What can music ensembles tell us about social cognition and interaction?

D'Ausilio et al. note that most studies of how cognition and brain organization is shaped by social factors have used subjects in defined experimental settings, rather than natural ones. The problem is that experimental rigor is inversely related to ecological naturalness. They suggest that musical ensemble performance offers a promising solution for balancing the trade-off between experimental control and ecological naturalness. Here is their list of features that make music a promising avenue for social cognition research.

Ecological validity: ensemble musicians participate in a socially-relevant interaction, obviating the need to introduce an artificial task, manipulation, or training to induce a social context.

Motivational factors: motivation is an inherent part of music and hence it is not necessary to employ extrinsic techniques (monetary compensation, competition, or response-contingent reward) to trigger the emergence of interaction.

Generalizability: musicality is a widespread human capacity, enabling almost everyone to sing together with others and to produce rhythms through body movements (e.g., simple drumming or dance).

Multi-level interactivity: information transfer is both continuous (body movements) and discrete (musical sounds). Furthermore, musicians’ movements that function to produce sound on an instrument can be dissociated from those that are not necessary for sound production (e.g., ancillary movements that serve expressive functions) [15]. These properties allow the investigation of the multi-level communicative functions (hierarchical musical structure and expressive intentions) of musical social interaction.

Temporal dependencies: information transfer is not only based on the content of an individual's instantaneous response but also is affected by rhythmic timing, tempo, and the degree of interpersonal synchrony.

Formal description of interaction: the musical score is a script-like description of the interaction that the experimenter can manipulate to control the emergence of social structures and different roles (e.g., leader vs follower) played by each musician.

The authors briefly review research that that traverses a continuum of ecological interaction. These classes include an individual interacting with a recording, a computer-controlled virtual partner that responds to the individual, another individual in a duo, multiple individuals in mixed ensembles (extending to large orchestras), and others in the presence of a live audience.