High-Probability Requests and a Preferred Item as a Distractor: Increasing Successful Transitions in Children with Behavior Problems.

Abstract

The performance of successful transitions by young children can be a critical factor in the provision of inclusive educational services. This study compared the effects of two interventions (high-probability requests and preferred item as a distractor) on the success of classroom transitions of two young children with behavior problems. Additionally, this study examined the social validity for the two procedures through the use of questionnaires and direct observations of the interventionist maintenance in using the strategies. Results indicate that both interventions were effective in increasing successful transitions. Educational implications and measures of social validity are discussed.

With an increasing emphasis on inclusive educational services for children with disabilities, the degree of independence of a child becomes a critical factor (Carla, Atwater, Schwartz, & Miller, 1990; Sainato, 1990). One such skill in determining independence is the successful transitions both within school and within classroom routines. Additionally, multiple daily transitions occur from daycare to school, to afternoon daycare! latch key, and finally home (Ostrosky, Donegan, & Fowler, in press). Difficulties encountered during transitions may emanate from a multitude of functions that include (1) a desire to escape the approaching activity/setting, (2) a desire to reestablish the activity that has just been terminated, (3) a generalized escape response to the verbal instruction associated with transition regardless of the task preference, and (4) generalized reaction to noise/confusion and increased activity level during transitions. The magnitude of interruption to a classroom can only be fully realized whe n one considers that transitions may occur during 20 to 30% of the classroom time in the typical preschool/early elementary child's day (Carta et al., 1990; Sainato & Lyon, 1983).

Over the last ten years, researchers have developed a milieu of proactive strategies that focus on challenging behavior that occurs during transitions. Among these are high-probability requests and preferred items as a distractor. High-probability (high-p) request sequences have been used as an antecedent strategy to increase appropriate behavior during transition from playtime to instructional group time (Singer, Singer, & Homer, 1987), task attempts (Homer, Day, Sprague, O'Brien, & Heathfield, 1991), compliant responding (Davis, Brady, Williams, & Hamilton, 1992; Mace, Hock, Lam, West, Belfiore, Pinter, & Brown, 1988; Mace & Belfiore, 1990) and social interactions (Davis, Brady, Hamilton, McEvoy, & Williams, 1994; Davis & Reichle, 1996). Typically, during a high-probability request sequence, the interventionist delivers 3 to 5 easy requests to which a student has a history of responding (i.e., high-probability requests) immediately before the delivery of a request to which the student does not typically res pond (i.e., low-probability request). The results of previous studies have clearly demonstrated that high-probability requests can be an effective strategy for students who engage in escape-motivated challenging behavior. Several reasons for the effectiveness of high-probability request sequences have been offered. Singer, Singer, and Homer (1987) suggested that high-probability requests change the density of teacher-delivered positive requests for engagement. As a result, activity preferredness may change over time. Alternatively, Homer et al. (1991) propose that a history of compliance to positive requests may create sufficient opportunities to promote a more generalized class of instruction-following that includes less preferred activities. A third explanation offered by Mace and his colleagues (1988) involves the establishment of behavioral momentum. Nevin, Mandell, and Atak (1983) draw an analogy between physical momentum and a behavior's resistance to change. That is, by increasing a behavior's response rate (velocity) and the rate of reinforcement, a momentum of responding is established such that responding is less likely to be interrupted in the presence of a lower-probability request. Mace and his colleagues propose that the delivery of three high-p requests increases the response rate and the delivery of reinforcement for responding to the requests increases the rate of reinforcement for that behavior, thus building a behavior's momentum.

Preferred item/event as a distractor has been offered as an intervention strategy when an activity associated with challenging behavior does not require active participation. For example, when a visual spectacle such as a quiet aquarium scene was presented during invasive dental procedures, clients were less likely to require pain management medication. Similarly, Reichle (1995) demonstrated that activities during transportation to and from school dramatically decreased challenging behavior associated with boredom (lack of activity engagement).

