Human and Non-Human Performance in the Experimental Analysis of Behavior

Abstract

Laboratory research with humans is common in Behavior Analysis, although the control of variables is different from the control possible with non-human animals. From the 1980s onwards, when research with humans increased in frequency in Behavior Analysis, debates arose about the necessity and viability of this research in producing knowledge on the basic principles of behavior. Specifically, discrepancies in research results that employed humans and non-humans to examine the fundamentals of behavior opened up different investigative paths. On the one hand, these discrepancies could be explained by quantitative differences, i.e., the quantity and imprecise control of variables affecting behavior, indicating a need for refinement and standardization of methods; on the other hand, the discrepancies could be explained by qualitative differences between species, highlighting a discontinuity of behavioral processes that could not be overcome by procedural controls, thus guiding the opening of new paradigms for human research. As a case study of a research agenda directly affected by such debates, the procedures and results reported in decades of experimental investigations into schedules of reinforcement were examined. Discrepancies in performance between humans and
non-humans could be attributed to differences in the degree of control of variables, not to distinctions in nature between species. The analysis of these data is in line with calls for greater standardization of basic research procedures with humans, similar to those with non-humans, as a way of preserving their importance for Behavior Analysis and not hastily discarding or being silent on the assumption of the continuity of behavioral processes.