Treatment of severe aggression in an adolescent with autism: Non-contingent Reinforcement and Functional Communication Training.

Abstract

The following article describes a two-part intervention addressing severe aggressive behavior exhibited by Richard, an 18-year-old male with autism enrolled in a community transition program. The two components of the intervention included: Noncontingent Reinforcement and Functional Communication Training. The intervention initially resulted in a dramatic reduction in aggression. However, a sharp increase was noted after 5 weeks. Additional analysis indicated that significant treatment drift had occurred in the implementation of Noncontingent Reinforcement. Staff members were re-trained in this intervention, and aggressions again dropped significantly. The results of the current case study suggest that the use of Noncontingent Reinforcement in coordination with Functional Communication Training had powerful effects in reducing Richard's aggressive behavior. However, these data also highlight the need for an ongoing assessment of treatment integrity and procedural reliability to ensure maximum and long-term success. Descriptors: Noncontingent Reinforcement; Autism; Aggression; Functional Communication Training. ********** Noncontingent reinforcement can be defined as the delivery of an aberrant behavior's known reinforcer on a response-independent basis (Carr, Coriaty, Wilder, Gaunt, Dozier, Britton, Avina, & Reed, 2000). Non-contingent reinforcement has its origins in basic research, but has become increasingly used in applied interventions. The effects of non-contingent reinforcement have been defined by basic research. Zeiler (1968) examined the difference in variable time vs. fixed time schedules in pigeons that had previously acquired an operant response, and found that FT schedules yielded consistent decreases in the rate of responding (while VT schedules had more erratic effects.) Rescorla & Skucy (1969) compared extinction to response-independent delivery of food (VT schedule). These researchers found that extinction decreased responses more effectively than response-independent food delivery. Lachter, Cole, & Schoenfeld (1971) found that dense FT schedules produced greater reductions in response than lean FT schedules in pigeons. These and other studies (e. g., Alleman & Zeiler, 1974; Dickinson & Charnock, 1985; Halliday & Boakes, 1971; Hammond, 1980; Hammond & Weinberg, 1984) support that changing from a response-dependent system of reinforcement to a response-independent system of reinforcement consistently produces a response reduction. In applied research, NCR has been used as a control for reinforcement procedures (O'Neill & Morris, 1979; Hart, Reynolds, Baer, Brawley, & Harris, 1968). In the Hart et al. study (1968), a five-year-old girl was given either contingent adult attention or non-contingent adult attention. An increase in cooperative play occurred only in the contingent attention condition. This lends support to the hypothesis that the contingency between the response and the reinforcer was responsible for the behavioral effect. Horner (1980) examined differential reinforcement procedures and the impact of these procedures on five individuals with mental retardation. In the NCR condition, the same enriched environment was provided, but social reinforcement was delivered independent of adaptive and appropriate behavior. The differential reinforcement procedures were responsible for the behavioral change. In the last decade, NCR has been used as part of functional analyses, to help assess the function of challenging behaviors. In this condition within a functional analysis, the individual receives frequent attention on a FI schedule (e.g., 30 seconds). Free access to preferred items and activities is provided, and no demands are placed on the individual. Iwata, Dorsey, Slifer, Bauman, & Richman (1994) examined the effects of NCR as a control procedure in functional analyses. NCR produced a reduction in SIB in 127 of 152 (86%) functional analyses. …