Research shows self-selected treatment, punishment reduces destructive behavior

Jun 21 2005

Punishment is often a necessary ingredient in any method aimed at making aggressive special-needs children less destructive, according to new research in the Journal of Applied Behavior Analysis.

However, punishment alone is not an effective method, said University of Kansas researcher Gregory Hanley, who published the research along with colleagues at Johns Hopkins University. Combining treatment methods selected by the children themselves with a nonviolent punishment, such as mild restraint, is more effective than relying on a single method or methods that have no punishment component, he said.

Hanley, an assistant professor of applied behavioral science, and his fellow researchers focused on their approaches on reducing the destructive behavior of a 5-year-old boy and an 8-year-old girl. The children, Jay and Betty, were in hospital units because of self-injuries, including bruises and bleeding. Both were diagnosed as having mental retardation and other conditions. Besides hurting themselves, they were aggressive with parents and other children.

With such children, clinicians sometimes train the children to communicate their needs rather than act out aggressively, says Hanley.

“ Because Betty and Jay's problem behavior was most likely to occur when parents attended to the behavior, we taught Betty to say things like 'play with me' or 'excuse me' to get parental attention,” Hanley said. “Jay was nonverbal, so he was taught to communicate by handing the researcher a card containing the written word 'play.'”

Another technique used in communication training involves “extinction,” or ignoring a problem behavior until it stops.

Both these strategies may decrease aggression, but they have mixed results, Hanley said. So he and his colleagues added one more to the mix: a punishment component.

“ This doesn't mean slapping or hitting children,” Hanley said.

The “punishment” for Jay was having his hands held in his lap by a researcher for 30 seconds. For Betty, it involved both restraining the hands and covering the eyes.

“ We saw nice reductions in the problem behavior, and the children continued to communicate,” Hanley said, “but most clinicians stay away from any mention of 'punishment.' A strong and vocal minority in our profession finds it too controversial.”

Yet it's equally controversial, Hanley said, not to use interventions that work.

The most important part of the study, Hanley said, was the researchers' decision to let the kids select their treatments. If they pressed a switch of one color, they got brief exposure to both communication training and extinction treatment. If they pressed a switch of a second color, they got communication training and punishment. Pressing a third resulted in punishment only. The switch pushed most often indicated the child's preference for treatment.

“ Both children preferred the communication training and punishment in combination,” said Hanley. Removing the punishment option would deny the child both the most effective treatment and the child's preferred treatment for reducing problem behavior, Hanley said.

He said that the field of applied behavioral analysis specializes in studies in which lots of data are gathered on few participants, not a little data on many.

“ Either other people come along later and replicate these studies,” he says, “or, alternatively, someone chooses a more typical group design and establishes the applicability of the finding that way.”

Hanley is KU faculty supervisor of the Educare preschool program at the Edna A. Hill Child Development Center. Besides studying problem-behavior prevention, he investigates young children's preferences for educational formats.