Information integration and the latent consciousness of human groups

We have applied Integrated Information Theory, a mathematical framework for studying consciousness, to human collectives. Specifically, we have analyzed soccer teams, drawing parallels between neuronal activity in the brain and player activity on the field. We investigated whether soccer teams integrate information in a quantifiable manner, and whether this predicts team performance. Utilizing player position data from 100 matches in the Spanish LaLiga, we derived activation time series of players throughout matches, akin to the activation time series of neuronal activity. For subsets of players, the measure of information integration Φ was calculated. Our results indicate that player groups exhibit non-zero and variable Φ values over time, suggesting that soccer teams integrate information during matches. Top-performing teams such as FC Barcelona, Real Madrid, and Atlético Madrid displayed significantly higher Φ values compared to other teams. Additionally, the difference in average Φ between competing teams correlated positively with relative shots on goal, implying that higher information integration is associated with better performance. Positive values of Φ suggest a “latent consciousness” at the group level, but because it is (far) lower than the level of consciousness of individual players, it is not realized. This approach offers insights into human collective behavior and practical implications for coaching strategies. This study also demonstrates that information integration can be quantified for complex systems generally, including animal collectives and larger human social structures, and in doing so, information integration can be thought of as a general measure of complexity.