The effects of rationales, differential reinforcement, and a guided compliance procedure to increase compliance among preschool children.
Previous research suggests that rationales, or statements describing why a child should comply with a caregiver-delivered instruction, are ineffective at increasing compliance. In the current study, we compared the effects of rationales to a differential reinforcement procedure and a guided compliance procedure. The results indicated that rationales and differential reinforcement were ineffective for both participants; the guided compliance procedure increased compliance for these children. Problem behavior varied within and across conditions, but was generally higher during the rationales and guided compliance conditions.
Keywords: Compliance, extinction, guided compliance training, noncompliance, preschool children, rationales
Noncompliance is defined as failure to follow a specific caregiver-delivered instruction (Forehand, Gardner, & Roberts, 1978; Kalb & Loeber, 2003). Although estimates of the prevalence of noncompliance vary, Kalb and Loeber (2003) suggest that 25 to 65% of children are noncompliant. In addition, noncompliance in childhood is correlated with some adolescent and adult psychiatric diagnoses such as conduct disorder and oppositional defiant disorder (Keenan, Shaw, Delliquadri, Giovannelli, & Walsh, 1988).
Interventions for noncompliance consist of two broad categories: consequence-based and antecedent-based interventions. Examples of consequence-based interventions for noncompliance include guided compliance procedures and differential reinforcement. Guided compliance procedures involve progression from less (e.g., verbal) to more (e.g., physical guidance) intrusive prompts by a caregiver to achieve compliance. Wilder and Atwell (2006) found that 3-step prompting, in which a verbal, then model, then physical prompt is used, increased compliance among 4 of 6 preschool children they evaluated. Although the behavioral mechanism responsible for the effects of guided compliance procedures probably varies by application and participant, escape extinction likely plays a role, assuming the noncompliance is motivated by escaping the requested task. That is, during guided compliance procedures, the child is physically guided to comply with the instruction if necessary. Engaging in behaviors other than compliance, which previously allowed the child to avoid or escape the task presented in the instruction, is no longer reinforced; the process of extinction increases compliance.
Differential reinforcement of compliance, or access to preferred items contingent upon compliance, has also been shown to be effective to increase compliance, at least among some children. For example, in one of the first applied evaluations of reinforcement procedures, Zeil-berger, Sampen, and Sloane (1968) found that differential reinforcement in the form of the delivery of small tangible items contingent upon appropriate behavior increased compliance in a 4-year-old boy. More recently, Marcus and Vollmer (1995) used differential negative reinforcement in the form of termination of instructions contingent on compliance, to increase compliance by a 5 year-old girl. Wilder, Harris, Reagan, and Rasey (2007) used differential reinforcement in the form of contingent access to coupons that were exchangeable for preferred items and activities to increase compliance among 2 preschool children. Other studies (e.g., Goetz, Holmberg, & LeBlanc, 1975) have also found differential reinforcement to be effective to increase compliance.
Examples of antecedent-based interventions for noncompliance include the high-p instructional sequence, advance notice or a "warning" of an upcoming instruction, and rationales. Although technically a blend of both antecedent and consequence-based procedures, the high-p instructional sequence includes the delivery of a series of instructions with which the child is likely to comply before the delivery of the target, instruction that has a low-probability of compliance. Because the intervention consists of changing the way the instruction is delivered, this strategy is often categorized as an antecedent-based intervention. Mace et al. (1988) evaluated the high-p instructional sequence to increase compliance. At least one of four adults with developmental disabilities participated across five separate experiments. The results showed that compliance improved when the high-p sequence was used.
Warning or providing advance notice of an upcoming instruction is another antecedent-based strategy that has been evaluated to increase compliance. Cote, Thompson, and McKerchar (2005) examined the effects of a warning to increase compliance with instructions to go to the restroom among three preschool children. In this study, a warning statement (e.g., "2 min to potty") was delivered to participants 2 min before an instruction to go to the restroom. The results showed that the warning was ineffective for all three participants. In another study, Wilder, Nicholson, and Allison (2010) found that a 1 min warning was also ineffective to increase compliance among preschool children.
The delivery of rationales, or statements describing why a child should comply with a caregiver-delivered instruction, is yet another antecedent-based intervention for noncompliance. For example, a caregiver might give a rational, such as "Let's get your room cleaned up so no one trips on these toys." Although no data exists on the frequency with which caregivers use rationales, they are commonly heard in day care centers, classrooms, and other early childhood settings. In fact, the use of rationales is recommended by some popular texts on managing child noncompliance (e.g., Kapalka, 2007; McMahon & Forehand, 2003; Nelsen, 2006). To date, however, only one study has empirically examined the effectiveness of rationales on compliance among young children. Wilder, Allison, Nicholson, Abellon, and Saulnier (2010) examined the use of rationales with 6 preschool children. Three participants received the rationale immediately before an instruction was delivered. Another three participants received a rationale immediately after an instruction was delivered. Unfortunately, the authors found that rationales did not increase compliance among any of the participants.
