The Naked Truth about Treating Behavior Problems
Institute’s Note: Our guest commentator, Tyler Nighbor, is a doctoral student in behavior analysis at West Virginia University. He earned his M.A. degree in behavior analysis from the University of the Pacific and is a Board Certified Behavior Analyst.
I once worked with an adult client who often disrobed in the common areas of the group home where she lived. You don’t have to have much imagination to see how this could be a problem, both for herself at many levels and for others around her. The question was what to do. The staff’s approach was “planned ignoring,” on the assumption that her disrobing had produced attention in the past, and that by not delivering attention, the behavior would extinguish (go away).
The staff's “solution” raises several issues. First and foremost is an ethical one. Addressing disrobing with a plan based on anecdotal assumptions does not keep the patient safe from harm, which is the first requirement of treatment. From the perspective of seeking a substantive solution, the staff’s unsubstantiated assumptions as to why such behavior was occurring needed to be replaced by an analysis grounded in science.
Not only did the ignoring intervention not work, but it actually increased the client’s frequency of disrobing. It’s easy to say in a case like this that the treatment should have worked except for an unresponsive client. (Blaming the client for failing to “behave properly” often happens when treatments don’t work.)
Ethical issues aside, we can ask why “planned ignoring” didn’t work. Extinction (ignoring, in this case) is supposed to eliminate behavior, but it didn’t here. No one bothered to assess whether attention from others really was what was maintaining the disrobing. “Planned ignoring” was based on someone’s intuition that it was a good treatment for such behavior. It wasn’t based on an examination of why the behavior was occurring (functional analysis). Functional analyses involve manipulating environmental events (e.g., demands, attention) systematically to discover the causes of the behavior. For example, to assess aggression, the therapist might ignore the child until the child hits the therapist, and contingent on hitting, the therapist attends to the child in the form of a verbal reprimand (e.g., “stop doing that!”) to determine whether attention reinforces hitting. If attention is indeed responsible for the problematic aggression, then (and only then) can an appropriate treatment can be designed.
Despite thirty years of success with experimental functional analyses, structural accounts of behavior (descriptive assessments) are still widely used today as the basis for treatment in practice. Recently, Oliver, Pratt, and Normand (in press) surveyed 682 Board Certified Behavior Analysts and found that 94% reported they always used descriptive assessments! In such assessments, no effort is made to determine what is maintaining the behavior before beginning an intervention; rather, observing and recording behavior provides a descriptive account of what is happening. This type of assessment is not only frequently used with behavior problems of children, but also in other behavior-analytic work, such as organizational behavior management in the workplace.
Although descriptive analyses are prevalent, in too many instances descriptive accounts fail to provide the needed information for changing problem behavior, namely, in failing to identify functional relations. Beyond the ethical issues noted above, there were several problems with this structural (“ignoring”) treatment for disrobing. First, an observed common antecedent to her disrobing was staff attention, which was baffling because one would expect that staff had probably been ignoring her, which would alter the value of staff attention and make it likely that she would make responses that had previously produced staff attention. Second, her disrobing, could be indicative of what is called an extinction burst, or increased responding characteristically following the discontinuation of a reinforcer (i.e., they had stopped attending to her disrobing recently). It also could be that staff (and perhaps other residents) hadn’t consistently ignored her disrobing and she was getting attention intermittently following disrobing. Or it could be that the wrong reinforcer had been identified.
We subsequently conducted a functional analysis in which we tested staff attention and escape from demands as potential reinforcers for her disrobing. As expected, the client did not disrobe when provided attention contingent on disrobing, but consistently disrobed during the conditions where disrobing resulted in escape from demands (such as “clean up your room”). Escape from staff demands (and staff leaving her alone) reinforced disrobing, not staff attention.
These results perfectly illustrate where structural accounts of behavior fail, and additionally, where adhering to treatments based on structural accounts of behavior may have counter-therapeutic outcomes. In this case, ignoring the disrobing actually provided temporary escape from staff demands, and instead of planned ignoring reducing instances of disrobing, it increased instances of disrobing. Pairing positive events while she engaged in small steps toward cleaning her room while fully clothed was a place to start. Attending to her while she was engaged in activities she liked, fully clothed, pairing our attention with good things, small steps at a time, was another.
Although we behavior analysts pride ourselves on emphasizing function-based treatments, structural accounts of behavior are still alive and well in daily practice, despite their relative ineffectiveness in treating behavioral problems. They are alive even in such stigmatizing and distressing actions for the patient as described above. It was not a matter of if she cared about being naked; it was a matter of what is appropriate for those who are caregivers over those most in need of compassionate and wise treatment. A quick experimental functional analysis can provide useful information for treatment and save time—and dignity in the long run. It’s the culturally and ethically right thing to do. Purely structural assessments are problematic because the data either are quite subjective with limited samples or take a lot of time to collect, don’t reveal the contingent causes of the behavior, and in cases like this, may result in treatment recommendations that place the patient in harm’s way, waste valuable time, and are counter-therapeutic.
The solution? If you’re a practitioner, when in doubt, conduct the functional assessment to determine what is maintaining problematic behavior. It will save weeks or months of time wasted on ineffective treatment, and certainly wasted time speculating about the person’s ‘reasons’ for doing what she or he is doing. If you’re a consumer of behavioral services, ask questions and ask to see the data, especially if you’re not seeing improvements in your child’s behavior or you and your team are still having the same problems that your organizational change expert came in to fix. Good judgment, along with timely function-based interventions, is required to keep any client safe from personal, social, and larger harm.