Clearly, there are a number of viable interventions that might be used to address challenging behavior exhibited during transitions. Unfortunately, few studies have directly compared the effectiveness of these strategies in decelerating challenging behavior. An equally important issue is the acceptability of the procedure from the perspective of the individuals who implement the procedure.

Within recent years, the field has made great strides in examining issues of social validity and treatment acceptability (Wolf, 1978; Schwartz & Baer, 1991; Polloway, 1996; Reiners, Wacker, Derby, & Cooper, 1995). To date, typical methods of social validity and treatment acceptability have consisted of surveys, rating scales, and in some more rigorous instances, a videotape and survey combination. This last method consists of participants viewing a videotaped example of a particular procedure and then answering questions regarding the validity of the procedures or behavior change (Peterson & McConnell, 1993). None of these procedures have taken into account the degree of correspondence between an individual's report and their actions. That is, while practitioners may rate a question such as "I would use this procedure in my classroom" high, their actual use of the procedure may be far less. Schwartz and Baer (1991) suggested that a much stronger and reliable measure of social validity may be the observation o f behavior change. Therefore, this study provided some direct observation measures of social validity.

The purpose of the study was twofold: (1) to compare the effects of the high-p request and preferred item as a distractor procedures on the success of classroom transitions, (2) to examine social validity for the two procedures by examining interventionist preference and the correspondence between the effectiveness of each of the two procedures and interventionist maintenance in using each of the two procedures.

Method

Participants and Settings

The participants in this study included two boys who exhibited severe challenging behavior during transition times in the classroom. Rob, a 6-year-old boy with Down syndrome, attended afternoon kindergarten at a public school in the suburbs of a large mid-western city. A standardized assessments (the PPVT-R) indicated Rob's expressive language to be in the severe range at a standard score of 44. His educational placement was a regular education kindergarten class which he attended with 27 typically developing peers. He received speech and language services, early childhood special education services, occupational therapy, and participated in adaptive physical education. Rob's daily schedule consisted of approximately 40 minutes receiving special education services and 1 hour 50 minutes in the regular education classroom. Transition times had been identified as a major problem by all of the educators and paraprofessionals who worked with Rob and was a behavioral objective on his IEP. Whenever Rob was asked to transition from one activity to another or one room to another, he might: (a) drop to the floor, (b) hit or kick at an adult close by, (c) scream/vocalize loud enough to be heard across the school building, or (d) any combination of these behaviors. These behaviors resulted in the delivery of full physical assistance, physical restraint for safety of others, and the disruption of classrooms across the school.

The second participant, Travis, was a 6-year-old boy who had been identified with emotional/behavioral disorder and mild mental retardation. Standardized test scores indicate Travis' full scale IQ to be 66 as reported by the WISC-R. Travis was enrolled in a multi-graded (K-3rd grade) segregated classroom serving students with emotional/behavioral disorders. The classroom was located on a regular education campus. However, Travis participated in recess, lunch, and music with regular education students. Transitions had been identified by all of Travis' educators as a problem area. Upon being requested to transition, Travis hit, dropped to the floor, screamed, and pouted. Typically these actions resulted in a brief period of extinction followed by physical guidance.

Interventionists. Interventionists participating in the current study included a paraprofessional (for Rob) and a teacher (for Travis). These interventionists were chosen because they were the individuals who interacted with the student during the transitions in which challenging behavior occurred. That is, these interventionists delivered the request to transition from one activity to another during their daily schedules which resulted in challenging behavior from the target student. In addition, the teacher and paraprofessional requested that they participate as the interventionist to facilitate consistency when the project had been concluded. All interventionists were trained to use both interventions two days prior to intervention. Training consisted of (a) listening to a description of the procedures, (b) practicing implementing the procedures by role playing with one of the authors, and (c) role playing with another student not participating in the study.

