Thomas C. Schelling and James M. Sakoda: The Intellectual, Technical, and Social History of a Model

Abstract

Abstract:The Journal of Mathematical Sociology (JMS) started in 1971. The second issue contained its most cited article: Thomas C. Schelling, "Dynamic Models of Segregation". In that article, Schelling presented a family of models, one of which became a canonical model. To date it is called the Schelling model -an eponym that affixes the inventor's name to the invention, one of the highest forms of scientific recognition. In the very first issue of JMS, James Minoru Sakoda published an article entitled "The Checkerboard Model of Social Interaction". Sakoda's article more or less went unrecognized. Yet, a careful comparison demonstrates that in a certain sense the Schelling model is just an instance of Sakoda's model. A precursor of that model was already part of Sakoda's 1949 dissertation submitted to the University of California at Berkeley. A substantial amount of evidence indicates that in the 1970s Sakoda was well known and recognized as a computational social scientist, whereas Schelling was an unknown in the field. A generation later, the pattern of recognition almost completely reversed: Sakoda had become the unknown, while Schelling was the well-known inventor of the pioneering Schelling model. This article explains this puzzling pattern of recognition. Technical and social factors play a decisive role. Some contrafactual historical reflection suggests that the final result was not inevitable.

Individual attitude change and societal dynamics: Computational experiments with psychological theories

“…With agent-based modeling we can resolve the social psychological challenge of going from individual information processing to the aggregated level of groups and societies (Hegselmann 2017;Helbing & Balietti, 2012;Jackson, Rand, Lewis, Norton, & Gray, 2017;Smith & Conrey, 2007;Squazzoni 2012). In an agent-based model, we construct a virtual society, where computer-simulated individuals ( agents ) repeatedly communicate with each other.…”

Social reinforcement with weighted interactions

“…In this framework, agents assign more weight to opinions that already conform to their beliefs, while neglecting to gather, or discounting, evidence that would tell against it [13, p.175]. While [26] captures this bias by allowing agents to regard only those peers who are suciently close in their opinion, we follow [27] by allowing agents being both close and far in their intrinsic preference to aect their peers' decision but with dierent weights. For instance, observing someone having the same preference adopting certain behavior is more persuasive than observing someone with a very distant preference.…”

On Entropy Dynamics for Active “Living” Particles

“…The dynamics of the entropy follows that of the distributions f ij and describes the specific features of the trend to equilibrium when it exists. Particularly important is analysis of segregation phenomena which might be described by Entropy functions [37]. The study of segregation phenomena is also very important in the study of collective learning dynamics [4,38], as it can denote the onset of clusters of extreme levels of education, namely very high and very low levels.…”