Basing the treatment of stereotypic and self-injurious behaviors on hypotheses of their causes.
Treat self-injury or stereotypy only after a mini functional analysis, then pick the reward that passed the test.
01Research in Context
What this study did
Three children with intellectual disability kept hitting themselves or rocking for hours. The team first watched each child in four quick test rooms. One room gave toys, one gave attention, one let escape work, and one stayed blank.
They used the test pattern to pick a treatment that fit the cause. A child who hurt herself for escape got help with hard tasks. A child who rocked for toys got toys only when he sat still.
What they found
Each child’s problem dropped fast when the reward matched the test clue. The same plan failed when it ignored the clue. For example, giving toys to the escape child did nothing.
Matched plans cut behavior to near zero in every case. Mismatched plans barely changed the numbers.
How this fits with other research
Lambert et al. (2022) ran the same idea in a busy university clinic. Their big caseload showed the match helps, but some kids still need tweaks. The 1988 paper is the seed; the 2022 paper shows the tree.
Weber et al. (2024) looked at 300+ clinical tests and found most show a clear cause, just like the three kids here. Their numbers back up the early small study.
Huguenin (2000) warned that pills alone rarely fix self-injury. The 1988 study agrees: know the cause first, then pick the fix, drug or not.
Why it matters
Before you write a behavior plan, run a 10-minute functional analysis. Pick the treatment that fed the test result, not the one that sounds nice. This habit turns long-lasting harm into quick silence.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Test one kid in four five-minute conditions today; start the matched reward tomorrow.
02At a glance
03Original abstract
Stereotypic and self-injurious behaviors are common forms of maladaptive responding demonstrated by severely handicapped persons. Various review papers suggest that no single treatment procedure is universally effective. Although there may be many reasons for this finding, one could be that people engage in these behaviors for various reasons, and that procedures that are incompatible with the cause of the behavior are unlikely to be effective. These studies also suggest many hypotheses for the development and maintenance of these behaviors, three of which are the self-stimulation, positive reinforcement, and negative reinforcement hypotheses. The purpose of this paper was to determine whether one of these hypotheses could be matched to the cause of the behavior and used as an effective treatment procedure. We therefore compared one hypothesis with one other for 3 subjects in a three-phase study. During baseline, data were taken in two classrooms for each subject, and a judgement was made about the hypothesis most likely to be related to the cause of the behavior. During the second phase, a treatment based on that hypothesis was used in one classroom, and a treatment based on another hypothesis was used in the second classroom. During the third phase, the treatment that was most effective in the second phase was used in both classrooms. Results showed that a successful treatment program can be developed on an hypothesis of why the behavior occurred during baseline. Results are discussed in terms of supporting the argument that treatment programs should be based on a functional analysis of the behavior in its environmental context.
Journal of applied behavior analysis, 1988 · doi:10.1901/jaba.1988.21-281