Behavioral Definitions

The dependent variable was percent of successful transitions (i.e., compliance to a low-probability request). A successful transition was defined as independently walking from point A (i.e., place of the request) to point B (i.e., new activity or area) without engaging in challenging behavior. In addition, percent compliance to high-p requests served as a dependent variable. The independent variables included high-probability requests and preferred item as a distractor.

Low-probability request. The low-probability (low-p) request was defined as the requests that, when presented to the student, predictably produced challenging behavior. A low-p request is an instruction or request which the participant has a history of responding to less than 40% of the opportunities. For the purpose of this study, all low-p requests were a request to move from one area or activity to another, that is, a request to transition. For example, during the transition from the bus to the classroom, Rob predictably engaged in challenging behavior when the adult stated, "It is time to go into the classroom."

High-probability request sequence. A high-probability (high-p) request sequence is a set of (3 to 5) instructions or requests to which the participant has a history of responding, delivered immediately prior to a low-p request. A high-p request is a request to which a student responds correctly during 80% to 100% of the opportunities. A pool of high-p requests were determined for each participant by first interviewing school staff and parents. Each high-p request reported by parents and school staff was then validated by the interventionist delivering approximately 10 opportunities for each high-p requests across three school days. Those requests in which the student responded to 80% of the opportunities or better were included in the final pool of high-p requests. High-p requests included but were not limited to, "Give me five," "Stomp your feet," and "Touch your [body parts]" for Rob, and "Give me five," "Point to the paper," "Pick up the [school materials]" for Travis. An example of the implementation of t his intervention is located in Table 1.

Preferred item as a distractor. A preferred item as a distractor involves the delivery of a preferred item to distract the child from conditions associated with challenging behavior (i.e., low-p requests). That is, immediately prior to the teacher delivering a request to transisiton, she delivered the preferred item as a distractor to the participant. Each student's "preferred items" were determined through: (a) interviews with their respective special education teachers, his paraprofessional, and his parents, (b) direct observations at free play and other activities during the day, and (c) forced choice procedures. Preferred items identified for Rob included a favorite peer, a stopwatch, blowing bubbles, a blue race car, a digital watch, and a magnifying glass. Travis' preferred items included a running watch, locker keys, a clipboard, and several action figures.

Data Collection

Data were collected during transitions occurring throughout the day that had been associated with problem behavior. Given that one of the purposes of this research was to examine the practicality of the use of the two interventions by teachers, data were collected in natural opportunities. Therefore, rather than setting a specific number of opportunities per day, data were collected when the identified transition occurred. This resulted in transitions occurring on the average of five times per day (range from two to six) according to the student's schedule. An event recording system was used to collect data on both participants' successful transitions (dependent variable). The data collected included: (a) the transition that occurred, (b) the intervention that was used, and (c) whether or not the student responded correctly to the request to transition. In the case of the high-p sequence, data were also collected on the students' responses to the high-p requests. A successful transition was marked by a "+" an d an unsuccessful transition was marked by a "-".

Data were collected on each intervention's procedure. For high-probability requests, data were collected on the (a) delivery of the high-probability request, (b) interprompt time between the delivery of the high-p requests, (c) delivery of the low-p request, and (d) interprompt time between the last high-p request and the low-p request. For the preferred item as a distractor, data were obtained on (a) delivery of the preferred item prior to the request to transition, and (b) the child's actions on or use of one of the identified preferred items.

Data collectors included two graduate students in special education for Rob and one of the authors and a paraprofessional for Travis. Functional assessment information was collected by one of the graduate students for Rob and one of the authors for Travis. Data were collected for both participants during the last three months of school (March, April and May).

Design and Procedural Overview

Initially, a functional assessment was implemented with each participant to isolate the function of challenging behavior. Subsequently, a comparison of the effects of high-p requests and using a preferred item as a distractor in promoting an increase in successful transitions were evaluated by using a within-subject, alternating treatments design. Treatments were randomized across all transitions (teaching opportunities) that occurred during each day of intervention. Additionally, for one student a final condition examining the effects of the preferred item as a distractor only was implemented.

