High-probability Request Research: Moving Beyond Compliance.
Practitioners are continually striving to develop and extend strategies to remediate the social and behavioral limitations of individuals with adaptive deficits. The field has been particularly influenced by a movement toward nonaversive and nonintrusive intervention strategies. One such strategy, the high-probability request sequence, fulfills this need in addition to being an effective strategy to increase compliant responding to requests. The applications of this research, however, have been mainly limited to compliance. This discussion outlines the research to date with the high-probability request sequence and proposes applications of this strategy to areas beyond compliance.
Promoting appropriate behaviors in children and adults with disabilities can be a complex task. Educators and clinicians are continually striving to develop and extend strategies to remediate the social and behavioral limitations (e.g., noncompliance, disruptive behavior, physical aggression, off-task behavior, etc.) of these individuals. For the past several years, the literature in the field of behavioral disorders has demonstrated numerous successful applications of intervention strategies to either increase appropriate responding or decrease inappropriate responding. Such strategies include time-out, contingent reinforcement, differential reinforcement, contingent praise, overcorrection, and physical guidance (see Alberto & Troutman, 1995; Cooper, Heron, & Heward, 1987 for a review of these strategies). Although the literature is replete with successful applications of all of these procedures, several limitations may reduce their effectiveness.
Often these interventions require close physical contact with an individual. For violent, aggressive, or large individuals, use of these procedures may be risky. The efficacy of reinforcing appropriate responding also has limitations if a more powerful reinforcer or richer schedule of reinforcement cannot be found to compete with the reinforcement that is maintaining the inappropriate behavior. Behavior reductive procedures also have been proven successful, but questions often arise regarding the ethics and appropriate implementation of these interventions. The most significant limitation to such contingency-management approaches is that they are reactive, rather than proactive. That is, the intervention takes place after the individual has already demonstrated the inappropriate behavior. Depending on the behavior emitted, this could be problematic, as the performance of a behavior may act as a discriminative stimulus for the continuation of that behavior (Can, Newsom, & Binkoff, 1976).
Consequently, behavioral investigations also have focused on examining antecedent interventions to alter the stimulus conditions that set the occasion for inappropriate behavior. Such strategies applied to the area of noncompliance include altering the rate of commands (Forehand & Scarboro, 1975; Plummer, Baer, & LeBlanc, 1977), altering the type of command (Elrod, 1987; Houlihan & Jones, 1990; Neef, Shafer, Egel, Cataldo, & Parrish, 1983), establishing eye contact (Hamlet, Axelrod, & Kuerschner, 1984; Hornik, 1987), and parent training to improve instructions (Kelley, Embry, & Baer, 1979). A continually developing area of research, based on the concept of a functional analysis of behavior, examines a combination of antecedent manipulations, contingency management, and contextual factors to assess functional relations between challenging behavior and specific environmental events (see Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1984; Lennox & Miltenberger, 1989; O'Neil et al., 1997 for review). Through the systematic assessment procedures employed within a functional analysis, interventionists gain a better understanding of the effects of environment on the occurrence and nonoccurrence of problem behaviors.
In line with such proactive efforts and the emphasis on positive behavioral support (Horner et al., 1990) for individuals with challenging behaviors, researchers have developed a group of operationally and functionally similar antecedent interventions, referred to in the literature as pre-task requests (Singer, Singer, & Homer, 1987), high-probability commands (Mace et al., 1988), high-probability requests (Davis, Brady, Williams, & Hamilton, 1992), interspersed requests (Horner, Day, Sprague, O'Brien, & Heathfield, 1991), or oftentimes, behavioral momentum (Mace et al., 1988; 1990). These interventions have received increased attention, as they have been used as the basis for a proactive strategy to facilitate compliant responding to requests. These strategies are all procedurally similar in that a sequence of simple requests (referred to as high-probability requests) are delivered just prior to delivering a request associated with noncompliant behavior (referred to as a low-probability request). High-probab ility requests have a history of frequent, compliant responding, as responses to such requests have been reinforced in the past. As a result, these responses increase and become part of one's behavioral repertoire. The probability of a response to such requests is high. Low-probability requests, however, have a history of infrequent responding, as the responses to these requests are associated with little or no prior reinforcement or success. Thus, the probability of a response to this type of request is low. The terms high-p request(s), high-p request sequence, or low-p request will be used throughout this discussion.
A growing body of literature has demonstrated the effectiveness of using a sequence of high-p requests to influence compliance to low-p behaviors. The strategy within the applied research to achieve this effect involves delivering the sequence of high-p requests (approximately 3 to 5 high-p requests) immediately prior to a low-p request. Compliance to high-p and low-p requests is followed by the delivery of reinforcement. For example, if an intervention was implemented in an attempt to have a child respond to requests to put a toy away, the command sequence could proceed as follows: "Touch the toy," "Clap your hands," "Pick up the toy," (high-p request sequence), followed by "Put the toy on the shelf" (low-p request). Again, reinforcement would be delivered contingently upon compliant responding to each of these requests. The high-p request sequence establishes a pattern of successful responding and increases the probability of responding to the low-p request. The following is a review of the literature in th e area of the high-p request sequence and a discussion of extensions of this intervention beyond the area of simple compliance to requests.
Review of the Applied Research
To date, much of the applied research with the high-p request sequence has focused on examining its effect on increased compliance to requests and increased engagement in appropriate behaviors. Specifically, interventions have focused on variables such as compliance and appropriate transitioning during classroom activities (Singer et al., 1987), increasing compliance to requests (Davis et al., 1992; Ducharme & Worling, 1994; Houlihan, Jacobson, & Brandon, 1994; Kennedy, Itkonen, & Lindquist, 1995; Mace, Mauro, Boyajian, & Eckert, 1997), decreasing task duration and task latency (Belfiore, Lee, Vargas, & Skinner, 1997; Mace et al., 1988), compliance to a prescribed medication regimen (Harchik & Putzier, 1990), increasing attempts at task completion and decreasing aggressive and self-injurious behaviors (Homer et al., 1991; Mace & Belfiore, 1990; Zarcone, Iwata, Hughes, & Vollmer, 1993; Zarcone, Iwata, Mazaleski, & Smith, 1994), increasing social interactions (Davis, Brady, Hamilton, McEvoy, & Williams, 1994; D avis & Reichle, 1996), and increasing communication behavior (Sanchez-Fort, Brady, & Davis, 1995).
