many rituals and customs. Collective co-performance of the script enables inter-subjectivity, which arguably contributes to the formation of neo-tribes. Shared-emotional experiences give material to the narrative of who we are.