Association for Behavior Analysis International

Like any science, a science of behavior seeks to measure its phenomena and explain them. Measurement entails ontological commitments, and from an ontological viewpoint, behavior is process, measured by the time it takes up. Since time is limited, activities must compete with one another for time. The Law of Allocation states that the relative time taken up by an activity equals the activity’s relative competitive weight. Explaining behavioral allocation means finding the determiners of competitive weight. The two basic determiners are induction and covariance. The Law of Induction states that behavior depends on environmental events that affect reproductive success—phylogenetically important events (PIEs). PIEs induce both adjunctive behavior and operant behavior: adjunctive behavior because of phylogenetic contingencies; operant behavior because of ontogenetic contingencies. The law of covariance applies to ontogenetic contingencies. A PIE induces an operant activity when the rate of that PIE covaries with the rate of the operant activity. Otherwise neutral events also induce operant activities and adjunctive activities when such events covary with PIEs. Such inducers have commonly been called discriminative stimuli and conditional stimuli. Induction far exceeds reinforcement in explanatory power. The three laws of allocation, induction, and covariance explain most known behavioral phenomena.

Behavior analysts of all stripes

As a branch of applied mathematics, information theory provides a set of tools and axioms for describing the patterning and structure of streams of events. Such streams are an indispensable form of data for learning theory. While measurements of disorder, or “entropy,” are now widely familiar to academics and laypeople, these only represent the tip of information theory’s iceberg. In concert with probability theory, its tools make possible the measure of how surprising an event is (contingent on some probabilistic expectation), how interrelated streams of events are, and how costly it is to translate between different sets of expectations. In this tutorial, I will give an overview of a practical toolkit of information-theoretic measures that can be used to make normative predictions about experimental designs. I will also introduce a strategy for how to measure the degree of correspondence between a model and data in a way that allows entirely different classes of statistical model to be compared with one another. Throughout, my emphasis will be on making these calculations practical for working experimentalists, so that these tools can be put to work in service of advancing learning theory.

Graduate students, experimental psychologists, and quantitative behavior analysts

Behavior analysts often rely on aggregate measures of behavior to conduct functional analyses and assess the efficacy of their treatments. These aggregate measures, nonetheless, obscure informative features of behavior. A more fine-grained analysis of how motivated behavior unfolds over time, for instance, suggests that, even in a relatively constant environment, motivated behavior is organized in bouts of engagement. Although this is likely in part an artifact of how behavior is measured, it reveals that (a) motivational, cognitive, and sensorimotor processes that underlie observable actions are, in principle, separable, and (b) the most important functional impact of reinforcement is not on conditioned responses or operants, but on behavioral states that give rise to those responses. This tutorial will discuss key insights on behavior that emerge from its temporal organization, the relation of this organization to theories of motivated behavior, and the analytical and procedural approaches to the estimation of organizational parameters.

Board certified behavior analysts and graduate students.