The Effect of Choice-Making Opportunities during Activity Schedules on Task Engagement of Adults with Autism

Journal of Autism and Developmental Disorders, Vol. 33, No. 5, October 2003 ( 2003) The Effect of Choice-Making Opportunities during Activity Schedules on Task Engagement of Adults with Autism Mari Watanabe 1 and Peter Sturmey 2,3 Increasing choice and participation by adults with autism spectrum disorders is an important, but neglected, aspect of research and services. This study evaluated the effects of choice-making opportunities, embedded within activity schedules, and contingent praise on the engagement of three adults with autism in a community vocational setting. In the baseline condition, staff assigned the order of the tasks. In the Choice condition and Maintenance phases, the participants chose the order of tasks that supervisors assigned to them. They made their own activity schedules by writing down the order of their tasks for that morning. Social praise was provided contingent on the participant s task completion. The same tasks were used in baseline, intervention, and maintenance phases. During the Choice and Maintenance conditions, client engagement was substantially higher than baseline for all three participants. Increasing choicemaking opportunities within activity schedules was an effective and socially acceptable way to increase choice and engagement in adults with autism. KEY WORDS: Choice; activity; schedule; engagement. Promoting choice making has been used to increase engagement in activities in people with disabilities. Moes (1998) provided choice-making opportunities to children with autism. That study found positive effects on the percentage of correct responding, percentage of intervals with disruptive behavior, rate of homework completion, and child affect. Dunlap et al. (1994) demonstrated that choice-making increased task engagement and decreased disruptive behavior of children with emotional and behavioral challenges. Although effective, both of these studies required substantial amounts of staff time and effort, as the teachers had to stay with the children all through the session. Furthermore, these studies used procedures that probably brought client behavior under the stimulus control 1 Department of Psychology, Queens College, Flushing, NY. 2 The Graduate Center, City University of New York. 3 Correspondence should be addressed to Peter Sturmey, Department of Psychology, Queens College, City University of New York, 65-30, Kissena Boulevard, Flushing, NY 11367. 535 of staff prompts. It is desirable, especially with adult participants, to find a means to improve independent choice-making and engagement. Activity schedules have been demonstrated to be effective in increasing independent behavior of people with disabilities (McClannahan & Krantz, 1999). MacDuff, Krantz, and McClannahan (1993) showed that activity schedules increased independent engagement in daily tasks of children with autism. Anderson, Sherman, Sheldon, and McAdam (1997) demonstrated that pictorial activity schedules increased engagement in activities of adults with mental retardation living in a group home. Anderson et al. (1997) concluded that choice-making and activity schedules worked together to enhance the subjects participation and independence in the study. However, the authors did not investigate the effect of reinforcers. During their schedule condition, they used tokens for one subject along with choice-making opportunities and activity schedules. For the other two subjects, the delivery of reinforcers was not reported. Therefore, the data were not sufficient 0162-3257/03/1000-0535/0 2003 Plenum Publishing Corporation