Although Wilder et al. (2010) delivered rationales immediately before or after an instruction, the delivery of rationales at other times may also be common. For example, parents and teachers may deliver rationales some time after a child exhibits noncompliance with an initial instruction or after a child vocally responds to an initial instruction (e.g., says "why," "no," or "I don't want to"). Compliance with an instruction in which a rationale is delivered some time after the instruction or after a vocal response from the child may be more likely because this type of rationale may have been more highly correlated with reinforcement (positive or negative) in the past. Thus, the primary purpose of the current study was to further evaluate the effects of rationales on compliance among preschoolers. In the current study (unlike the previous study on rationales), rationales were delivered 10 s after a child exhibited noncompliance with an initial instruction or vocally responded to an initial instruction.
As described above, a number of consequence-based procedures have been evaluated as interventions for noncompliance. However, despite the existence of this research, few studies have directly compared consequence-based procedures. Therefore, a second purpose of this study was to compare the effects of rationales with two consequence-based interventions: differential reinforcement (i.e., access to preferred items contingent upon compliance) and guided compliance.
Participants and Setting
Craig and Hank, both 4-year-old boys, participated in the study. The participants were typically developing with age-appropriate language skills. Participants attended the same pre-school, and their teachers nominated them as being "noncompliant with instructions," particularly those involving surrendering preferred toys. Sessions were conducted in a small, private therapy room at the preschool. Four-to-eight sessions were conducted per-day, two days per week. A female graduate student, unknown to participants before the study began, served as therapist.
Definitions and Data Collection
Compliance, the dependent variable, was defined as independently initiating or completing the activity requested in the instruction within 10 s. Compliance with the instruction was recorded on each trial, and all sessions consisted of 5 trials. Data on problem behavior, defined as aggression (e.g., hitting, pinching, kicking), property disruption (e.g., throwing toys, hitting objects), whining, or crying were also collected. Compliance and problem behavior were recorded on a trial-by-trial basis. That is, during each trial, after the therapist presented the instruction, the observer recorded compliance and the occurrence of problem behavior (the two were not mutually exclusive). During the rationale condition, compliance was scored if the child handed the toy to the therapist either a) after the initial instruction was delivered, or b) within 10-s of the rationale being delivered (see description of rationale condition below). Data are reported as the percentage of trials in which compliance and problem behavior occurred.
Interobserver agreement (IOA) data were collected during at least 70% of sessions for each participant and was obtained by comparing two observers' records on a trial-by-trial basis. An agreement was defined as both observers recording an instance of either compliance or noncompliance, or, for problem behavior, an instance of problem behavior or no problem behavior, on a given trial. Mean agreement was determined by dividing the number of agreements by the number of agreements plus disagreements and converting the ratio to a percentage. For compliance, mean agreement was 96% (range, 94-100%) and 98% (range, 96-100%) for Craig and Hank, respectively. For problem behavior, mean agreement was 88% (range, 82-100%) and 90% (range, 84-100%) for Craig and Hank, respectively.
Data on independent variable (IV) integrity were collected during at least 30% of sessions for both participants. IV integrity data were collected on the delivery of rationales, the delivery of the model prompt, and the use of hand-over-hand guidance during the guided compliance procedure. IV integrity values were 100% for all sessions during which these data were collected for both participants.
Procedures and Experimental Design
First, one paired stimulus preference assessment (SPA) (Fisher et al., 1992) with edible items and one SPA with tangible items was conducted separately for each participant to identify high and low-preference foods and toys. The most preferred edible items were M&[Ms.sup.TM] and gummy candy for Craig and Hank, respectively. The most preferred toys were a toy motorcycle for both Craig and Hank. Craig's lowest preference toy was a video game; Hank's was a toy dinosaur.
Reversal designs were used to evaluate the effects of rationales, contingent access, and the guided compliance procedure on compliance. Each trial consisted of a pre-instruction period, the instruction, and a post-instruction period. Participants engaged with high-preference toys identified via the preference assessment. During baseline, after the (approximately) 1-min pre-instruction period, the therapist delivered an instruction to "give me the toy". Compliance resulted in the therapist saying, "Thank you," and a low-preference toy (also identified via the preference assessment) was available during the (approximately) 2-min post-instruction period. If the child did not comply with the instruction, the therapist did nothing (i.e., did not remove the high-preference toy) for the remainder of the post-instruction period. No programmed contingencies were in place for problem behavior (i.e., aggression, property destruction, whining, crying) during baseline; it was simply ignored.
