COMPUTATIONAL SOCIAL SCIENCE

Department of Computational Social Science Seminar Abstract

Friday, April 13, 3:00 p.m.
Center for Social Complexity Suite
Third Floor, Research Hall

Re-thinking Kauffman's NK Fitness Landscape: Going from Artifact & Groupthink to Weak-tie Effects

Bill McKelvey
Professor of Strategic Organizing and Complexity Science
UCLA Anderson School of Management


Stuart Kauffman’s NK model defines fitness as the average of all the agents (genes, employees, units) having frequent interconnections. Because of the formula he uses, NK-model results appear to be artifacts preordained by the code rather than by theory-based experiments—a fundamentally egregious problem, if true. Furthermore, as K (# of contacts among agents) increases, all agents become interconnected—similar to Janis’s groupthink—and fitness trends toward the mean and adaptive failure; what Kauffman calls “complexity catastrophe.” Inadvertantly, the 35+ studies applying it to management presume groupthink predominates—a misrepresentation of reality in many firms. The biological basis of the NK model leads to the two foregoing outcomes, artifact and groupthink. But employees aren’t genes. Recent findings in the strong-tie/weak-tie literature suggest the notion of a tipping point. The NK model is changed in just one way so that it produces results showing the transition from strong- to weak-ties: F, the frequency of contacts per year, is added; it varies from once a week to once a year. Our results clearly show that the F variable negates the standard NK effect of emergent complexity catastrophe. And, yes, there is a tipping point.