Procedures

Pre-assessment. Prior to the implementation of the intervention, an A-B-C assessment was used in an attempt to identify the stimuli (antecedents and consequences) that surrounded transitions. For both children, the A-B-C assessment identified that requests to transition resulted in challenging behavior and failure to independently transition from one activity to another. As a result, low-p requests were identified. See Behavioral Definitions for complete definition of low-p requests. Identified transitions for Rob included: (a) moving from the bus to the school building, (b) moving from the hallway to the regular education classroom, (c) going to circle or large group, (d) moving from a large group to a small group activity, (e) moving from the special education classroom back to the regular education classroom, (f) moving from outdoor play back to the regular education room, and (g) moving from the bathroom back to the classroom. Identified transitions for Travis included: (a) moving to the discourse room (a n instructional computer system), (b) moving back to the classroom from the gym, (c) moving out to the bus to go home, (d) moving to lunch, and (e) changing from reading to computer time.

Baseline. During baseline, the interventionist delivered a low-p request. That is, the interventionist asked the student to transition from one activity to another. The delivery of requests occurred naturally at the onset of a transition. If the student did not successfully complete the transition (respond to the low-p request), the interventionist was instructed to treat the transition as she typically would. This resulted in verbally and physically prompting the student to his destination.

Intervention. The setting and the context of the activities remained the same as in the baseline. However, the interventionist was instructed to deliver either a high-p request sequence or preferred item as a distractor immediately prior to a request to transition. It should be noted that the interventionist was instructed prior to the transition which intervention to use by the primary data collector. Using the high-p request procedure, the interventionist initiated a sequence of high-p requests prior to the low-p request. The sequence of events in the implementation of high-p requests included: (a) delivering three high-p requests (e.g., "Give me five," "Clap your hands," "Let me see you hop") selected from each participants' pool; (b) delivering verbal or gestural praise (i.e., thumbs up, wink, head nod) for each performance of a high-p request; and (c) delivering the low-p request (e.g., "Let's go to class" or "Time to go to group") immediately following the delivery of the reinforcement for performance o f the last high-p request. If a participant did not respond to one of the high-p requests, the interventionist paused, then continued with the next high-p request in the sequence. If he did not respond to the low-p request, the interventionist paused 5 seconds and verbally and physically prompted the child to his destination. It should be noted that all high-prequests were selected from each participant's pool of requests and were varied across trials (Davis & Reichle, 1996).

During the preferred item as a distractor procedure, the interventionist delivered a preferred item (e.g., a clipboard) immediately prior to the delivery of the low-p request (i.e., "Here is the clipboard, it's time to go into the classroom"). If the student engaged in challenging behavior prior to the delivery of a low-p request, the interventionist paused 5 seconds, did not deliver the intervention, and verbally and physically prompted the student to the next activity. This was done to avoid reinforcing a chain of behaviors that included: (a) challenging behavior, (b) delivery of the intervention, (c) physical assistance.

Measures of Social Validity

Social validity measures were obtained to help determine the practicality of the interventions when used in classroom settings. Social validity measures in Rob's classroom were obtained using a survey (see Figure 1) comprised of eight questions on which the service personnel either rated the question on a 1-to-5 rating scale or circled one of two response choices. Measures of social validity for interventions used with Travis were (a) the survey used previously, as well as (b) direct observations of the teachers during transitions. As with Rob's interventionists, Travis' interventionists were asked to complete the survey on the last day of the data collection. In addition, four weeks after the study was completed in Travis' classroom, a new observer was sent to observe the classroom under the auspices of a college "class assignment." This observer had no known connection to authors other than being a student in the first author's graduate class. The observer used data collection procedures that had been imple mented during intervention. Data were collected on (a) the proportion of transitions in which the interventionist used one of the interventions, (b) whether or not the interventions were successful, and (c) whether or not the interventions were delivered correctly.

