Statistical Physics of Human Cooperation
| Author | Matjaž Perc et al. |
|---|---|
| Published | 2017 |
| Format | Research paper |
| Original | Read the original |
TL;DR — A 2017 review of how Monte Carlo simulation and phase-transition theory from physics explain when cooperation, punishment and reward win or fail in structured populations.
Written by Matjaž Perc, Jillian Jordan, David Rand, Zhen Wang, Stefano Boccaletti and Attila Szolnoki, with affiliations at the University of Maribor, Yale and the Hungarian Academy of Sciences. The arXiv version is dated May 2017. Its premise is that humans cooperate with unrelated strangers on a scale no other species does, and that the outcomes of many interacting individuals can be studied the way physicists study interacting particles near a phase transition.
The review has two halves. The first covers human experiments: how laboratory games measure prosocial behaviour in groups and pairs, how punishment (by the injured party or by third parties) and rewards enforce it, what changes when the structure of the interaction network changes, and where experiments and models fail to line up. The second half sets out the models. The spatial public goods game serves as the baseline, extended with peer and pool punishment, peer and pool rewards, self-organised incentives, antisocial strategies and tolerance, all studied with Monte Carlo methods.
The results are spatial and dynamical. Peer punishment works through “indirect territorial competition”. Peer rewards can produce cyclic dominance, where strategies chase each other in rock-paper-scissors fashion. Pool punishment produces complex spatiotemporal patterns, while institutional rewards on their own barely work. Antisocial punishment can unravel cooperation enforcement. The paper ends with a summary and an outlook for future research, and the figures show the spatial patterns from simulations.
What it is: A peer-reviewed review article (published in Physics Reports in 2017) by established researchers in the field. It is a synthesis of simulation and model results, with a section on experimental evidence. The phase-transition findings are properties of stylised models, not observations of real societies.
Where this came from
- Will Manidis’s “most underrated PDF” thread, July 2023 — routing record: The Most Underrated PDF on the Internet (July 2023)
48 pages. A copy is archived locally against link rot; the header links the original source.