Can Behavior Analysis Benefit from Reductionism?

Abstract

Our concern is whether or not the arbitrary philosophic value of reductionism provides immediate usefulness for behavior analysis. In order to work with stimuli and behavior relations efficiently, one need be concerned minimally with underlying physical, chemical, or neuroscience correlates, and primarily focus on the observable antecedents, behaviors and consequences. Selecting the appropriate level of analysis for a research question is paramount and the reductionist approach adds little value to the parsimonious methods of behavior analysis that operate at the level at which the behavior occurs. When moving between levels, the probability of introducing an error increases as the gap between the input and output variables increase. It is therefore suggested that in order to minimize the possibility of making these errors, analyses should be performed at the same level at which the event is being observed. Summary variables that operate at different levels to the behaviors should be used sparingly and, when introduced, should always emphasize both the environment and the behavior. Although reduction to neurophysiological events may be possible, and even useful, this may be viewed as at qualitatively different levels of analysis that, while providing interesting data, would usually not be critical for the behavior-analytic approach.

Key words:

• reductionism
• molecular
• molar
• levels of analysis
• summary variables
• neurological activity