Interobserver Agreement

Procedural reliability was assessed during 25% of all opportunities and was calculated by using an [Agreements/Agreements + Disagreements] x 100 formula. Procedural reliability was recorded as an agreement if the observer recorded the correct delivery of the intervention prior to the low-p request, and the delivery of the designated low-p request. For the high-p intervention data were collected on the (a) delivery of the designated high-p's. (b) delivery of social praise (e.g., thumbs up, "good job") for performance of the high-p request, and (c) delivery of the designated low-p request within 10 seconds of the social praise for performance of the last high-p request. For the distractor intervention data were collected on the (a) delivery of the designated preferred item as a distractor, and (b) delivery of the distractor before the low-p request. Procedural reliability for the high-p intervention was 100% across all three categories for both participants. For preferred item as a distractor, procedural reliab ility for delivery of the designated preferred item was 100% for both participants while procedural reliability for the delivery of the distractor prior to the low-p request was 100% for Rob and 98% for Travis. That is, Travis' interventionist once stated the low-p request before she delivered the distractor to the student, "Travis, it is time to line up for the bus. Here is the clipboard."

Interobserver agreement was collected on the participants' responses to (a) high-p requests, (b) acceptance of the distractor, and (c) requests to transition (i.e., low-p requests). Interobserver agreement for particpant's responses were collected on at least 25% of the opportunities for each condition for each participant. For Rob, interobserver agreement for responses was collected on 27%, 30% and 30% of the opportunities for baseline, HPR, and Distractor, respectively. For Travis, interobserver agreement for responses was collected on 27%, 25%, 25%, 27% and 27% of the opportunities for each of the conditions. It should be noted that reliability was collected on the percentage of opportunities across a condition. For example, Travis had 40 opportunities in the HPR condition, 25% of which interobserver agreement was collected (10 opportunities). Therefore, interobserver agreement for low-p responses are reported as one score. Interobserver agreement for responses to high-p requests was 98% and 100% for Rob a nd Travis, respectively. Agreement for acceptance of distractor was 100% for Rob and 100% for both conditions in which the distractor was presented for Travis. Interobserver agreement for responses to the low-p requests was 100% for Rob for each of the conditions. For Travis, interobserver agreement for response to low-p requests was: 100% for baseline, 90% for the HPR intervention, 90% for the Distractor intervention, 100% for return to baseline, and 100% for the Distractor only condition.

Results

Rob

Results of the high-probability request sequences and the preferred item as a distractor for Rob are represented in Figure 2. During baseline, Rob successfully transitioned from one area or activity to another during 1 of 15 opportunities (a range of 0% to 20% with a mean of 7%). The percent of successful transitions in baseline was low and stable. Increases in Rob's responses to low-p requests (i.e., a request to move from one area or activity to another) occurred after high-p request sequences were delivered by the interventionist (a range of 80% to 100% with a mean of 94%) across blocks of five opportunities. Responses to requests to transition remained high throughout the high-p sequence condition. Responses to high-p requests for Rob averaged 97% across all blocks of five and ranged from 87% to 100%.

Increases in successful transitions also occurred when preferred item as a distractor was delivered. Rob's responses to requests to transition when using the preferred item as a distractor ranged from 60% to 100% with a mean of 91%. Although the mean percent of successful transitions was high when the distractor intervention was delivered, some variability was displayed in the last three data points.

A secondary analysis of Rob's intervention data was conducted to determine if noncompliance might be attributable to a specific transition. That is, when noncompliance occurred was it associated with one or two specific transitions. The analysis found noncompliance to be evenly distributed across transitions for either intervention (i.e., HPR or Distractor). In fact, no two occurrences of Rob's noncompliance occurred across any one transition.

