Problem behaviors in persons with developmental disabilities are disturbing and isolating. These behaviors cause the person with disabilities, their families and other caregivers great stress. They increase the probability of restrictive school and living placements, and they are the leading reason why people remain in state residential facilities; in turn, this means increased costs, estimated at more than $3.5 billion annually.
Today, problem behaviors are viewed as forms of aggression, directed toward self, others, or the environment. We’ve discovered that it’s more effective to treat them as a group of interrelated behaviors rather than to focus on self-injury by itself, for example. Support strategies need to be long-term and intensive. There appears to be a genetic basis for some problem behaviors, although that relationship still requires further research. Functional assessments are providing researchers new insights into the environmental causes of problem behavior, which in turn will provide a basis for developing more effective treatments.
A person behaves in a particular way in response to something positive or negative in the environment. We observe a problem behavior to determine what positive stimuli occur or what negative stimuli are removed after that behavior. Although this type of analysis may sound simple, it’s complex, because of the different patterns of reinforcement and wide range of stimuli involved across individuals. The goal of functional behavior analysis assessments is to identify the type of feedback mechanism maintaining the problem behavior. Reinforcers need to be defined individually and functionally. Functional assessments are often like solving a detective mystery. This assessment approach is successful in identifying environmental or social causes of problem behavior in about 70% of cases.
Although researchers know a great deal about the variables that maintain problem behaviors, we know little about their origins. Current views of their origins are emergence from stereotypy, emergence from health care needs, and response induction.
Stereotypies—behavior like body rocking, or hand waving or flapping—emerge early in typical and atypical development and are maintained by sensory consequences. Recent research indicates that these behaviors take on a social function and children learn to use them as a form of communication.
Once a behavior is troubling enough to require intervention, a caregiver may try to “extinguish” a behavior by ignoring it. This will likely increase the behavior’s frequency and variability. Ignored, a child who had been hand flapping gets more agitated, might hit himself, someone intervenes with social attention, and a new problem behavior is reinforced. The “extinction” process has induced a problem behavior.
Another cause is health issues. People with developmental disabilities have higher incidence rates of many health issues, such as sleep problems, gastrointestinal problems, allergies, and chronic ear infections. Individuals may be unable to communicate when they are in pain. At the same time, when a person is evaluated as having mental retardation, that takes precedence over their having other health conditions, which results in under-identification of health issues. Unaddressed health care needs can establish certain behaviors as functional, and problem behaviors may emerge.
A third cause of problem behavior is response induction. For example, I might have been taught to use a sign to request a cookie. On some occasion I’m in a situation where people are ignoring my request, I become agitated and knock over a drink; to calm me down I’m given a cookie. I’ve just learned it’s more effective to create a fuss to get that cookie than to sign.
Future research at the Vanderbilt Kennedy Center will address several areas. First, we need to identify causes of problem behavior in the 30% of cases where functional assessments are unsuccessful in identifying its purposes and reinforcers. Second, behavioral processes leading to the genesis of behavior problems need to be documented. Third, we should identify genetically based predispositions toward behavior problems. Finally, health, education, and behavior-analytic assessments must be linked to guide comprehensive interventions.
Craig H. Kennedy, Ph.D., is associate professor of special education, Vanderbilt Kennedy Center investigator, and director of the Vanderbilt Kennedy Behavior Analysis Clinic.