Figure 1 and Table 1 include a review of key applied high-p request research. Figure 1 provides a graphic, organizational representation of the progression of high-p request research, from the originating framework and early applications, to current extensions of the procedure. As some of the applied research overlaps into different areas discussed in this review (e.g., methodological extensions, challenging behaviors, generalization and maintenance, social skills, communication skills and academics), note that some investigations are cited more than once in the figure. Table 1, arranged chronologically from 1987 to the present, highlights the participants, setting, dependent variable and measurement, and results of these key investigations.
Early Applications of the High-p Request Sequence
In order to induce generalized compliant responding, Engelmann and Colvin (1983) proposed an antecedent strategy, the "hard task" procedure, that involved delivering a series of easy and familiar requests immediately prior to the presentation of a task having a low likelihood of compliance. The basis for this procedure is that altering the context in which a task occurs influences the likelihood of compliance with the task. The requests presented should be delivered in rapid succession, require a short response time, be followed by praise, and be delivered immediately prior to the presentation of a task having a low probability of completion. Delivering high and low probability tasks in a series requires the learner to fulfill the same compliance requirements for all of the requests. Referring to their procedure as "pre-task requesting," Singer et al. (1987) implemented the Engelmann and Colvin technique with four elementary school children with moderate to severe disabilities and a history of noncompliance t o teacher requests and aggressive behaviors (see Table 1). The purpose of the investigation was to facilitate appropriate transitions between activities in their classroom. Pre-task requesting involved delivering an individualized set of short, easy, and familiar requests (i.e., "Give me five," "Shake hands," "Say your name") in rapid succession immediately prior to a request to transition (e.g. "Go to group now"). Compliance to each request was followed by verbal praise. The results of this investigation indicate that the use of pre-task requesting increased compliance to requests when the classroom teacher requested the students to return to work after recess.
The use of this strategy was extended by Harchik and Putzier (1990) to increase compliance to a medication regimen by a woman with severe mental retardation and a seizure disorder. The participant would either refuse to take her medication, spit it out, or vomit after presented with the medication. The high-p request sequence was incorporated into the participant's daily medication routine, where residential staff presented a series of requests immediately prior to the low-p request to take her medication. Compliant responses to all requests were reinforced with tokens which could be traded in at a later time for items or activities. Use of the high-p request sequence resulted in a substantial increase in taking the medication and a decrease in vomiting or spitting out the medication. These results were maintained at a six month follow-up.
Methodological Extensions of High-p Request Research
Mace et al. (1988) elegantly conducted a series of five studies implementing the "high-p command sequence" (see Table 1 and Figure 1). Four adult males with mental retardation, living in a residential setting, had a history of noncompliance to requests or were slow to initiate or complete tasks. The high-p command sequence was implemented prior to low-p "do" (e.g., "Please put your lunch box away.") and "don't" (e.g., "Please don't leave your lunch box on the table.") requests (see Neef et al., 1983 for a review of compliance training with "do" and "don't" requests). The results from the five investigations indicate the following results: the high-p command sequence (a) effectively increased compliance to low-p requests, (b) increased compliance to low-p requests when compared to a procedure that simply provided the participants with social attention prior to a low-p request (e.g., neutral statements such as "I'm going to wash my car today."), (c) increased compliance when the interprompt time (i.e., the duration of time between reinforcing compliance to the last high-p command in a sequence and the delivery of a low-p command) was 5 s vs. 2 0 s, (d) decreased latency to initiate tasks (such as household chores), and (e) decreased the duration of tasks (i.e., showering and dressing) when compared with verbal prompting and a contingency management procedure.
The Mace et al. (1988) investigation has prompted several direct and systematic replications of the research (see Figure 1). Houlihan et al. (1994) replicated the third experiment from Mace et al. investigating varied interprompt times of 5s and 20s, with a young boy with autism who occasionally responded to verbal requests. Sessions were conducted in his special-needs preschool classroom during work time and structured play time. During each session, the investigator prompted the child to choose a work or play activity, and requests were issued during the course of the activity. The high-p request sequence was issued either 5 s or 20 s before the delivery of a low-p request. The results indicate that compliant responding to low-p requests increased when a 5 s inter-prompt time was in effect for both work and play times. No improvements in compliance, when compared to baseline, were noted when the 20 s interprompt time was in effect. These results are consistent with the Mace et al. investigation and suggest that the effectiveness of the high-p request sequence depends on the temporal contiguity between the high-p request sequence and the low-p request. A shorter interprompt time results in a greater increase in compliant responding to requests.
In another extension of Mace et al. (1988), Kennedy et al. (1995) compared the use of high-p requests with social comments (similar to the social attention control procedure utilized with Mace et al.), on compliant responding to requests. Two students with severe disabilities and histories of noncompliance, participated in the investigation. The examiners compared the effects of the strategy on compliant responding within three different experimental conditions: (a) interspersed requests, (b) social comments (2 s), and (c) social comments (15 s). For the interspersed requests condition, the investigator presented a series of high-p requests before a low-p request. The social comments (2 s) condition was similar to the interspersed requests condition, except that four brief social comments were delivered to the participants, prior to a low-p request. There were no response requirements by the participants for social comments, and low-p requests were delivered 2 s after the last social comment. The social comme nts (15 s) condition was similar to the social comments (2 s) condition, except that the low-p request was issued 15 s after the last social comment. The results indicate that compliant responding to low-p requests increased during both interspersed requests and social comments (2 s) conditions. Unlike the Mace et al. investigation, these results indicate that presenting social comments in close temporal proximity to low-p requests can have effects similar to the high-p request sequence (i.e., increased compliance). The authors caution, however, that the results should not be interpreted to indicate that social comments could replace the high-p request sequence as these results are preliminary.