Travis

Figure 3 presents a graphic representation of Travis' performance. During baseline, Travis successfully completed a mean of 10% of transitions across all baseline sessions (range 0% to 40% of opportunities). The percent of successful transitions was variable but low. Similar to Rob, upon the implementation of the high-p request sequences, Travis' responding to low-p requests (i.e., requests to transition) increased to a mean of 72.5% (ranging from 40% to 100%) of opportunities. Responses to high-p requests were high and stable ranging from 87% to 100% with a mean of 97% of the opportunities. Increases in successful transitions also occurred when the preferred item as a distractor was implemented. Travis' responses to low-p requests in this condition ranged from 60% to 100% and averaged 87.5% of the opportunities. Upon a return to baseline, Travis' responding decreased to a mean of 26.6% (ranging from 20% to 40%) of the opportunities. When the distractor intervention was implemented alone, Travis' responding t o requests to transition increased to an average of 86.6% of opportunities.

A secondary analysis was conducted to determine if the noncompliance during intervention was associated with any specific transition. As with Rob, noncompliance for both interventions were evenly distributed across transitions. When the interventions were examined individually, no two occurrences of noncompliance occurred during any one transition. When all transitions were examined as a whole, noncompliance was evenly distributed across all transitions (i.e., 2 per transition).

Social Validity

Surveys were given to four of the service personnel, including the interventionist, who worked with Rob. Teachers gave an average rating of 4.75 (excellent) to the question, "To what extent was the intervention successful?". Questions regarding the successfulness of individual interventions yielded a mean rating of 4.5 for both the distractor and the high-p sequence. When asked, "would you use the [intervention] in the future?," all participants answered yes for both interventions. Finally, when given a choice between the two interventions, three of the four participants (including the interventionist) selected high-p request sequences as the intervention they would use in the future.

Results of the social validity measures for Travis are displayed in Figure 4. Each probe constituted 4 hours of classroom observation. Generally, the results suggest that the interventionist chose to use one of the proactive interventions rather than no intervention during 13 of 15 transition opportunities. Of these 13 opportunities, 9 involved the use of high-probability request sequences while 4 involved the use of preferred item as a distractor.

Discussion

The results of this study indicate that the two interventions, high-p sequences and preferred item as a distractor, were effective in increasing the percent of successful transitions for two young students with behavior problems. Additionally, results from the social validity survey of Rob's teachers and the direct observations conducted of Travis' teachers indicate the use of these two interventions are socially valid and, to that end, practical for teachers to use on a daily basis in their classroom. This study replicates and extends the literature in two separate but distinct ways. First, the effectiveness of these two antecedent-based interventions have been empirically validated. Second, steps were taken to measure the social validity of these interventions by teachers.

As a result of the low percent of successful transitions prior to intervention, the amount of time the paraprofessional or special education teacher spent attempting to get the two students off the floor and to their destination had drastically limited the amount of time the students were able to participate in given activities. In addition, the extreme noncompliance exhibited by the students during trasitions, if continued, would have severely limited their access to various environments. As the students grew older and larger, the educators working with these students simply could not continue to physically prompt (i.e., move) the students to their next activity without risking injury to themselves. Results of the current investigation suggest that a practitioner could use either of the interventions to increase successful transitions, thereby generating additional time that could be spent in educational activities and reducing the liklihood of injury. What is not clear are the effects of these two intervent ions when used alone. Clearly, a limitation to this study is the use of an alternating treatments design and the possibility of multi-elements treatment interference. In particular for Rob, since both interventions yielded similar results and a reversal was not conducted, we are unable to state conclusively that one is more effective than the other or comment on the effectiveness of one intervention used without the other. However, the use of high-p request sequences has been well documented to increase responding among young children (Davis et al., 1992, 1994; Ducharme & Worling, 1994; Singer, Singer, & Horner, 1987). Given the literature in this area, we can state with relative assurance that high-p requests alone are an effective antecedent-based strategy that increase responding to requests. Conversely, there are few studies that examine the use of the antecedent strategy of preferred item as a distractor. The results of this study are encouraging regarding the use of this strategy to increase responding to requests in a transition setting; however, we cannot speak to the effectiveness of this strategy used in isolation on a daily basis with Rob. That is, if this strategy were the only intervention used by the interventionist, would the same strength of effectiveness be the result?

