Tipping points among social learners: Tools from varied disciplines

Michael J O’Brien,R Alexander Bentley

doi:10.1093/czoolo/58.2.298

Michael J O’Brien, R Alexander Bentley

Open Access

https://doi.org/10.1093/czoolo/58.2.298

Copy DOI

Journal: Current Zoology	Publication Date: Apr 1, 2012
Citations: 39	License type: cc-by

Affiliation: University of Missouri, University of Bristol

Abstract

Abstract There is a long and rich tradition in the social sciences of using models of collective behavior in animals as jumping-off points for the study of human behavior, including collective human behavior. Here, we come at the problem in a slightly different fashion. We ask whether models of collective human behavior have anything to offer those who study animal behavior. Our brief example of tipping points, a model first developed in the physical sciences and later used in the social sciences, suggests that the analysis of human collective behavior does indeed have considerable to offer [Current Zoology 58 (2): 298–306, 2012].

Highlights

Animal behavior is one arena where social learning, defined as learning by observing or interacting with others (Heyes, 1994) as opposed to learning individually, is studied in its most uncomplicated form (e.g., Laland, 2004; Couzin et al, 2005)
The record of the past is our best view into the future, and in the science of behavior we are often presented with a time series of events, from abundances of fossil species to climatic events or the incidence of certain animal behavior observed in the lab or the field (e.g., Ormerod, 1998, 2006; Michie, 2006; Long, 2009; Willis et al, 2009)
This space of potential cascades, which is dependent on interconnections versus individual thresholds to change, is the same as that yielded by a model of information cascades across a network of social agents

Summary

Scaling Up

Care needs to be taken in scaling up observations on flocks of birds or schools of fish to human groups, they provide potentially profound hypotheses about collective decisions in human groups, in that the most persuasive or influential property may not necessarily be logical superiority in skill or status—increased working memory, ability to delay gratification, and the like—but rather the persistence of the message, a strategy of which opinion leaders and marketers make explicit use. This effect may arise from the network rather than from a special individual. Even simple forms of animal social learning give a group higherorder computational capacities to respond to its environment (Couzin, 2007)

Scaling Down

Tipping Points

Time Series

System Interconnections and Thresholds

Distributions

Applications to Current Zoology