536 Watanabe and Sturmey to conclude that the combination of choice-making and activity schedules caused the change in client behavior. This evaluated the effect of choice-making within activity schedules on participants task engagement. This study extended previous literature, first, by including adults with autism in a community vocational setting and, second, by separating the effects of activity schedules and choice-making from contingency management effects. METHOD Setting and Participants The study took place in an adult services program for people with developmental and behavioral disorders located in an urban area. Three men with autism Mark, 22 years old; Bob, 40 years old; and Nick, 30 years old were selected as participants. At the beginning of the study, they had participated in the program for 1 year and 1 month, 2 years and 8 months, and 6 years and 3 months, respectively. These men belonged to the program s highest-functioning group, which consisted of nine adults with autism and other developmental and behavioral disorders. These participants exhibited inappropriate behavior more often than other members of this group when they were not engaged in tasks. The inappropriate behavior included making noises, handflapping, and talking to themselves. Their room was approximately 20 20 feet. There were five long tables and 15 chairs for the clients. The supervisor s desk and chair were in front of the room facing the clients. Behind the supervisor was a blackboard. A clock was hung above the blackboard so all the clients could see it. Procedure The clients tasks included, but were not limited to, math drills, reading comprehension, handwriting practice, personal hygiene check, job search, and letterwriting. None of these tasks was novel for the participants. Each task involved writing on paper. The study was simultaneously conducted with three participants from 9 A.M. to11:45 A.M. for 23 days (23 sessions).one session consisted of three tasks. The participants on-task behavior was recorded using 1-minute momentary time sampling. Observations were made during a 30-minute observational period which started at 9:30, 10:15, or 11:05 A.M. The starting time was randomly selected each day of the study. The observer marked the participants on-task behavior (+) and off-task behavior ( ) onadata sheet. Percent of on-task behavior was calculated by dividing the number of scored intervals by the total number of intervals. On-task behavior was defined as reading the task paper or activity schedule sheet (indicated by the participant looking at the paper with eyes moving), writing on paper or activity schedule sheet appropriately, correcting writing with an eraser, walking toward a drawer to get a pencil and walking back to the seat, going to the next room to use a pencil sharpener, and asking questions about the task. Off-task behavior was defined as looking at the paper without eyes moving, doodling or scribbling, leaving the seat for reasons other than getting a pencil or sharpening a pencil, and engaging in inappropriate behavior such as vocal noise, hand-flapping, noncontextual speech, body-rocking, and sleeping. Experimental control was demonstrated using a multiple-baseline-across-subjects design. Each subject had a baseline phase in which no choices were made, an intervention Choice phase, and a Maintenance phase. The independent variable of the study was choicemaking opportunities provided to the participants during activity schedules. The independent variable was the percentage of on-task behavior. During baseline, the experimenter wrote the morning schedule on the blackboard and gave the task papers to the participants. The participants had three tasks in one session. It was expected that the tasks would be completed within 40 minutes. There were 5- to 10- minute breaks between the tasks. The experimenter told the participants when to start tasks and take breaks (e.g., It s 10:55. Please take a 5-minute break, or It s 11:05. Let s go back to the activity. ). Other prompts were not given. The experimenter gave verbal praise such as, Good job, or, You tried hard, didn t you? with smiles and eye contact to the participants when they finished the task during the scheduled period. If the participant did not finish the task within the scheduled time, the experimenter told him to take a break and start working on the next task after the break. If the participant completed the task earlier, he was told to take a long break until the next period started. The Choice condition was conducted in the same manner as in Baseline, except at the beginning of the session, the experimenter said, Please make today s schedule. The experimenter then gave a schedule sheet and a list of nine activities to the participants. On the schedule sheet, there were three empty columns next to the times 9:30 10:10, 10:15 10:55, and 11:05 11:45. The participants wrote the names of the tasks they had chosen in the columns to make the day s activity schedule. The experimenter told the participants to leave a schedule sheet on the table where they could always

The Effect of Choice-Making Opportunities during Activity Schedules on Task Engagement 537 see it. No schedule was written up on the chalkboard. During Maintenance, the sessions were conducted in the same manner as in Choice condition, except the experimenter did not provide the verbal prompts. A second observer simultaneously and independently recording the behavior performed interobserver agreement checks. The checks were made in each phase and for 30.4% of all sessions. Agreement scores were calculated by dividing the number of agreements by the number of agreements plus disagreements and multiplying it by 100%. The mean score for all three participants was 92.9% (range = 86.7 to 100%). The respective mean scores for Baseline, Choice condition, and Maintenance phases for each participant were 97.3%, 91.7%, and 86.7% for Mark; 100%, 97.3%, and 97.3% for Bob; and, 100%, 100%, and 87.7% for Nick. RESULTS Results are shown in Figure 1. Mark, Bob, and Nick, respectively, had 6, 9, and 12 days of Baseline; 7, 8, and 7 days of Choice condition; and 5, 4, and 3 days of Maintenance phase. Mark was absent or not available on days 7, 12, 13, 14, 15, and 23. Bob was absent on days 5 and 15. Nick was absent on day 13. There was a 2-month interval between day 15 and 16, during which the participants were often expected not to be available for the study. Mark s average time on task was 19.4% during Baseline (range = 1.3% 46.7%; SD = 16.7), 50.8% during the Choice condition (range = 36.7% 73.3%; SD = 12.3), and 59.0% during Maintenance (range = 40.0% 70.0%; SD = 12.3). Bob s average time on task was 23.9% during baseline (range = 0% 63.3%; SD = 22.1), 67.5% during the Choice condition (range = 28.0% 90.0%; SD = 19.3), and 63.6% during maintenance (range = 56.7% 63.3%; SD = 3.3). Nick s average time on task was 40.8% during Baseline (range = 1.7% 76.7%; SD = 24.3), 58.6% during the Choice condition (range = 35.3% 70.0%; SD = 12.1), and 76.7% during Maintenance (range = 70.0% 90.0%; SD = 11.5). The mean percentages of on-task behavior for all three participants were 28.0% in Baseline, 59.0% in the Choice condition, and 65.4% during Maintenance condition. Thus, participants spent approximately twice Fig. 1. Percentage of 1-minute momentary time samples scored as on-task behavior in Baseline (no choice), Intervention (choice), and Maintenance phases. There is a 2-month interval between days 15 and 16.

