Printer Friendly

Nonaversive treatment of high-rate disruption: child and provider effects.

Nonaversive Treatment of High-Rate Disruption: Child and Provider Effects Numerous position papers criticizing aversive interventions for persons with severe disabilities have been published (e.g., La Vigna & Donnellan, 1986). Not surprisingly, these papers have generated countercriticisms (e.g., Mulick & Linscheid, 1988). Unfortunately for many of these people, the debate may distract mental retardation professions from the primary issue: the need for effective treatment (Van Houten et al., 1988). The crucial questions concern the effects of nonaversive treatments; useful answers will be produced by data-based research, not by arguments.

An important question is whether strictly nonaversive treatment produces clinically significant changes in high-rate problem behaviors. The most frequently studied nonaversive treatment involves differential reinforcement of other and/or incompatible behavior (DRO/DRI). Research on DRO/DRI used on high-rate behavior problems in children with severe disabilities has produced mixed results. Some comparative experiments have suggested that these methods are substantially less effective for behavior problems than are aversive techniques such as overcorrection (Harris & Wolchik, 1979) or contingent restraint (Gaylord-Ross, Weeks, Lipner, & Gaylord-Ross, 1983). Other research has suggested that DRO/DRI methods are effective only when used in combination with aversive methods such as reprimands (Rolider & Van Houten, 1984), brief restraint (Friman, Barnard, Altman, & Wolf, 1986), or time-out (Repp& Deitz, 1974). But at least some research has shown that DRO/DRI used alone can reduce behavior problems effectively (Barton, Brulle, & Repp, 1986; Luiselli & Slocumb, 1983). Although the evidence appears to favor combining aversive and nonaversive components for effective response reduction, more research on the effects of DRO/DRI used alone is needed.

An important, yet virtually unstudied, question is how providing nonaversive treatment affects provider behaviors. Data are usually taken only on child behaviors; yet most treatment, training, or educational situations for children with severe disabilities involve interactions between children and their providers. Evaluating changes that occur in provider behavior as providers deliver treatment could increase knowledge about treatment effects.

Another infrequently studied question is whether nonaversive interventions produce unwanted side effects. Side effects are often addressed in research on aversive interventions (cf. Matson & Taras, 1989), but rarely in research on reinforcement-based reductive procedures (Homer & Peterson, 1980). Although side effects of nonaversive interventions seem improbable, a potential for adventitious reinforcement of untargeted behavior problems exists (Repp & Deitz, 1974). Therefore, monitoring responses other than those targeted by nonaversive treatment would also add to knowledge about its effects.

This experiment addressed these questions by evaluating the direct effects of DRI on disruptive behavior in a boy with severe disabilities. As well, the study evaluated the indirect effects of DRI on the boy's providers' use of physical restraint and the boy's appropriate toy play.



Providers. The providers in this experiment were three graduate trainers for children with severe disabilities: Provider 1 was a 30-year-old graduate student in psychology; Provider 2 was a 26-year-old graduate student in occupational therapy; and provider 3 was a 28-year-old post-doctoral intern in psychology. All three, who were enrolled in a university-affiliated program, were learning one-to-one training methods for children with severe disabilities.

Child. The child in this study was Tom, a 4-year-old boy who was hyperactive and highly disruptive. He had no functional expressive speech and extremely delayed receptive speech. He could walk independently, and he often ran away from his caregivers. His activity level was very high, and he resisted or ignored most verbal and gestural instruction. Tom had a short attention span. His interests and corresponding movements had a random appearance; for example, when he entered a room, he usually ran around the perimeter, touched all walls, touched all objects within his reach, and attempted to climb on all surfaces. While conducting these activities, he usually yelled or laughed loudly. If a caregiver attempted to inhibit him, he actively resisted by yelling or crying, kicking, spitting, and throwing objects. If these methods failed he would passively resist by falling limply to the floor.

Although Tom's fine motor skills were severely delayed, the parents reported he could complete simple four-piece puzzles and large six-piece peg boards. He completed no puzzles or pegboards during baseline, however.

Tom's parents brought him to a psychology clinic to obtain assistance in finding an appropriate developmental preschool. The parents had contacted several programs thought suitable for him, but all had denied Tom enrollment because of his disruptive and maladaptive behaviors. The primary deterrent to enrollment was his reluctance (refusal) to sit in a chair. The mother complained he would not even sit for meals (she described her feeding sessions as "meals on wheels").