Applications of the High-p Request Sequence to Other Challenging Behaviors
More recent investigations into the high-p request sequence have moved beyond the realm of basic compliance to noncompliance accompanied by other challenging behaviors (e.g., stereotypic behaviors, aggression, and self-injury) maintained by escape from demands (Horner et al., 1991; Mace & Belfiore, 1990; Zarcone et al., 1993, 1994) (see Figure 1). Homer et al. extended the use of the high-p request sequence to reduce aggression and self-injurious behavior (SIB) during instruction with three adolescents with mental retardation. Each participant had an extensive history of displaying challenging behaviors during instruction (e.g., self-injurious behavior, aggression, or destructive behaviors). The procedure involved presenting tasks assessed as easy and hard for each participant. Easy tasks, defined as those tasks consistently and correctly performed, included pouring water, following a "do this" request, and putting on a t-shirt. Hard tasks, those seldom performed without assistance, involved sorting silverwar e, following two-step instructions, and putting on underwear. The investigator observed the occurrence of aggression and self-injury during (a) easy trials, (b) hard trials, and (c) hard trials accompanied by interspersed requests. Additionally, the investigators measured the participants' attempts to perform the task. The intervention used by Horner and his colleagues consisted of interspersing a series of short and simple high-p requests following any indication of resistance by the participants during hard tasks. This investigation differed from other studies on high-p requests as the high-p request sequence was delivered after approximately every third instructional trial, as opposed to after every instructional trial. In effect, their research attempted to investigate the durability and persistence of responding from a single high-p request sequence. The use of interspersed requests effectively decreased disruptive behaviors during instruction and increased the participants' attempts at responding. A fur ther extension involved replicating the procedures with novel trainers and tasks. Lower occurrences of disruptive behaviors and higher rates of attempts were observed with all participants under these novel conditions.
Mace and Belfiore (1990) investigated the effectiveness of the high-p command sequence to reduce stereotypic behavior negatively reinforced by escape from task-related demands in a woman with mental retardation. A descriptive analysis of the behavior during uncontrolled conditions revealed that the rate of stereotypic behaviors was higher when demands were placed upon the participant, and lowest during a no demand situation when neutral comments (e.g., "It might rain today.") were presented to the participant. During the intervention, the high-p request sequence was delivered prior to a task-related request. The participant was not allowed to escape from the task if she emitted stereotypic behaviors. The results indicate that the rate of stereotypic responding was reduced and collateral compliance to demand situations substantially increased.
Zarcone et al. (1993) proposed that the high-p command sequence does not include a specific procedure to interrupt the escape contingency maintaining the inappropriate behavior. The effectiveness of using the high-p request sequence is in question when individuals attempt to escape from an instructional situation. As previously indicated, the participant in the Mace and Belfiore (1990) investigation was not allowed to escape from the task contingent upon stereotypic behaviors. Therefore, the reduction in escape behavior may have likely resulted from escape extinction. Thus, Zarcone et al. conducted a systematic replication of the Mace and Belfiore investigation, using the high-p command procedure proposed by Mace et al. (1988) with and without an extinction procedure, to treat self-injurious behavior (SIB) maintained by escape (see Table 1). The participant, a 33-year-old woman with mental retardation, had a history of escape-motivated SIB. The investigators measured latency to the occurrence of SIB and compl iance to instructions during three instructional conditions: (a) the high-p request instructional sequence alone, (b) the high-p request instructional sequence combined with escape extinction, and (c) escape extinction alone. Under the high-p request instructional sequence condition, latency to SIB decreased and little improvement was observed for compliant responding. When the high-p request instructional sequence was combined with escape extinction, latency to SIB increased and compliance to instruction was greatest. When escape extinction was used alone, latency to SIB and compliant responding were also high. Zarcone and colleagues suggest, however, that the escape extinction procedure may have only been responsible for the reduction in SIB, whereas the high-p request sequence facilitated increases in compliance.
In a subsequent investigation, Zarcone et al. (1994) more fully examined the effects of a high-p instructional sequence on compliant responding to requests and escape-maintained behavior (see Table 1). Two adult males with mental retardation served as participants for the investigation; both had a history of escape-motivated SIB and exhibited a variety of other disruptive behaviors. Data were collected during all experimental conditions where the investigator engaged in activities with each participant. During the high-p instructional sequence phase, high-p instructions were delivered and if escape behavior occurred, the investigator terminated the instructional trial. During the high-p instructional sequence plus escape extinction condition, the investigator physically guided the participant through the task, and continued with the session if any SIB occurred. The results indicate that when the high-p instructional sequence was implemented without escape extinction, the rates of SIB increased and compliant r esponding decreased. When the high-p instructional sequence plus extinction condition was implemented, rates of SIB decreased, and compliant responding increased. The findings of the investigations by Zarcone and her colleagues (1993, 1994) suggest that escape extinction may be a significant component of treatment when escape behavior accompanies noncompliance.
As Zarcone et al. (1993, 1994) reported marginal improvements in compliant responding in the presence of escape-motivated behavior with the use of the high-p request sequence alone, Mace et al. (1997) propose that the effectiveness of the high-p request sequence must be increased as it is contraindicated in highly aggressive clients (as those discussed in Zarcone et al., 1993, 1994). Mace et al. propose that the effectiveness of the intervention may likely be improved with the use of "higher quality reinforcers" provided contingently upon compliance to high-p requests. Typically, praise is used as reinforcement for compliance to requests. However, the effectiveness of a reinforcer is determined by its effect on the target behavior. Therefore, it is reasonable to assume that if Zarcone et al. (1993, 1994) obtained marginal gains in compliance, verbal praise was a poor reinforcer for the participants in those studies. Thus, Mace et al. propose that by varying the quality of reinforcement for compliance to reque sts, greater treatment gains may be achieved. Mace et al. implemented the high-p request sequence and evaluated its effectiveness under three different reinforcement conditions (praise, food, and praise plus food). The results of the investigation indicate that (a) using food as a reinforcer to compliance to high-p instructions produced greater compliance, and (b) compliance to successive low-p instructions persisted more under conditions where food (versus praise) was presented contingently upon compliance to high-p instructions. It can be argued that "quality" of reinforcers is an irrelevant point, as a stimulus presented after the occurrence of a response either increases or decreases the future probability of responding. If praise, food, or a combination thereof does or does not increase the future occurrence of a response, it either is or is not a reinforcer under given conditions. The Mace et al. investigation, however, presents valuable implications for future research on carefully selecting effective reinforcers (such as through a reinforcer assessment) and varying reinforcers with high-p request research.
