Thomas Robiglio (IT-U University, Linz)

The interplay between causal mechanisms and emerging collective behaviors is a central aspect of understanding, controlling, and predicting complex networked systems.

Existing methods to study each of the two facets mostly adopt lower-order descriptions: pairwise network representations for mechanisms, and low-order information-theoretic metrics for behaviors. Despite their success, these low-order methods often fail to fully capture the intricate nuances inherent to many complex systems, thus beyond-pairwise methods are being developed: higher-order network representations and higher-order behavorial metrics.

As both low-order and higher-order mechanisms can determine the observation of both low and higher-order behaviors, the connection between behavioral observables and microscopic mechanisms in systems with pairwise and group interactions is not trivial; a systematic investigation of this complex relationship across different orders of interactions is needed.

In this talk, I will present recent work exploring the link between higher-order mechanisms and higher-order behavioral observables in two representative models with group interactions: a simplicial Ising model and a social contagion model. Our findings reveal that group (higher-order) interactions give rise to emergent synergistic (higher-order) behaviors in both systems.

Finally, I will argue that fully disentangling behaviors and mechanisms in complex systems—especially those with group interactions—requires principled approaches based on generative models and statistical inference.

REFERENCES:

• Robiglio, T., Neri, M., Coppes, D., Agostinelli, C., Battiston, F., Lucas, M., & Petri, G. (2024). Synergistic signatures of group mechanisms in higher-order systems. arXiv preprint arXiv:2401.11588.
• Peel, L., Peixoto, T. P., & De Domenico, M. (2022). Statistical inference links data and theory in network science. Nature Communications, 13(1), 6794.