During preliminary observations of Tom's behavior, the validity of his mother's complaint (and that of the various preschools who had rejected his application) was evident. He refused to be seated, and he kicked and hit at those (including his mother) who attempted to seat him. Not being seated left him free to run around and sometimes away from the study setting, unless he was restrained by an adult. Once before the experiment and once just before a baseline session, he ran from the study setting and out into the main hospital before he was retrieved. Once after a baseline session, he ran out of the hospital itself and was retrieved in a hospital parking lot. Tom was selected for this experiment because his behavior provided a rigorous test of a strictly nonaversive intervention. Quite simply, Tom's behavior prompted restraint from all his caregivers, including his parents. Because of the rate and the range of Tom's escapes from instructional settings, it was apparent that for Tom, inseat behavior would not only be an important prerequisite to preschool instruction but a survival skill as well.


The study was conducted in an empty classroom (approximately 3 X 3.6 meters, m), which contained one desk, two chairs, and an array of toys (e.g., dolls and puzzles) and small food substances (e.g., fruit punch and raisins). At the rear of the classroom was a .6 X .9 m one-way mirror that permitted unobtrusive video recording of sessions.

Behavior Definitions

Behavior definitions were kept standard across all observers, settings, and conditions. This experiment focused on two of Tom's behaviors, out-of-seat behavior and appropriate toy play; and one provider behavior, use of restraint.

Out-of-Seat Behavior. Out-of-seat behavior was defined as an occasion when Tom's gluteus maximus (extending from buttock to mid-thigh) rose completely off the flat bottom surface of his chair or when his waist rose above the work table.

Appropriate Toy Play. Appropriate toy play was defined as an occasion when Tom touched a toy with two or more fingers of one hand in an attempt to manipulate it while seated.

Provider Use of Restraint. Restraint was defined as an occasion where a provider put his or her hand on any part of Tom's body and exerted pressure in an attempt to get Tom to his seat or keep him in his seat.


All sessions were videotaped, and measures of each response were obtained directly from the tapes. The observers viewed each tape three times. During the first viewing, a cassette recording signaling 1-minute (min) intervals and the videotape were played simultaneously. Observers recorded the number of occurrences of out-of-seat behavior in 1-min intervals. During the second and third viewing, the cassette recorder simultaneously signaled 10-second (s) intervals. The observers recorded occurrence or nonoccurrence of physical restraint in each interval during the second viewing and the occurrence or nonoccurrence of appropriate toy play in the third. Inappropriate behaviors were observed incidentally.


Reliability was established through interobserver agreement by having a second observer make periodic simultaneous observations using the same recording procedure as the primary observer. Agreement for all behaviors of interest was established by comparing the observer's data sheet on an interval-by-interval basis and then dividing the agreement intervals by the sum of agreement and disagreement intervals and multiplying by 100. Agreement data were obtained in two baseline sessions and two intervention sessions for all behaviors and all providers. The data obtained yielded overall agreement means of 91% (range 66%-100%) for out-of-seat behavior, 94% (range 91%-100%) for appropriate toy play, and 89% (range 82%-98%) for use of physical restraint.

Integrity of the Independent Variable

Establishing that the intervention being evaluated is actually used is an important issue in behavior analysis (Peterson, Homer, & Wonderlich, 1982). Estimates of the schedule of reinforcer delivery were obtained by observing the tapes of all sessions and applying the following formula: session length in seconds/number of reinforcer deliveries. In the fixed 10-s schedule, the average time between deliveries was 11.1 s (90%) for trainer 1, 11.5 s (87%) for trainer 2, and 12.5 s (80%) for trainer 3. In the fixed 20-s schedule, the average times were 21.6 s (93%) for trainer, 1, 22.3 s (90%) for trainer 2, and 23.2 (86%) for trainer 3. In the varied 20-s schedule, the average was 21.1 (94%) for trainer 3.


Tom's program was conducted four times a week for 6 weeks. Providers were given 3 min each to orient themselves and Tom before video recording. The instant recording began, an audio recorder began to announce timed intervals. For all baseline sessions and the first three treatment sessions, the recorder announced an interval every 10 s for 5 min. After the third treatment session, the interval was expanded to 20 s, and the sessions were lengthened to 10 min. The follow-up session was lengthened to 20 min. In this session an average interval length of 20 s was maintained, but the length of individual intervals was varied from 4 s to 38 s.

Baseline. The providers worked with Tom on a one-to-one basis. Each provider was told that Tom had been denied enrollment in preschool because he would not remain seated and that their task was to teach him to sit at a table and play with toys. During baseline, they were given no specific instruction about how to accomplish their task.

DRI. Following baseline sessions, a research assistant complimented the providers individually on their teaching efforts and then instructed them to deliver an edible reinforcer whenever Tom remained seated for an entire interval as announced by the audio recorder. They were told to provide the reinforcer even if physical assistance (i.e., restraint) was used to keep him seated.

Follow-up. Provider 3 conducted a 20-min follow-up session 4 months after the intervention. He was chosen because the other providers had graduated. The follow-up conditions were the same as those used during the experiment, with the exception of the varied intervals.