Generalization and Maintenance of Responding
The use of multiple trainers on the acquisition and generalization of appropriate responding to low-p requests has also been examined (see Figure 1). Davis et al. (1992) worked with two boys with behavior disorders and severe disabilities in their classroom setting. During the course of the study, multiple trainers delivered low-p requests. During intervention, the high-p command sequence was implemented, and the use of a multiple baseline design across trainers allowed evaluation of the generalization of appropriate responding to each trainer. Results indicate that generalized responding across trainers occurred, even by trainers who had not previously implemented the high-p request sequence with a participant, and appropriate responding was maintained during follow-up sessions.
Although the use of a high-p request sequence has been effective for increasing the occurrence of compliant responding, Ducharme and Worling (1994) expressed concerns for the durability of compliant responding, as much of the research has shown that compliance to low-p requests typically returns to baseline levels when the high-p request sequence is withdrawn (Homer et al., 1991; Mace et al., 1988; Singer et al., 1987). This concern prompted their investigation on fading the high-p request sequence, yet maintaining the same high level of compliance typically obtained during intervention. The participants for their investigation were a 5-year-old boy and a 15-year-old girl, both with mental retardation. Both children had been referred to an outpatient treatment center for their noncompliant behavior in the home. The investigators measured the percentage of compliant responding to low-p "do" and "don't" requests (Neef et al,. 1983). Low-p requests were those related to hygiene, social interactions, and dressing , and were delivered by the parents of each participant within the home. Following the second experimental condition using the high-p request sequence, high-p requests were faded over six separate phases, by gradually reducing the number of high-p requests delivered, and gradually increasing the duration of time between the last high-p request in a sequence and the subsequent low-p request. Finally, a maintenance condition replicated baseline conditions. The results indicate that not only was the high-p request sequence successful in increasing compliant responding to requests, but the request sequence could be successfully faded without compromising the compliance gains achieved during treatment conditions. For both participants, high levels of compliance were maintained for 16 weeks after fading the high-p request sequence, providing evidence that stimulus control could be transferred from the high-p request sequence to low-p requests.
Application of the High-p Request Sequence to Social Skills
The applicability of the high-p request sequence to areas beyond compliant responding has been extended to social interactions of young children with severe disabilities (Davis et al., 1994) (see Figure 1). The participants in this investigation were three boys with severe disabilities, ranging in age from 5 to 6 years, and diagnosed with autism, mental retardation, and speech handicaps. Eight children without disabilities were also selected for participation in the study as recipients of social interactions by the students with disabilities; four of these children were designated as "training peers" and the remaining as "generalization peers." The training peers were instructed in identifying social interactions made by the students with disabilities. The investigation was conducted in each participant's regular integrated play setting. Low-p requests involved those requiring a target child to extend a social initiation to a training peer without disabilities (e.g., "Max, give the toy to [peer].", "Tim, thro w the ball to [peer].", "Alan, push the car to [peer]."). During the course of the study, the investigator delivered all requests to the children with disabilities and measured peer initiations, peer responses, ongoing interactions, and compliant responses to low-p requests. The results indicate that the use of high-p requests increased the participants' responsiveness to requests to initiate social behavior with peers. Furthermore, an increase in unprompted initiations and extended interactions with training peers and generalization peers, and unprompted interactions in non-training settings were also noted.
Davis and Reichie (1996) conducted a similar study using training peers to increase the social interactions between these peers and children with emotional-behavioral disorders. Other researchers (Zarcone et al., 1994) noted that over time, participants complied less often to high-p requests, hypothesizing that when repeatedly pairing high-p requests with low-p requests, high-p requests acquire the same aversive properties as low-p requests. Consequently, the frequency of compliance decreases. Thus, by varying the order of high-p requests, this problem may be avoided. In this investigation, the children with disabilities extended their social interactions to other recipients (e.g., "Peter [target peer], give the pizza to Sara [social peer]."). Social initiations were comparable to compliant responses to requests. However, as a methodological extension of the current research in high-p requests, the investigators compared the difference in responding to low-p requests when high-p requests were delivered in an invariant sequence (i.e., high-p requests delivered in the same order) or variant sequence (i.e., high-p requests delivered in a different sequence). As in the Davis et al. (1994) study, sessions were conducted during play sessions. With all children, greater compliance gains were observed when high-p requests were presented in a variant sequence.
Application of the High-p Request Sequence to Communication Skills
Researchers have extended the use of the high-p request sequence into a teaching procedure to increase the communication behavior of children with speech and language impairments (Sanchez-Fort et al., 1995) (see Figure 1). The first participant was an 8-year-old girl with severe disabilities, who communicated though pointing, making sounds, and manual signs. The second participant, a 4-year-old girl with moderate disabilities, had speech that was difficult to understand, and she used some manual signing upon request. The emission of low-p communication behavior was the dependent variable for each child. Multiple trainers implemented the teaching procedure with each child. Low-p requests included words such as "help," "water," "music," "more," and " want." Furthermore, the investigators assessed each child's responses to untrained words such as "play," "drink," and "restroom." The first participant was required to present the manual sign and vocalize each word. The second participant could either manually sign or vocalize each response.
During baseline, a trainer presented a verbal cue requesting each child to emit the target word while also presenting a model. During treatment conditions, the same procedure as in baseline remained in effect, but the high-p request sequence was issued immediately prior to the delivery of low-p requests. The investigators also observed and measured responding to untrained words in order to assess generalization effects. The high-p request sequence was never applied to the untrained words. The results of the investigation indicate that using the high-p request sequence resulted in a significant increase in the use of communication behavior. The findings also indicate increases in the use of non-trained words. Anecdotal reports indicate that each child increased communication with words not targeted in the study.
Application of the High-p Request Sequence to Academics
Although investigators have examined the effect of the high-p request sequence during instruction, until recently it had not been extended to academic behavior (Belfiore et al., 1997). Two adolescent females participated in this investigation. Both had been expelled from the regular school setting due to academic and social noncompliance, and were enrolled in a community-based alternative education school. The investigators assessed the preference of each participant for multiple-digit (e.g., 587 x 625) and single-digit (e.g., 8 x 7) multiplication problems. The results of the preference assessment indicated that both participants preferred the single-digit problems. These problems served as high-preference problems (similarly to high-p requests). Low-preference problems were the multiple-digit problems. The intervention consisted of presenting thee single-digit math problems prior to a multiple-digit math problem, and measuring the latency to initiate the low-preference problem. The results indicate that lat ency to initiate the multiple-digit problems decreased when compared to baseline levels. Returns to baseline resulted in lower mean latencies when compared to the initial baseline, and the authors discuss the possibility of repeated practice effects. This study extended the high-p request paradigm into academic behavior by requiring functional paper and pencil type responses, thus permitting additional opportunities to practice tasks through the use of the high-p request sequence. Additionally, no feedback (e.g., verbal praise, tangible reinforcers) was provided to the participants contingently upon task completion. Previous research has involved working one-on-one with an individual to deliver requests and subsequent verbal praise. These procedures have implications for the use of the high-p request sequence in other educational settings where one-on-one attention is neither feasible nor practical.
Implications for Educational and Behavioral Programming
The high-p request sequence has gained popularity within applied and research settings as the intervention has proven itself as an effective, non-aversive, and proactive intervention to increase compliant responding to requests by children and adults. When teachers or trainers implement strategies to facilitate compliant responding in those with whom they work, a variety of educational and behavioral options become available. A child's compliant responding reduces unpleasant scenarios that become inappropriate models for others. Compliant responding facilitates positive interactions between individuals, fostering a relationship conducive to instruction and learning. Most importantly, when a child is compliant, the teacher is free to work on skills that facilitate the acquisition of more functional skills.
The ease of implementing the technique has also been demonstrated in several investigations where teachers, paraprofessionals, direct care staff, and parents have successfully applied the technique in a variety of settings and situations. Undoubtedly, the procedure requires little instruction to implement, suggesting that it can be incorporated easily into basic training curriculums, such as parent training and group home training programs. Because of the ease of implementing the procedure, its utility can be extended readily into work or play activities, or academic time, without disrupting ongoing activities. A significant feature of the intervention is that it does not require removal of an individual from the natural setting and interruption of interactions with peers and the natural environment. Most importantly, because of the nature of the intervention, it can be easily tailored to meet an individual's specific educational or behavioral needs.
A rapidly expanding area of research is focusing on the need to attend to the responsiveness of learners during teaching and training. The design and execution of the high-p request sequence is fitting for this goal. The research discussed in this review calls for further research to extend the utility and effectiveness of the high-p request sequence beyond simple compliance to requests. A review of the existing research with high-p requests indicates that oftentimes the focus of responding has been on increasing simple, isolated requests, outside of a functional context. Although such research has made a significant impact on and contribution to the literature in this area, efforts to make the high-p request strategy effective, functional, and socially valid must move into a wider variety of domains. The high-p request sequence is offered as an additional strategy for interventionists to expand their repertoire of instructional techniques. When combined with other traditional strategies (e.g., task analysis , prompting, errorless learning, corrective feedback, reinforcement procedures, etc.) the high-p request sequence offers promise for instructional programming to increase attempts and responding with a task, and continue interaction and sustained engagement in an activity.
Moving Beyond Compliance
Undoubtedly, researchers have demonstrated remarkable gains in the efficacy of the high-p request sequence when applied to noncompliant behavior. As the inherent nature of the intervention facilitates reciprocal interactions and task engagement it is not unreasonable to assume that the strategy can be easily applied to a variety of areas beyond simple compliance. Researchers are just beginning to expand the utility of this strategy beyond simple compliance to requests (see Davis et al., 1994; Davis & Reichle, 1996; Sanchez-Fort et al., 1995 for review). In these investigations, the participants were not noncompliant with requests, per se, in the sense that they refused to comply with requests. These participants simply did not attempt, initiate, engage in, or sustain interaction with tasks or activities. These investigations thus demonstrate the utility and potential of the high-p request sequence beyond simple compliance to requests; from managing behaviors that interfere with task completion (i.e., verbal r efusal, self-injury, aggression), to teaching new skills, or increasing an individual's involvement with an existing skill.
Future extensions of the high-p request sequence might focus on changing other behavioral dimensions, improving communication and functional and adaptive skills, and facilitating more effective rehabilitation, and improved academic instruction.
To date, high-p request research has primarily focused on increasing the frequency of compliant responding. A natural extension of this focus would be to examine the effects of the strategy on other behavioral dimensions, such as latency, duration, accuracy, topography, and magnitude. The high-p request sequence has already been successfully used to decrease the latency to respond to requests (Mace et al., 1988), yet these results have not been replicated. The implications of the results obtained by Mace et al. prove promising if integrated into training curriculums, such as social skills training, to facilitate the necessity of immediate responding central to reciprocal interactions. Integration into vocational training programs would promote increased task initiation and productivity. The results have implications for the classroom, as well. For example, following teacher directions increases a student's access to more reinforcement and socially rewarding interactions. Facilitating faster task initiations i ncreases the rate of responding and provides additional opportunities for drill and practice. As rate of responding provides information on the proficiency of performance (Cooper et al., 1987), using the high-p request sequence to influence rate of responding would be particularly relevant in academics or vocational skills to ensure fluent responding. As the high-p request sequence has successfully facilitated task engagement through actual compliance to requests (e.g., engagement in tasks) with individuals who would typically refuse any interaction, a natural progression would be to use the high-p request sequence to foster continued task engagement. Again, the benefits to socialization, vocational productivity, and academic engagement are clear. Task engagement would be especially relevant for individuals with attention deficits.
Finally, the process of behavioral shaping (e.g., differentially reinforcing successive approximations to a target behavior) could be easily merged with the high-p request sequence. By integrating the high-p request sequence into the shaping process, each successful response to high-p requests could bring the individual to successively approximate low-p responses. The integration of shaping and the high-p request sequence could naturally be extended to the magnitude and topography of a response. The use of the high-p request sequence could focus on gradually increasing or changing the magnitude or topography of responding in other therapeutic milieus, such as speech and language therapy, recreational therapy, physical therapy, or occupational therapy.
