8/8 Want to go deeper? Read our paper!

🔗 doi.org/10.1038/s420...

Many thanks to @asteixeira.bsky.social and @anxosanchez.bsky.social.

7/8 And more beautifully: we can use simple mathematical models to describe this type of complex human social behavior.

6/8 Why does this matter? Because if a network is in equilibrium, a single network snapshot can reveal stable dynamical patterns. That validates tools like Exponential Random Graph Models, usual in non-longitudinal settings. Huge for social science!

5/8 We argue this behavior is rooted in the way in which humans allocate their cognitive resources to create functional relationships able to cover their social necessities, and in the competition between the mechanisms that drive the network evolution.

4/8 People constantly form and break ties, leave and join the network, and still all network snapshots are statistically equivalent to one another. This implies that what matters is not who is in the network, but the rules governing how relationships evolve.

3/8 It means that despite all the complexity, the temporal evolution of the network can be described by simple mathematical models and, more importantly, its behavior can be predicted with very high accuracy.

2/8 They are not! In our last work we analyze the empirical evolution of a social network, and we provide evidence that relationship networks can exhibit a behavior akin to equilibrium dynamics in the statistical physics sense. What does this mean?

Always a pleasure to work together!

In this account we fully support any type of spherical cow