Experimental Design

The results of treatment were assessed using a multiple-baseline-across-providers design (Barlow & Hersen, 1984). The multiple baseline was chosen over the stronger withdrawal design for two reasons. First, Tom's behavior and the reactions it provoked were so disruptive that it seemed inappropriate to allow any positive changes produced by DRI to extinguish during a withdrawal when a design not requiring a withdrawal was available. Second, the participation of three providers allowed for three direct replications of the intervention, a number that meets recommended criteria for demonstrations of experimental control (Barlow & Hersen, 1984).


Out-of-Seat Behavior

Figure 1 shows the effect of DRI on Tom's out-of-seat behavior. During baseline, out-of-seat behavior occurred at a mean rate per minute of .52 for provider 1,.93 for provider 2, and .54 for provider 3. After DRI, these rates were reduced to zero for provider 1,.04 for provider 2, and .06 for provider 3.


Figure 2 shows the collateral effect of DRI on provider use of physical restraint. During baseline, the mean percentage of intervals in which providers used restraint was 13% for provider 1, 30% for provider 2, and 27% for provider 3. After DRI, these rates were reduced to zero for provider 1, 1% for provider 2, and 3% for provider 3.

Appropriate Toy Play

Figure 3 shows the collateral effect of DRI on appropriate toy play. Before the introduction of reinforcement procedures, the mean percentage of intervals in which appropriate toy play occurred was 94% for trainer 1, 73% for trainer 2, and 81% for trainer 3. After DRI, these percentages increased to 99% for trainer 1, 95% for trainer 2, and 98% for trainer 3. Spitting, hitting, and kicking did not occur after intervention.


During the follow-up session, Tom remained seated for 20 min, playing appropriately with toys; and the provider did not use restraint.


The purpose of this study was to assess the direct effects of DRI on Tom's out-of-seat behavior and its indirect effects on provider use of restraint and Tom's appropriate toy play. The multiple baseline showed that each provider encountered high rates of out-of-seat behavior until administering DRI, which was immediately followed by a near-zero rates of the target behavior. In fact, the duration of in-seat behavior increased from less than 1 min during baseline to 10 min during the intervention, and to 20 min during follow-up. Although baselines were expanded by only one data point across providers, the sequenced replication of the effect meets accepted standards for experimental control (Barlow & Hersen, 1984, p. 225 and 242). The results also show provider use of restraint reduced to near zero rates and Tom's appropriate toy play increased to almost 100% of intervals. A final related result was Tom's enrollment in a pre-school (which had previously denied his application) at the experiment's end.

This study contributes to the literature in at least three ways. First, it adds to the sparse research showing that strictly reinforcement-based interventions can substantially reduce high-rate behavior problems in children with severe disabilities. Most of the research shows that such interventions require an aversive component to produce clinically significant changes (e.g., Friman et al., 1986; Gaylord-Ross et al., 1983; Harris & Wolchik, 1979; Repp & Deitz, 1974; Rolider & Van Houten, 1984).

Second, the study shows that DRI also produced desirable changes in provider behavior. Specifically, each provider reduced his or her use of physical restraint to near zero rates even though it was not incompatible with DRI and was not prohibited by the investigators. The providers reported (independently during debriefing) resorting to restraint because it produced brief periods of sitting. They reported further that DRI, because it resulted in sustained sitting, rendered restraint unnecessary.

Third, the study shows that DRI did not produce unwanted side effects through adventitious reinforcement (cf. Repp & Deitz, 1974). Quite the contrary, the data indicate an increase in appropriate toy play across providers, and incidental observations indicate a decrease in several untargeted inappropriate behaviors (e.g., spitting, mouthing, throwing, and kicking). It should be noted that Tom's baseline levels of appropriate toy play are somewhat inflated because the response definition required only that Tom manipulate a toy while seated. It was written this way because preexperimental observations revealed that Tom's toy play was very primitive. As a result, several incidents involving throwing or mouthing toys were scored as toy play during baseline because they involved manipulation of toys while Tom was seated. Despite the highly sensitive response definition, appropriate toy play increased sharply across providers (see Figure 3). And, because the level and topography of appropriate toy play changed so substantially during DRI (e.g., Tom completed puzzles and pegboards without mouthing or throwing the pieces), a more specific definition would have yielded even more dramatic effects.

Clearly the findings are limited. For example, although three providers were studied, they worked with only one child. Also, despite the high level of Tom's disruption, his behavior was not as severe as behaviors often treated with aversive methods (e.g., self-injury). Future research should include additional children with more severe problems. Another limitation is that all sessions were conducted in an experimental and not a regular classroom. The DRI was normalized somewhat by lengthening it to an average of 20 s and varying it from 4 s to 38 s, resulting in a follow-up session where Tom remained seated for 20 min playing appropriately. Future research could normalize it further by conducting probes in regular settings.

