Selections on the Empirical and Theoretical Investigations of Behavioral Variability: An Introduction to the Special Issue

Abstract

A. Blaisdell & W. D. Stahlman Special Issue Introduction Selections on the Empirical and Theoretical Investigations of Behavioral Variability: An Introduction to the Special Issue W. David Stahlman University of Mary Washington, U.S.A. Aaron P. Blaisdell University of California, Los Angeles, U.S.A. This special issue of the International Journal of Comparative Psychology is dedicated in memory of Seth Roberts who passed away before its publication. The ideas contained herein are in part the result of, and greatly inspired by, his innovative and important work on behavioral variability . “Under carefully controlled experimental circumstances, an animal will behave as it damned well pleases.” -The Harvard Law of Animal Behavior Though the Harvard Law is much a tongue-in-cheek, wry statement for researchers to evoke when their investigations do not go as planned, it holds a critical reminder for those who would set out to forecast the behavior of organisms. Namely, it suggests an important role for the unpredictability inherent to behavior. The Law subtly indicates that this unpredictability is neither due to measurement error, nor due to an incomplete understanding of the factors that control behavior, nor owing to improper control of the relevant observational parameters. Instead, due to intrinsic stochasticity, behavior retains a degree of necessary unpredictability. In other words, while researchers can readily make accurate predictions regarding typical behavior when collecting many data, they must necessarily struggle to accurately predict a single organism’s behavior at a given instant in time. Behavior appears to be probabilistic in nature, not determined. The experimental investigation of behavioral variability as an important phenomenon, rather than a nuisance, has really only entered the scientific literature in very recent decades. Certainly, there were investigations that predate the so-called cognitive revolution (e.g., Antonitis, 1951; Dashiell, 1930); however, the systematic investigation of behavior has traditionally focused on the strength of discrete response classes, rather than the examination of the control of the form of behavior. Much of this focus may be attributable to the giant intellectual forebears in the studies of classical conditioning and instrumental learning. Thorndike’s (1911) law of effect focused on the lawfulness of how Stimulus-Response (S-R) learning follows from Correspondence concerning this article should be addressed to W. David Stahlman, UMW Department of Psychology, 1301 College Ave., Fredericksburg, VA 22401. (Email: wdstahlm@umw.edu)