Reading M3 – Labeling behavior
You really can’t talk about behavior without labeling the piece of it under discussion. Behavior is such a broad, varied, and pervasive phenomenon that we have no choice but to get specific about the particular instances of interest. So, whether it’s a casual conversation or a professional presentation, we identify the features of behavior at issue with a label, usually no more than a word or two. These summary labels tend to crudely point to an operant or response class, which is typically the most important thing to distinguish.
Informally, these labels seem to work well enough in daily discourse. If there is any confusion for the listener, it’s easy enough to add some clarification. This is sufficient for informal communication because speakers and listeners tend to share the same verbal community. When a parent talks about their child “having a melt-down” at the grocery store, other parents – not to mention everyone who has been a witness to such episodes – have a pretty good idea of the scene. Colloquial labels such as this are endlessly varied, but this shared verbal community is what makes them more or less effective.
Professional communication requires more clarity, however. “Having a melt-down” isn’t sufficiently specific for researchers or practitioners. They need to know much more about what behavior is included or excluded. Researchers are searching for reliable and general relationships between behavior and features of the environment, and their findings are limited by the particulars. Experimental evidence that a certain procedure will yield certain behavioral outcomes is not likely to hold for just any behavior. Practitioners are searching for procedures that will change particular behaviors under particular conditions. The details of the target behavior are necessary to guide selection of treatment procedures.
We can’t get there from here
In fact, behavior analysis has some rules about how we identify the aspects of behavior under discussion. These rules are not as specific as those in other fields, however. Take biology, for example. The technical labels that subdivide their subject matter are quite detailed. Instead of referring to “vultures,” for instance, biologists refer to a specific grouping of birds, such as the families Aegypiidae and Cathartidae, or even to a specific genus and species. Among biologists, these labels have very narrow definitions, and there is little chance of misunderstanding what kinds of birds are being referenced.
The science and technology of behavior has no such taxonomic vocabulary of behavior, nor is it even possible to come up with a technical label for every possible behavior. This limitation stems from the fact that each response class is unique in important ways from one organism to another. Furthermore, because behavior necessarily adapts to changing environmental circumstances, any response class is always changing for each individual, even if only in subtle features. Together, these characteristics mean that there are probably an infinite number of response classes in an organism’s behavior, let alone across all members of a species.
The limited generality of labels
When a researcher or practitioner attaches a shorthand label to a carefully defined target behavior, colleagues can be pretty clear about what the label refers to in that instance. The problem is that when others use the same label, it will refer to somewhat different definitions used for different participants in their studies or clinical projects. In other words, no matter how carefully each professional selects a particular label for his or her situation, that label will often identify target behaviors that differ in important ways. Nevertheless, such labels often take on a life of their own and may be even used to refer to an entire area of research literature, such as self-injurious behavior.
These unavoidable variations in how particular labels are used raise an important question: How well do the findings of different studies focusing on the same labeled behavior hold across all such studies, not to mention practical circumstances? Given that a particular label can accommodate importantly different features as it is used in different projects and interventions, what generality does it allow? In other words, if a researcher publishes a study investigating cooperative play behavior in preschool children, how can other researchers or practitioners know if they will get the same results if they use the same procedures addressing cooperative play behavior in their own circumstances? There is no easy answer to this kind of question. However, the label we attach to particular target behavior definitions used in each experiment or practical intervention can make a difference in how we interpret their results.
What can we do?
What, then, can researchers and practitioners do to communicate clearly about the particular features of behavior they are working with? First, using clearly stated definitions of each target behavior will probably help more than anything else. Defining the behavior functionally so that all responses in the class share common relations with the environment confers a further advantage. Second, when deciding what to call target behaviors, it can help to be aware of how others may interpret or use the label. Calling a behavior “hitting others” is less likely to be misinterpreted than the more general label “aggression.” Similarly, “typing letters” or “calling clients” is probably a better label than “working.” Third, choosing labels that clarify the relation of the behavior to the environment may also encourage narrow interpretations. For example, the term “playing” covers an endless variety of behavior-environment possibilities. In contrast, “playing with blocks” or “playing house with peers” narrows possible interpretations.
Although these suggestions may aid clear communication about behavior, there is no real solution to this challenge. The nature of behavior guarantees endless variety in its forms and relations with the environment, and everyday language labels – no matter how carefully articulated – cannot meet the requirements for scientific specificity. Nevertheless, it is our task as researchers and practitioners to translate our investigations into the language of the culture. As we come to better understand the factors that determine the generality of our research findings, we will be better prepared to refine the way we describe target behaviors.