Despite its limitations, this study shows that high-rate behavior problems can be decreased nonaversively even in children with a severely diminished capacity to learn. More imporant, the findings suggest that efforts expended in criticism of aversive procedures may be more productively channeled toward experiments showing the benefits of nonaversive procedures for children and providers.


Barlow, D. H., Hersen, M. (1984). Single case experimental designs (2nd ed.). New York: Pergamon.

Barton, L. E., Brulle, A. R., & Repp, A. C. (1986). Maintenance of therapeutic change by momentary DRO. Journal of Applied Behavior Analysis, 19, 277-282.

Friman, P. C., Barnard, J. D., Altman, K., & Wolf, M. M. (1986). Parent and teacher use of DRO and DRI to reduce aggresive behavior. Analysis and Intervention in Developmental Disabilities, 6, 319-330.

Gaylord-Ross, R. J., Weeks, M., Lipner, C., & Gaylord-Ross, C. (1983). The differential effectiveness of four treatment procedures in suppressing self-injurious behavior among severely handicapped students. Education and Training of the Mentally Retarded, 18, 38-44.

Harris, S. L., & Wolchik, S. A. (1979). Suppression of self-stimulation: Three alternative strategies. Journal of Applied Behavior Analysis, 12, 185-198.

Homer, A. L., & Peterson, L. (1980). Differential reinforcement of other behavior: A preferred response elimination procedure. Journal of Applied Behavior Analysis, 11, 449-471.

Lavigna, G. L. & Donnellan, A. M. (1986). Alternatives to punishment: Solving behavior problems with nonaversive strategies. New York: Irvington.

Luiselli, J. K., & Slocumb, P. R. (1983). Management of multiple aggressive behaviors by differential reinforcement. Journal of Behavior Therapy and Experimental Psychiatry, 14, 343-347.

Matson, J., & Taras, M. E. (1989). A 20 year review of punishment and alternative methods to treat problem behaviors in developmentally delayed persons. Research in Developmental Disabilities, 10, 85-104.

Mulick, J. A., & Linscheid, T. R. (1988). (Review of Alternatives to punishment: Solving behavior problems with non-aversive strategies). Research in Developmental Disabilities, 9, 317-330.

Peterson, L., Homer, A. L., & Wonderlich, S. A. (1982). The integrity of the independent variable in behavior analysis. Journal of Applied Behavior Analysis, 15, 477-493.

Repp. A. C. & Dietz, S. M. (1974). Reducing aggressive and self-injurious behavior of institutionalized retarded children through reinforcement of other behavior. Journal of Applied Behavior Analysis, 7, 313-325.

Rolider, A., & Van Houten, R. (1984). The effects of DRO alone and DRO plus reprimands on the undesirable behavior of three children in home settings. Education and Treatment of Children, 7, 17-31.

Van Houten, R., Axelrod, S., Bailey, J. S., Favell, J. E., Foxx, R. M., Iwata, B. A., & Lovaas, O. I. (1988). The right to effective behavioral treatment. Journal of Applied Behavior Analysis, 21, 381-384.

PATRICK C. FRIMAN is an Assistant Professor of Psychology in Pediatrics, University of Pennsylvania School of Medicine, Philadelphia.

The preparation of this manuscript was supported in part by DHEW grant MCT-000944 and NIH Grant No. 025 28/10 to the Children's Rehabilitation Unit at the University of Kansas Medical Center, and by Project 405, a Maternal and Child Health Services Grant to the Meyer Children's Rehabilitation Unit at the University of Nebraska Medical Center. The author wishes to thank Karl Altman, Keith Allen, Bill Warzak, and George Williams for help with the research and the manuscript. Requests for reprints should be sent to P. C. Friman, Pediatric Psychology, Children's Seashore House, Philadelphia Center for Health Care Sciences, 3405 Civic Center Blvd., Philadelphia, PA 19104.
COPYRIGHT 1990 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Friman, Patrick C.
Publication:Exceptional Children
Date:Sep 1, 1990
Previous Article:Postsecondary experiences of young adults with severe physical disabilities.
Next Article:Prediction is not the only measure of a plan: a response to Marston.

Related Articles
The effects of initial interval size on the efficacy of DRO schedules of reinforcement.
The relationship between instructional variables and problem behavior: a review.
First Inhaled Corticosteroid for Children 4 Years and Older Cleared for Use
New studies confirm efficacy of atomoxetine for ADHD children. (Still a Role for Stimulants).
Improved Treatment for Acute Lymphoblastic Leukemia Offers Hope for Eliminating Irradiation; St. Jude clinical trial XIIIB was based on stringent...
Response-class hierarchies and resurgence of severe problem behavior.
A full picture of methamphetamine.
Exposure to violence disrupts kids' health.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters