References:

1
Neural correlates of social coordination
Spatiotemporal analysis of brain and behavior
Armin Fuchs, Emmanuelle Tognoli, Daniela
Benites, J.A. Scott Kelso
Center for Complex Systems and Brain Sciences
239.17
Florida Atlantic University, Boca Raton, FL 33431
Segmentation Procedure
Introduction Social coordination is manifested
on different levels, e.g. in situations where
people have to respond and anticipate each other
as in sports games or dancing, or in experiments
designed to investigate cooperation and
competition between individuals. Neuromarkers of
such behaviors have been identified in the a-band
in EEG 1 and also correlated with EEG segments
obtained by visual inspection of the brain
recordings 2. So far, the segmentation
procedure had to be carried out manually and was
extremely time consuming. Here we present an
automated process based on the rotating wave
approximation and temporal correlations in the
resulting signal. Results indicate that the
segmentations obtained from this procedure are as
powerful in identifying neurocorrelates of social
coordination as those form visual inspection by a
trained analyst.

• Fig. 1 Experimental setup
• Dual EEG recordings
• t0-20s screen opaque
• t0-40s screen transparent
• Coordination modes
• in-phase
• anti-phase
• ignore partner

• Conclusions
• An algorithmic segmentation was developed that
  can replace time consuming manual procedures
• The algorithm extracts segments from EEG data
  whose boundaries exhibit correlations between
  brain activity and behavioral data in social
  coordination
• Correlation were found between single subjects
  behavior and EEG in the a-range as well as
  between movement of the subject that adapts its
  rate and the EEG of the subject that continues
  moving at the same frequency
• Further analysis of other coordination conditions
  and subject pairs will allow to fine-tune the
  procedure.

Fig. 2 Coordination behavior during a single
trail from the in-phase condition. Normalized
movement for subjects labeled red and blue
showing no coordination and wrapping of the
relative phase (magenta) for tlt20s (screen
opaque), a transition at t20s and in-phase
coordination for tgt20s (screen transparent).
Note that in-phase coordination is established
by subject red changing its frequency and
adjusting to subject blue who continues moving at
the same rate. This behavior (red adjusts to
blue) is observed in almost all of the 20 trials
of this condition.
References 1 E. Tognoli, J. Lagarde, G.C.
DeGuzman, J.A.S. Kelso (2006). The phi complex as
a neuromarker of human social coordination.
PNAS 104 8190-8195 2 E. Tognoli, J.A.S. Kelso
(2009, pending) System and method for analysis of
spatio-temporal data. WO 2010/016992 A2.
Publication under Patent Cooperation Treaty.
World Intellectual Property Organization.
Acknowledgement Work supported by NINDS grant
48299 (JASK).