Children who have deficits with expressive and receptive communication skills are certain to encounter daily difficulty and frustration in virtually all aspects of their learning and personal adjustment. The high-p request sequence may be a means to minimize this difficulty and frustration and facilitate communication responses in typical environments, such as home and school, and during naturally occurring routines and activities (e.g., meal preparation, hygiene, grooming, play, leisure). Applying the high-p request sequence could facilitate the reciprocity central to socialization, thus reinforcing and maintaining it in relevant settings. The high-p request sequence has been integrated into a teaching procedure to increase verbal responding in two children with disabilities (Sanchez-Fort et al., 1995). The high-p request sequence would naturally fit with other efforts to improve communication skills. Communication disorders comprise a variety of manifestations. The high-p request strategy could be logically extended into teaching procedures for speech disorders. For example, voice disorders are evident when the loudness or pitch of one's voice is inappropriate. The high-p request sequence could be used to increase or decrease dysfunctional learned speech patterns. The strategy would also be amenable for use with articulation disorders in facilitating the production of more accurate speech sounds. For example, a child may distort speech sounds, producing the word "zleep" for "sleep." By asking the student to perform high-p responses such as other /s/ sounds or words and then requesting the low-p /s/ sound found in the word "sleep," appropriate responding may be likely to persist under these conditions.
Future research should examine the utility of the high-p request sequence when integrated into teaching functional and adaptive skills such as dressing, bathing, leisure activities, vocational skills, and social/communication skills. These skills, designed to make students more independent and less reliant on others for the execution and completion of daily tasks, have not been fully examined within a high-p request sequence. Integration of the high-p request sequence into teaching strategies could focus on developing a chain of responses to facilitate completion of functional activities. For example, a functional skill, such as bathing or hand washing could be broken down into responses (through task analysis) and the high-p request sequence, when integrated into the task sequence, could be used to facilitate engagement and task completion. As discussed with communication skills, it is possible that a high-p request-based procedure could develop into naturally occurring routines and activities, as functiona l and adaptive skills are so frequently demanded within the natural environment.
An additional potential extension of the high-p request sequence is its integration with rehabilitation, such as physical, occupational, and recreational therapy. Extensions of the high-p request into the area of rehabilitation would reflect directly on the social validity of the procedure. Often not only is it necessary to increase the frequency of engagement in these skills, but to use them correctly and functionally. Physical therapists often focus on increasing muscle control, developing muscular function, or engagement in exercise or play programs. Occupational therapists often focus on one's functional participation in an activity or task. The main focus of these professionals is to increase the ability to sustain engagement in an activity, increase the accuracy of performance, or integrate sensory modalities (i.e., touch, movement, vision, etc.). Given the ease with which high-p request sequences facilitate interaction with a task, a logical extension in this area would be its application to rehabilitation.
Educational research (particularly special education) suggests that teachers and researchers should attend more to the responsiveness of learners during instruction. The literature is replete with examples facilitating effectiveness and motivation during instruction while stressing the need for functional skills (Dunlap & Koegel, 1980; McGee, Krantz, & McClannahan, 1985, 1986). The use of the high-p request sequence is another attempt to improve instruction and reduce collateral challenging behaviors. Horner et al. (1991) utilized the intervention during instruction to increase students' attempts to perform difficult tasks. The use of the high-p request sequence would likely promote more effective instruction when used to facilitate engagement in less preferred instructional activities, transition from more to less preferred activities, or engagement in more demanding tasks. The high-p request sequence has already been successfully used as a peer mediated strategy (Davis et al., 1994; Davis & Reichle, 1996). A logical extension of the high-p request sequence as a peer-mediated strategy into academics could involve using the strategy during cooperative learning activities or peer tutoring. Future research with the intervention could investigate the relation between persistence of responding (the tendency for responding to low-p requests after the high-p request sequence) and fluency. Binder (1994) indicates that greater speed of responding results in greater resistance to distraction. As an extension of rate of responding and fluency, the role that accuracy serves in the persistence of behavior is worthy of investigation.
The need for strategies to promote compliant behavior is apparent in the literature. It is time, however, to expand the range of applications of the high-p request sequence beyond the realm of simple compliance into other domains. The high-p request sequence emerged from influences in special education to individualize strategies that maximize effective instruction, appropriate socialization, and community inclusion. The continued limited applications of this powerful intervention, however, stagnate the range of possibilities in providing individuals with disabilities with effective intervention strategies in a variety of areas such as rehabilitation, communication, and academics. This discussion has examined several future applications of the intervention. By expanding the range of outcomes for the high-p request sequence, practitioners can develop functional programming options to improve the educational validity of individuals with disabilities.
Alberto, P. A., & Troutman, A. C. (1995). Applied behavior analysis for teachers. (4th ed.). Columbus, OH: Prentice-Hall
Belfiore, P. J., Lee, D. L., Vargas, A. U., & Skinner, C. H. (1997). Effects of high-preference single-digit mathematics problem completion on multiple-digit mathematics problem performance. Journal of Applied Behavior Analysis, 30, 327-330.
Binder, C. (1994, May). Behavioral fluency: Evolution of a new paradigm. Presentation at the meeting of the Association for Behavior Analysis, Atlanta, GA.
Carr, E. G., Newsome, C. D., & Binkoff, J. A. (1976). Stimulus control of self-destructive behavior in a psychotic child. Journal of Abnormal Child Psychology, 4, 139-153.
Cooper, J. O., Heron, T. E., & Heward, W. L. (1987). Applied Behavior Analysis. Columbus: Merrill.
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. P., Williams, R. E., & Hamilton, R. (1992). Effects of high-probability requests on the acquisition and generalization of responses 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, 471-482.
Ducharme, J. M., & Worling, D. E. (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.
Dunlap, G., & Koegel, R. L. (1980). Motivating autistic children through stimulus variation. Journal of Applied Behavior Analysis, 23, 619-627.
Elrod, M. M. (1987). Children's understanding of indirect requests. Journal of Genetic Psychology, 248, 63-70.
Engelmann, S., & Colvin, G. T. (1983). Generalized compliance training: A direct-instruction program for managing severe behavior problems. TX: Pro-Ed.
Forehand, R. L., & Scarboro, M. E. (1975). An analysis of children's behavior. Journal of Abnormal Child Psychology, 3, 27-31.