For Travis, we attempted to validate this strategy in isolation and control for multi-element interference. After returning to baseline and implementing a preferred item as a distractor only phase, Travis' appropriate transitions increased. This condition, although not conclusive, supports the use of a distractor in isolation. Future investigations should continue to examine the successful use of these two interventions and control for multi-element interference by including within subject reversals or conditions that either maintain a baseline comparison or test one intervention in isolation.

The second area in which this study extends the literature is that of obtaining some measure of social validity. For Rob, we attempted to obtain some measure of social validity by asking the service providers (i.e., teachers, paraprofessionals, and speech pathologist) to rate the effectiveness of the interventions. Although the service providers rated both interventions as successful, it is interesting to note that all of the service providers in a forced choice rated high-p request sequences as the intervention that they would most likely use in the future. This is surprising given the amount of effort it takes to deliver the high-p request sequences compared to the preferred item as a distractor. To deliver the distractor, the interventionist delivers one prompt; to deliver the high-p request sequence, the interventionist must deliver at least three prompts prior to the delivery of the request to transition (i.e., low-p request). The preferred item as a distractor intervention appears to require the least a mount of interventionist effort.

In Travis' case, we attempted to obtain observational data to determine which intervention strategy was more likely to be maintained. One month after the study had concluded in Travis' classroom, a new observer (who was not previously associated with the provision of technical assistance) was sent in to observe the classroom under the auspices of a college "class assignment." The information validated the continued use of the two interventions after the end of the study. This information suggests that while practitioners might state that they would use the interventions in the future, in the case of these two interventions with this practitioner, she actually did continue both interventions. Results of the current investigation suggest that teachers maintain their implementation of antecedent-based strategies after the removal of technical assistance associated with the antecedent-based procedural implementation. Future investigations might address the influence that this type of activity and the teacher's av ailability might have on the selection and use of each of the two intervention strategies addressed in the current study.

Future research should compare the use of multi-component and single component interventions using dependent measures that not only address student outcomes but also "instructional efficiency" from a teacher's perspective. Most intervention studies presume that the interventionist is highly motivated to implement the intervention procedure being examined. Offering teachers choices among viable procedures may allow teachers to better match an instructional procedure with their teaching style. Little is known about the effect of choice-making on teacher performance. Future investigations should focus on identifying what variables influence whether or not teachers use or continue to use specific interventions.

This study continues to support high-probability request research and extends the preferred item as a distractor research by demonstrating that a distractor or high-p requests can be used to increase a student's responsiveness to requests during undesired events and activities that he typically tries to escape or avoid. These procedures will offer teachers and parents an easy, efficient, and antecedent way to increase positive responses to requests during transitions (i.e., moving from one area or activity to another). Finally, and most importantly, the use of these interventions might increase the child's probability of independent functioning and successful integration into the classroom and community.

References

Carta, J.J., Atwater, J.B., Schwartz, I.S., & Miller, P.A. (1990). Applications of ecobehavioral analysis in the study of transitions across early education settings. Education and Treatment of Children, 13, 298-316.

Davis, C. A., Brady, M. P., Hamilton, R., McEvoy, M. A., Williams, R. E. (1994). Effects of high-probability requests on the social interactions of young children with severe disabilities. Journal of Applied Behavior Analysis, 27, 619.637.

Davis, C. A., Brady, M., Williams, R, & Hamilton, R. (1992). Effects of high-probability requests on the acquisition and generalization of responding to requests in young children with behavior disorders. Journal of Applied Behavior Analysis, 25, 905-916.