538 Watanabe and Sturmey as much time independently engaged during the Choice and Maintenance conditions compared with Baseline. Inspection of Figure 1 indicates that time on task was a function of the introduction of the Choice condition. During the Maintenance phase, although its length was relatively shorter than the other conditions, time on task remained long and the performance was least variable for all three participants. DISCUSSION This study demonstrated that the introduction of choice-making embedded within activity schedules increased time on task for all participants. Furthermore, this outcome was maintained during the Maintenance phase when staff prompts were eliminated. Thus, this study replicates the results of other studies demonstrating that choice-making can be an effective antecedent-based intervention to increase time on task (Anderson et al., 1997; Dunlap et al., 1994; Moes, 1998). Furthermore, we isolated the effect of choice making from the effects of contingency management, which was confounded with the introduction of choicemaking in earlier studies (Anderson et al., 1997). Thus, the availability of praise and other social consequences in Baseline of our study did not result in increases in on-task behavior during that phase. Finally, in the Maintenance phase, we were able to reduce client dependence on staff prompts still further, without decreasing time on task. Thus, we conclude that choice-making embedded within activity schedules was a powerful intervention that caused in substantial increase in time on task for all three participants. In this study, the primary objective was to improve the participants task engagement. Therefore, the percent of on-task behavior was selected as the dependent variable. Although not directly recorded, some other positive effects were observed. First, as the participants on-task behavior improved, their inappropriate behavior decreased. As Anderson et al. (1997) mentioned, people are less likely to exhibit maladaptive behavior when exhibiting on-task behavior. Second, productivity of the participants seemed to have improved, as the participants appeared to complete their tasks within the 40 minutes more often during the Choice and Maintenance phases. There may be at least one prerequisite for the choice-making procedures. Parsons, Reid, Reynolds, and Bumgarner (1990) pointed out that participants must demonstrate skills to choose tasks for studies on choice-making. The effectiveness of the procedures for participants may depend on participants having choicemaking skills. If one lacks such skills, intervention to teach appropriate choice-making behavior may be needed (Parsons et al., 1990). In this study, the activity schedules took different forms in Baseline and in the Choice condition. Future studies should use the same form of schedule in all conditions. Future research also could focus on further component analyses of choice-making procedures. For example, we do not yet know the relative contribution of individual elements of choice-making, such as identifying tasks to engage in, choosing the sequence of those tasks, and setting goals to complete a task. In addition future research could also widen the range of dependent variables used to evaluate choice-making in adults with autism to include measures of maladaptive behaviors and productivity measures. REFERENCES Anderson, M. D., Sherman J. A., Sheldon, J. B., & McAdam, D. (1997). Picture activity schedules and engagement of adults with mental retardation in a group home. Research in Developmental Disabilities, 18, 231 250. Dunlap, G., deperczel, M., Clarke, S., Wilson, D., Wright, S., White, R., & Gomez, A. (1994). Choice making to promote adaptive behavior for students with emotional and behavioral challenges. Journal of Applied Behavior Analysis, 27, 505 518. Guess, D., Benson, H. A., & Siegel-Causey, E. (1985). Concepts and issues related to choice-making and autonomy among persons with severe disabilities. Journal of the Association for Persons with Severe Handicaps, 10, 79 86. MacDuff, G. S., Krantz, P. J., & McClannahan, L. E. (1993). Teaching children with autism to use photographic activity schedules: Maintenance and generalization of complex response chains. Journal of Applied Behavior Analysis, 26, 89 97. McClannahan, L. E., & Krantz, P. J. (1999). Activity schedules for children with autism: teaching independent behavior (Topics in Autism). Bethesda, MD: Woodbine House. Moes, D. R. (1998). Integrating choice-making opportunities within teacher-assigned academic tasks to facilitate the performance of children with autism. Journal of the Association for Persons with Severe Handicaps, 23, 319 328. Parsons, M. B., Reid, D. H., Reynolds, J., & Bumgarner, M. (1990). Effects of chosen versus assigned jobs on the work performance of persons with severe handicaps. Journal of Applied Behavior Analysis, 23, 253 258.

Copyright of Journal of Autism & Developmental Disorders is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.