Exploring the innards of human social coordination: a dynamical analysis of continuous EEG oscillations

J. A. Scott Kelso, Emmanuelle Tognoli, Daniela Benites, Gonzalo C. de Guzman


The brain is fundamentally a dynamical system; its functions emerge when patterns of activity change over time. In several minutes of brain recordings that are typically aggregated to obtain “reliable” estimates of spectral power, hundreds to thousands of oscillatory patterns succeed to one another in the 10Hz band alone. Key information about the brain machinery is engulfed in the process of averaging, for instance: -what is the unique functional contribution of a pattern?-, -which patterns follow which others and why?-, -how pattern duration and recurrence interplay with functional efficacy?-, etc. The loss of information from temporal averaging is all the more critical in complex activities such as social coordination. To understand the brain mechanisms unfolding during behavioral coordination between people, we focused on the transition between coordinated and uncoordinated social behavior. Pairs of subjects performed rhythmic right index finger movements in full view of one another under instructions to establish one of three collective behaviors: inphase (both fingers flex/extend in synchrony), antiphase (one subject extends while the other flexes) and intrinsic (both subject maintain their own movement and do not intend to synchronize). Behavioral transitions were identified and classified according to collective behavior (transition to synchronized or unsynchronized behavior), agency (i.e., who effects the transition), task goal and behavioral strategy. We applied a 4D colorimetric mapping to both subjects’ EEG to identify brain spatio-temporal patterns associated with these behavioral transitions. Each transient network of brain activity creates a spatio-temporal pattern of EEG that recurs under renewed functional demand. In each family of spatio-temporal patterns, we studied the distribution of dependent and independent behavioral variables and identified combinations of factors that were significantly more frequent than chance. We also examined the interactions between activities in both brains by analyzing the tendency for coincident switching between oscillatory patterns. Results are put in the context of future work on the full temporal organization of brains’ functional processes of social behavior.  [PDF].