Hamlet, C. C., Axelrod, S., & Kuerschner, S. (1984). Eye contact as an antecedent to compliant behavior. Journal of Applied Behavior Analysis, 17, 553-557.
Harchik, A. E., & Putzier, V. S. (1990). The use of high-probability requests to increase compliance with instructions to take medication. Journal of the Association for Persons with Severe Handicaps, 15, 40-43.
Horner, R. H., Day, H. M., Sprague, J. it, O'Brien, M., & Heathfield, L. T. (1991). Interspersed requests: A non-aversive procedure for reducing aggression and self-injury during instruction. Journal of Applied Behavior Analysis, 24, 265-278.
Horner, R. H., Dunlap, G., Koegel, R. L., Carr, E. G., Sailor, W., Anderson, J., Albin, R. W., & O'Neill, R. E. (1990). Toward a technology of "nonaversive" behavioral support. Journal of the Association for Persons with Severe Handicaps, 15, 125-132.
Hornik, J. (1987). The effect of touch and gaze upon compliance and interest of intervention. Journal of Social Psychology, 127, 681-683.
Houlihan, D., Jacobson, L., & Brandon, P. K. (1994). Replication of a high-probability request sequence with varied interprompt times in a preschool setting. Journal of Applied Behavior Analysis, 27, 737-738.
Houlihan, D., & Jones, R.N. (1990). Exploring the reinforcement of compliance with do and don't requests and their side effects: A partial replication and extension. Psychological Reports, 67, 439-448.
Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1982). Toward a functional analysis of self-injury. Analysis and Intervention in Developmental Disabilities, 2, 3-20. Reprinted in Journal of Applied Behavior Analysis, 1994, 27, 197-209.
Kelley, M. L., Embry, L. H., & Baer, D. M. (1979). Skills for child management and family support. Behavior Modification, 3, 373-396.
Kennedy, C. H., Itkonen, T., & Lindquist, K (1995). Comparing interspersed requests and social comments as antecedents for increasing student compliance. Journal of Applied Behavior Analysis, is, 28, 97-98.
Lennox, D. B., & Miltenberger, R. G. (1989). Conducting a functional assessment of problem behavior in applied settings. Journal of the Association for Persons with Severe Handicaps, 14, 304-311.
Mace, F. C., & Belfiore, P. (1990). Behavioral momentum in the treatment of escape motivated stereotypy. Journal of Applied Behavior Analysis, 23, 507-514.
Mace, F. C., Hock, M. L., Lalli, J. S., West, B. J., Belfiore, P., Pinter, E., & Brown, D. K. (1988). Behavioral momentum in the treatment of noncompliance. Journal of Applied Behavior Analysis, 21, 123-141.
Mace, F. C., Lalli, J. S., Shea, M. C., Lalli, E. P., West, B. J., Roberts, M., & Nevin, J. A. (1990). The momentum of human behavior in a natural setting. Journal of the Experimental Analysis of Behavior, 54, 163-172.
Mace, F. C., Mauro, B. C., Boyajian, A. E., & Eckert, T. L. (1997). Effects of reinforcer quality on behavioral momentum: Coordinated applied and basic research. Journal of Applied Behavior Analysis, 30, 1-20.
McGee, J. J., Krantz, P. J., & McClannahan, L. E. (1985). The facilitative effects of incidental teaching on preposition use by autistic children. Journal of Applied Behavior Analysis, 18, 17-32.
McGee, J. J., Krantz, P. J., & McClannahan, L. E. (1986). An extension of incidental teaching procedures to reading instruction for autistic children. Journal of Applied Behavior Analysis, 19, 147-158.
Neef, N. A., Shafer, M. S., Egel, A. L., Cataldo, M. F., & Parrish, J. M. (1983). The class specific effects of compliance training with "do" and "don't" requests: Analogue analysis and classroom applications. Journal of Applied Behavior Analysis, 16, 81-99.
O'Neill, R. E., Horner, R. H., Albin, R. W., Sprague, J. R, Storey, K., & Newton, J. S. (1997). Functional assessment and program development for problem behavior. (2nd ed.). Pacific Grove, CA: Brooks/Cole.
Plummer, S., Baer, D., & LeBlanc, J. (1977). Functional consideration in the use of procedural time-out and an effective alternative. Journal of Applied Behavior Analysis, 10, 689-705.
Sanchez-Fort, M. R, Brady, M. P., & Davis, C. A. (1995). Using high-probability requests to increase low-probability communication behavior in young children with severe disabilities. Education and Training in Mental Retardation and Developmental Disabilities, 30 (2), 151-165.
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.
Zarcone, J. R., Iwata, B. A., Hughes, C. E., & Vollmer, T. R. (1993). Momentum versus extinction effects in the treatment of self-injurious escape behavior. Journal of Applied Behavior Analysis is, 26, 135-136.
Zarcone, J. R., Iwata, B. A., Mazaleski J. L., & Smith, R. G. (1994). Momentum and extinction effects on self-injurious escape behavior and noncompliance. Journal of Applied Behavior Analysis, 27, 649-658.
