The ecological assessment of an autistic child

The behavior of an autistic child was observed in his natural environment. Observations were made in three settings, over approximately 6 months. Sixteen behaviors of the child and six behaviors of adults and peers were recorded. A cluster analysis was performed to identify response classes within the behavioral structure of the child and to see if correlations existed between response and stimulus events. To examine intra- and intersetting changes, Spearman rank correlations and ttests were also computed. This autistic child exhibited a diversity of behavior over time in a given setting and across settings. He persistently ignored other nonautistic children, although he was responsive to adults. Response classes reflected an inverse correlation between the child's selfstimulatory behaviors and his attentiveness to the environment. Elevated levels of adult attention were consistently related to reduced levels in most self-stimulatory behaviors. Two response classes showed inverse relationships between self-stimulatory behaviors. Reducing the frequency of some self-stimulatory behaviors might cause increases in others.This study is based on a dissertation submitted by the first author to The University of Tennessee in partial fulfillment of the requirements for the Ph.D. degree. The research was conducted under the chairmanship of the second author. The authors wish to express their appreciation to Dr. Edward E. Cureton for his help with the statistical analyses.