Association for Behavior Analysis International

Behavior analysis has established its subject matter as operating on the individual, whole-organismal level and consisting of environment-behavior relations that form analytic units selected and shaped by contingencies. However, behavior is situated firmly in biological strata whose physiological processes constitute component parts that are active within the behavior of the whole organism. Conversely, contingencies that control behavior are often social in that the behavior of other organism(s) acts as the controlling environment, and the resulting social systems can serve to determine the very contingencies that then shape individual repertoires. Behavior analysis have often disagreed about how to approach role of physiological units in behavioral phenomena as well as the investigation into potential group-level analytic units arising from multi-player scenarios. Criticism about culture-behavioral concepts suggest that principles of cultural relations can be expressed via operant concepts, but that the opposite is not true. In this symposium, we will use conceptual tools and empirical data to explore potentially fruitful ways to investigate the connections between levels of analysis and the effort to determine when new analytical units might be in play.

Behavior Analysis has rightfully resisted efforts to inappropriately reduce behavioral phenomena to physiological processes, though there can be no doubt that these processes are seamlessly integrated with (and sometimes constitute component parts of) behavior. It is also clear that manipulating them in some cases impacts behavior. At the same time, many Behavior Analysts argue that social, “cultural”, and human-systems analyses can best be entirely accounted for via component operant processes. In this talk, we will use the approach described by Bechtel and Hamilton (2007) to discuss how to disambiguate levels of analysis according to principles governing their respective functional relations. We will then talk about some ideas on how to bridge concepts between these three levels without inappropriate reduction. We will then discuss the difficulties inherent in postulating new levels of analysis (especially group-level units) as they have been suggested both in evolutionary biology and in culturo-behavioral science, and some empirical approaches to testing such ideas. Finally, we will present and reflect on potential homologies and analogies in principles across these levels that seem to permeate the natural sciences.

The Darwinian perspective on natural selection place humans as part of a continuum in evolutionary history. Within species, the ability to learn from environmental changes during an individual life span has great adaptive value in the race for survival, thus learning is inseparable from natural selection. In addition to natural selection and the selection of behavior, cultures can also be studied from a selectionist perspective. To understand selection at the level of cultures, it is paramount to identify how environmental events are functionally related to i) social behavior, ii) cooperation and iii) the coordination of responses of two or more individuals in interlocking behavioral contingencies. This presentation discusses a conceptual framework to understand the evolutionary continuum from natural selection to the selection of cultures as a co-evolutionary process. The presentation expands on the differentiation between the selection of cultures, as settings of contingencies, and cultural selection as a within-group selection of individuals’ behavior (Couto and Sandaker, 2016) and discusses the role of behavior analysis from a broader evolutionary perspective.

We discuss the need to distinguish units of analysis that describe individual behavior and units describing social behavior. We highlight two lines of evidence from our lab that suggest important differences from simple operant behavior to social behavior. The first line of evidence came from research about inequity aversion, the preference for no gain over an unfair distribution. Our data shows that previous interaction with cooperative or uncooperative can alter aversion to disadvantageous inequity in a within-subject and between-subject design. The same participant may be or may be not averse to disadvantageous inequity due to recent past cooperative history. Another source of evidence came from experimental study of coordination. Our strategy used rates of individual responding maintained by ratio or interval schedules of reinforcement as a baseline to evaluate the effect of consequences contingent on coordinated behavior of participants in groups of three. Results showed that contingency on coordinated behaviors resulted in an increased rate of coordination along with a decreased rate of responding to individual schedule. Both lines of investigation suggest that we need to distinguish units of analysis that describe individual behavior and units describing social behavior, although these units may occur simultaneously