Behavioral Artistry: Identifying the Characteristics of Effective Therapists in the Applied Behavior Analysis Treatment of Autism

Background: Richard Foxx's article “Twenty-five years of Applied Behavior Analysis: Lessons Learned” summarized insights from decades of research and treatment in the field of ABA, concluding that the effectiveness of therapy may be negatively impacted by limited repertoires among today’s behavioral interventionists. Foxx believes there are important differences in the outcomes of ABA treatment between persons who deliver services strictly in a traditional way (“Behavioral Technologists”) and those who demonstrate a broader set of humanistic/interpersonal behaviors (“Behavioral Artists”) including care, flexibility, optimism, and humor, among others. Our previous research standardized the reliable measurement of Foxx's Behavioral Artistry (BA) traits using Cattell's Sixteen Personality Factors (16PF) Questionnaire, and socially validated BA by demonstrating parents of children with autism significantly preferred descriptions of behaviors aligning with BA characteristics.

Objectives: The purpose of this study was to investigate differences in the quality of ABA treatment for children with autism when delivered by therapists who have higher or lower levels of Behavioral Artistry.

Methods: ABA therapists at a university-based autism center completed the 16PF questionnaire and were assigned an overall percentage of BA traits based on their corresponding 16PF scores. Therapists with the highest (75% or greater) and lowest (25% or lower) BA percentages were observed during multiple 10-minute videotaped DTT and NET therapy sessions. There were 13 therapists in both the High-BA and Low-BA groups. Foxx's characteristic "Likes People" subsumes a majority of the BA traits. Thus, Likes People was operationally defined, including four behavioral components: Pleasant Facial Expression, Positive Tone of Voice, Sustained Gaze (at client), and Body Position/Orientation. For each therapist, partial-interval recording (10-second intervals) was used to determine a percentage of occurrence for each behavioral component across three observations. Additionally, a subjective rating (0-100) of Likes People was scored by data collectors for each observation session. A percentage of Behavioral Technologist behaviors (defined as fidelity of implementation of trained components of DTT and NET therapy) was also scored. Differences by therapist gender, university major, number of months of experience, and type of therapy session were analyzed.

Results: Therapists in the High-BA group scored higher on all objective and subjective ratings of Likes People, and on Behavioral Technology, than those in the Low-BA group. Differences were statistically significantly higher (p=0.05) for Pleasant Facial Expression. Females scored higher than males in both groups. There were significantly fewer ABA majors in the High-BA group. Low-BA therapists had significantly more experience. Therapists in the Low-BA group scored lower on Positive Tone of Voice during NET sessions. IOA for Likes People interval scoring was 94.2% overall.

Conclusions: Therapists with higher levels of interpersonal behaviors associated with Behavioral Artistry were rated qualitatively better in their delivery of ABA treatment for children with autism, suggesting important implications for hiring, training, and supervising effective ABA interventionists. Hypothetically, therapists who are warm, attentive, creative, optimistic, and persevering will generate more positive outcomes for individuals with autism in schools, clinics, and homes. Future research should investigate whether High-BA therapists have better client outcomes and whether BA can be effectively trained.