Davis, C. A. & Reichle, J. (1996). Variant and invariant high-probability requests: Increasing appropriate behaviors in children with emotional-behavioral disorders. Journal of Applied Behavior Analysis, 29, 471-481.

Ducharme, J., & Worling, D. (1994). Behavioral momentum and stimulus fading in the acquisition and maintenance of child compliance in the home. Journal of Applied Behavior Analysis, 27, 639-647.

Horner, R., Day, H. M., Sprague, J., O'Brien, M., & Heathfield, L. T. (1991). Interspersed requests: A nonaversive procedure for reducing aggression and self-injury during instruction. Journal of Applied Behavior Analysis, 24, 265-278.

Mace, F., & Belfiore, P. (1990). Behavioral momentum in the treatment of escape-motivated stereotypy. Journal of the Applied Behavior Analysis, 23, 507-514.

Mace, F., Hock, M., Lalli, J., West, B., Belfiore, P., Pinter, E., & Brown, D. (1988). Behavioral momentum in the treatment of noncompliance. Journal of Applied Behavior Analysis, 21, 123-141.

Nevin, J. A., Mandell, C., & Atak, J. (1983). The analysis of behavioral momentum. Journal of the Experimental Analysis of Behavior, 39, 49-59.

Ostrosky, M. M., Donegan, M. M., & Fowler, S. A. (in press). Facilitating transitions across home, community, work, and school. In A. M. Wetherby, S. F. Warren, & J. Reichle (Eds.), Transitions in prelinguistic communication: Preintentional to intentional and presymbolic to symbolic. Baltimore: Brookes.

Peterson, C., & McConnell, S. (1993). Factors affecting the impact of social interaction skills interventions in early childhood special education. Topics in Early Childhood Special Education, 13, 38-56.

Polloway, E. (1996). Treatment acceptability: Determining appropriate interventions within inclusive classrooms. Intervention in School and Clinic, 31, 133-144.

Reichle, J. (1995). Using preferred items to compete with attention-motivated challenging behavior during bus rides. Unpublished manuscript. University of Minnesota, Minneapolis.

Reiners, T., Wacker, D., Derby, K. M., & Cooper, L. (1995). Relation between parental attributions and the acceptability of behavioral treatments for their child's behavior problems. Behavioral Disorders, 20, 171-178.

Sainato, D. M. (1990). Classroom transitions: Organizing environments to promote independent performance in preschool children with disabilities. Education and Treatment of Children, 13, 288-297.

Sainato, D. M., & Lyon, S. (1983). A descriptive analysis of the requirements for independent performance in handicapped and nonhandicapped preschool classrooms. In P.S. Strain (Chair), Assisting behaviorally handicapped preschoolers in mainstream settings: A report of research from the Early Childhood Research Institute. Symposium presented at the HCEEP/DEC Conference, Washington, DC.

Singer, G. H. S., Singer, J., & Horner, R. H. (1987). Using pretask requests to increase the probability of compliance for students with severe disabilities. Journal of the Association for Persons with Severe Handicaps, 12, 287-291.

Schwartz, I., & Baer, D. (1991). Social validity assessments: Is current practice state of the art? Journal of Applied Behavior Analysis, 24, 189-202.

Wolf, M. (1978). Social validity: The case for subjective measurement. How applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11, 203-214.

[Graph omitted]

Table 1.

High-probability requests implementation example.


Request (high and low-p)          Student Response

"Give me five." High-p            S. gives T. five
"Touch your [body part]." High-p  S. touches body part
"Point to your shirt." High-p     S. points to shirt
"Walk into the classroom." Low-p  S. walks to classroom



Request (high and low-p)          Teacher Response

"Give me five." High-p            T. praises S. response
"Touch your [body part]." High-p  T. praises S. response
"Point to your shirt." High-p     T. praises S. response
"Walk into the classroom." Low-p  T. praises S. response