Table 1 Summary of Applied High-probability Request Research Study Participants/Setting Singer, Singer, 4 students (7-10 years) and Horner (1987) with moderate to severe disabilities and aggressive behaviors, after recess in special education classroom Mace, Hock, Lalli, Adult males with West, Belfiore, Pinter, moderate to severe and Brown (1988) mental retardation in community residential facility Harchik and Putzier 23-year-old female (1990) with mental retardation and seizure disorder in community residential facility Mace and Belfiore 38-year-old female with (1990) severe disabilities and history of escape-motivated stereotypic behavior in community residential facility Homer, Day, Sprague, 4 adolescents with O'Brien, and Healthfield moderate/severe mental (1991) retardation and challenging behavior in community residential facility Study Dependent Variable & Measurement Singer, Singer, Frequency of compliance and Horner (1987) to low-p teacher requests to transition Mace, Hock, Lalli, Frequency count of compliant West, Belfiore, Pinter, responding to low-p "do" and Brown (1988) and "don't requests; compliant responding to longer and shorter interprompt times; response latency and duration Harchik and Putzier Count of number of times (1990) took medication, and spit out or vomited medication Mace and Belfiore Compliance to low-p requests (1990) and occurrences of stereotypic behaviors Homer, Day, Sprague, Occurrences of aggression or O'Brien, and Healthfield self-injury during easy and (1991) hard tasks; attempts to complete easy and hard tasks Study Results Singer, Singer, Increased occurrence of and Horner (1987) compliance Mace, Hock, Lalli, Increase in frequency of West, Belfiore, Pinter, compliance to "do" and "don't" and Brown (1988) commands; increased compliance during shorter interprompt times; shorter response latency and duration Harchik and Putzier Increase in number of times (1990) participant took medication and decrease in number of times spit out or vomited medication; response maintenance after 6 months Mace and Belfiore Increases in compliance to (1990) low-p requests and decreases in stereotypic behaviors Homer, Day, Sprague, Lower occurrences of aggression O'Brien, and Healthfield and SIB during easy an hard + (1991) interspersed tasks; higher occurrences of attempts to perform tasks during easy and hard + interspersed phases; similar responding with new trainers and new bard tasks Davis, Brady, Williams, 7-year-old boy with Down and Hamilton (1992) Syndrome 5-year-old boy with autism and mental retardation; both with aggressive, destructive, and stereotypic behaviors; enrolled in life skills program Zarcone, Iwata, Hughes, 33-year-old female and Vollmer (1993) with profound mental retardation and history of escape-motivated SIB Davis, Brady, Hamilton, 3 boys with autism, McEvoy, and Williams mental retardation, (1994) and speech disorders; integrated training and non-training play settings Ducharme and Worling 5-year-old boy with mental (1994) retardation, 15-year-old girl with severe disabilities living with parents Houlihan, Jacobson, 5-year-old boy with autism, and Brandon (1994) in special education preschool classroom Davis, Brady, Williams, Percentage of responding to and Hamilton (1992) low-p requests without presence of challenging behaviors Zarcone, Iwata, Hughes, Count of compliant responding and Vollmer (1993) to low-p requests and SIB; latency in minutes to the first occurrence of SIB Davis, Brady, Hamilton, Responses to low-p McEvoy, and Williams requests, social initiations, (1994) social responses, and continued interactions Ducharme and Worling Compliant responses to 'do,, (1994) and "don't" requests; continued compliance when high-p request sequence faded Houlihan, Jacobson, Percentage of compliance to and Brandon (1994) low-p requests when delivered 5s or 20 s after the high-p request sequence Davis, Brady, Williams, Increased responding to low-p and Hamilton (1992) requests, and generalized responding across trainers; response maintenance at 4-weeks Zarcone, Iwata, Hughes, When presented alone, high-p and Vollmer (1993) requests did not reduce SIB; when escape extinction was implemented alone or with the high-p request sequence, SIB decreased and compliance increased Davis, Brady, Hamilton, In training settings, increase in McEvoy, and Williams rate of prompted initiations (1994) engagement in interactions, and initiated interactions; in non-training settings, increase in unprompted initiations Ducharme and Worling For Participant 1, compliant (1994) responding maintained for "do" and "don't" requests; For Participant 2, compliance maintained for "do" reuests, and compliance increased when "don't" requests converted to "do" requests Houlihan, Jacobson, Increase in percentage of and Brandon (1994) compliant responding with 5 s interprompt time; no compliance gains with 20 s interprompt time Zarcone, Iwata, 38-year-old and 45-year-old Mazaleski, and males with profound mental Smith (1994) retardation and SIB in state facility Kennedy, Itkonen, 18-year-old male and and Lindquist (1995) 19-year-old female with severe disabilities Sanchez-Fort, Brady, 8-year-old and 5-year-old and Davis (1995) girls with communication disorders in integrated public school classroom Davis and Reichle (1996) 2 boys and 2 girls, 4-5 years old with emotional/ behavioral disorders in kindergarten classroom Mace, Mauro, Boyajian, 2 adolescent boys with and Eckert (1997) moderate mental retardation in dorm itory-style room Belfiore, Lee, Vargas 2 adolescent girls in and Skinner (1997) alternative education setting Zarcone, Iwata, Occurrence of SIB and compliance; Mazaleski, and responses per minute of SIB; Smith (1994) all during high-p request sequence and high-p request sequence + escape extinction Kennedy, Itkonen, Occurrence of compliant and Lindquist (1995) responding to low-p requests delivered 2 s or 20 s after the delivery of social comments vs. low-p requests preceded by high-p request sequence Sanchez-Fort, Brady, Emission of low-p training and Davis (1995) words; emission of low-p generalization words Davis and Reichle (1996) Compliant responses to low-p requests preceded by variant and invariant high-p request sequence; low-p requests were requests to initiate social bid to a peer Mace, Mauro, Boyajian, Cumulative frequency of compliance and Eckert (1997) to low-p requests with varying reinforcer quality" (e.g., praise, food, praise + food) Belfiore, Lee, Vargas Mean latency (in seconds) and Skinner (1997) to initiate multiple digit multiplication problems Zarcone, Iwata, During high-p request sequence Mazaleski, and phases, SIB increased, compliance Smith (1994) decreased, and escape behavior increased; during high-p request sequence + escape extinction phases, SIB decreased and compliance increased Kennedy, Itkonen, High-p request sequence increased and Lindquist (1995) compliant responding to requests; compliant responding to low-p requests delivered 2 s after social comments was greater than compliant responding to low-p requests delivered 20 s after social comments Sanchez-Fort, Brady, Increase in emission of low-p words and Davis (1995) for each participant; increase in emission of generalization words for second participant Davis and Reichle (1996) Increase in social bid to peer with use of variant high-p request sequence; decrease in social bids with invariant high-p request sequence Mace, Mauro, Boyajian, Increase in cumulative frequency of and Eckert (1997) compliant responding to low-p requests with "higher quality reinforcers" Belfiore, Lee, Vargas Decrease in latency to initiate low and Skinner (1997) preference multiplication problems
|Printer friendly Cite/link Email Feedback|
|Publication:||Education & Treatment of Children|
|Date:||Nov 1, 1999|
|Previous Article:||The Instructional Assistants Program: A Potential Entry Point for Behavior Analysis in Education.|
|Next Article:||Increasing the Social Interactions of Preschool Children with Disabilities During Mealtimes: The Effects of an Interactive Placemat Game.|