The Physics of Groups: Spoof or Proof

Group dynamics is remarkably interdisciplinary.  Everyone, from political scientists who investigate policy making/planning groups (such as the famous JFK executive committee) to economists investigating choice in multi-player games, have joined social psychologists in the pursuit of knowledge about groups. We now have, for example, team science emerging as a new way of examining how task-focused groups with stable memberships work and experts in social network analysis applying the methods with renewed enthusiasm to groups (I say “renewed” since, after all, Moreno was pretty busy with SNA back in his day).

But I wasn’t ready for the physics of groups. Yes, the physics of groups. I wanted to see who had cited the classic 1969 Moscovici, Lage, and Naffrechoux paper, which found evidence of delayed influence of minority in a “reverse-Asch” situation. Cited by 295 other papers, I had a look at those published more recently–after 2008.  And the title that stood out on the second page of the list: “Social consensus through the influence of committed minorities” by Xie, Sreenivasan, Korniss, Zhang, Lim, and Szymanski published in Physical Review E, a journal of the American Physical Society. Physicists, it turns out, have decided that human groups are interesting systems, and that their theories of structure, change, and dynamics apply to collections of people as much as they do to collections of rocks, planets, or microscopic particles. And, because I have not kept up my subscription to Physica, International Journal of Modern Physics, and Review of Modern Physics, I was unaware of the strides being made by the real scientists. Mason, Conrey, and Smith (2007) called out “heads up” in their analysis of social influence as a multidirectional flow system of dynamic networks, but even their careful review of four separate models of such systems did not prepare me for the degree of theoretical interest exhibited by physicists in group processes.

The Xie et al.  paper does not present data that they collected examining opinion shifts. Rather, the work is theoretical, applying principles based on physical systems to interpersonal ones. Specifically, they apply the “two-opinion variant of the naming game (NG)” and test its fit using Erdos-Renyi random graphs. Apparently, earlier investigations by physicists of social influence (Professor Galam is mentioned frequently) found this model described hypothetical changes in attitudes in social networks well, but their work focused on adding “rigidity” to the model: the individual, who they label the “zealot” who is immune from influence.  Their abstract sounds like social psychology (“We show how the prevailing majority opinion in a population can be rapidly reversed by a small fraction of randomly distributed committed agents who consistently proselytize the opposing opinion and are immune to influence.”), even if the paper reads like Bibb Latane’s 1981 American Psychologist social impact theory paper on steroids.

Intrigued, I cast the Google Scholar net more widely, and uncovered an entire subfield of physics devoted to group-level processes. Fortunately, a very detailed paper by Castellano, Fortunato, and Lorento (2009)  provided a very comprehensive review of attempts to apply method and theory from physics to “collective phenomena emerging from the interactions of individuals as elementary unites in social structures” (p. 591). They examine such processes as “the dynamics of opinions,” “cultural dissemination,” “crowd dynamics,” “the emergence of hierarchies,” “social spreading phenomena,” and “what is becoming established as ‘human dynamics’. (p. 634).  I’ll quote a few sections, to give a flavor of the work:

• p. 594:  Generally speaking, the drive toward order is provided by the tendency of interacting agents to become  more alike. This effect is often termed “social influence” in the social science literature (Festinger et al., 1950) and can be seen as a counterpart of ferromagnetic interaction in magnets. Couplings of antiferromagnetic type, i.e., pushing people to adopt a state different from the state of their neighbors, are also important in some cases and will be considered.

• p. 596: Beyond its relevance as a physics model, the Ising ferromagnet can be seen as a simple model for opinion dynamics, with agents influenced by the state of the majority of their interacting partners. Consider a collection of N spins (agents) si that can assume two values ±1. Each spin is energetically pushed to be aligned with its nearest neighbors. The total energy is  H = – 1/2 sum (i, j) Si/Sj, where the sum runs on the pairs of nearest-neighbors spins.

• p. 598:  The first opinion dynamics designed by a physicist was a model proposed by Weidlich
  1971. The model is based on the probabilistic framework of sociodynamics, discussed in Sec. II.C. Later on, the Ising model made its first appearance in opinion dynamics
  Galam et al., 1982; Galam and Moscovici, 1991.

• p. 624:  To study the collective motion of large groups of organisms, the concept of self-propelled particles
  SPP has been introduced
  Vicsek et al., 1995; Czirók and Vicsek, 1999, 2000. SPP are particles driven by an intrinsic force, produced by an energy depot that is internal to the particles, as occurs in real organisms.

• p. 629: Dominance relationships seem to be determined by the outcome of fights between individuals. Laboratory experiments on various species hint at the existence of a positive feedback mechanism
  Hogeweg and Hesper, 1983; Chase et al., 1994; Theraulaz et al., 1995, according to which individuals who won more fights have an enhanced probability to win future fights as compared to those who were less successful
  winner or loser effects.

I know.  It seems like a spoof.  They mention in their analysis this caveat: “It is worth remarking that, even if we have done our best to mention relevant social science literature and highlight connections to it, the main focus of this work remains a description of the statistical physics approach to social dynamics.”  To me, that would be like me saying something like “we are going to discuss interplanetary motion and celestial dynamics. We will do our best to mention the work of astronomers, but we will probably just talk about conclusions we drew by standing outside in the back yard on a dark night. “