During the rationales condition, the therapist presented the instruction as in baseline. If the child did not comply within 10 s of the initial instruction or if the child vocally responded to the initial instruction (e.g., said "why," "no," or "I don't want to"), the therapist randomly delivered 1 of 3 rationales followed by re-presentation of the instruction. The three rationales (including the instruction) were: "Someone might trip on that and get hurt. Give me the__.", "It's time to do another activity. Give me the__.", "Someone else needs to use this room after us so it should be clean. Give me the__." During this condition, a response was scored as compliant if the child handed the toy to the therapist either a) after the initial instruction was delivered, or b) within 10-s of the rationale being delivered. However, the vast majority of compliance observed in this condition occurred after the rationale was delivered. The consequences for compliance and noncompliance in this condition were identical to baseline.
During the contingent-access condition, the therapist presented the instruction as in baseline while holding a small piece of the participant's most preferred edible item in her hand. If the child complied, the therapist said, "Thank you" and immediately gave the edible item to the child. If the participant did not comply with the instruction, the therapist did nothing for the remainder of the 2-min post-instruction period. The contingent access condition was evaluated prior to the guided compliance condition because it was less intrusive (i.e., did not involve physical guidance), and because the teachers and parents of the participants requested that it be attempted first.
During the guided compliance condition, the therapist presented the instruction as in baseline. If the child complied, the therapist said, "Thank you." If the child did not comply after 10 s, the therapist obtained eye contact with the participant, repeated the instruction, and demonstrated giving a toy. If the child did not comply after the demonstration, the therapist repeated the instruction a third time while simultaneously guiding the participant to relinquish the toy. A response was scored as compliant only if the child handed over the toy after the initial instruction. No programmed contingencies were in place for problem behavior (i.e., aggression, property destruction, whining, crying) during any of the treatment conditions described above; problem behavior was simply ignored.
Figure 1 shows the results of the compliance evaluation; both compliance and problem behavior are depicted. Craig's mean percentage of trials with compliance during all baseline sessions was 10% (range, 0-60%). His mean percentage of trials with compliance during all rationale sessions was 29% (range, 0-100%); this compliance occurred after the rationale was delivered, and not immediately after the original request. Craig's mean percentage of trials with compliance during the contingent-access condition was 53% (range, 20-100%). Craig's compliance increased during the guided compliance sessions (M=66%; range, 0-100%). Craig's mean percentage of trials with problem behavior was 6% (range, 0-40%), 21% (range, 0-80%), 7% (range, 0-20%), and 20% (range, 0-80%) during baseline, rationale, contingent-access and guided compliance sessions, respectively.
[FIGURE 1 OMITTED]
Hank's mean percentage of trials with compliance during all baseline sessions was 7% (range, 0-40%). His mean percentage of trials with compliance during rationale sessions was only slightly higher (M=13%; range, 0-60%). Hank's compliance increased during the contingent-access condition (M=52%; range, 20-80%) and during guided compliance sessions (M=56%; range, 20-100%). His mean compliance during the final guided compliance condition was above 90%. Hank's mean percentage of trials with problem behavior was low during baseline sessions (M= 9%; range, 0-40%), rationale sessions (M=8%; range, 0-60%), and contingent-access sessions (M=8%; range, 0-20%). His problem behavior increased during guided compliance sessions (M=24%; range, 0-80%).
The results of this study suggest that rationales delivered 10 s after noncompliance to the initial instruction, or after participants vocally responded to the initial instruction by saying "why," "no," or "I don't want to," were ineffective in increasing compliance. In addition, contingent access to preferred edibles was relatively ineffective at increasing compliance. A guided compliance procedure was necessary to increase compliance to acceptable levels with these two children. Problem behavior varied within and across conditions, but was generally higher during rationale and guided compliance conditions, relative to baseline and contingent access conditions.
The results of this and previous research suggest that delivering rationales is an ineffective method of increasing compliance for many children. The current study extends Wilder et al. (2010) by showing that rationales are ineffective when used after a child asks "why" or replies "I don't want to" to the initial instruction. This was the case despite the fact that the definition of compliance during the rationale sessions was more liberal than during the other conditions in that compliance was recorded if the child complied after the original instruction or within 10 s of the delivery of the rationale. One possible reason for the lack of effectiveness is that no consequence is described as part of the rationale. Rationales might represent what Malott, Whaley, and Malott (1997) describe as "incomplete rules." Rules are "if-then" statements that specify the contingencies involved in behaving in a particular manner. For example, a rule related to compliance might be "If you comply, you will be able to play with your favorite toy." In this example, the consequence for compliance is clearly specified. Rationales, at least as evaluated in this and the previous Wilder et al. (2010) study, may represent incomplete rules in that the consequence for behavior is not clearly described as part of the rationale. Rationales that include a description of a consequence, and if necessary, involve the therapist implementing the consequence described in the rationale, may be more effective. Indeed, if and when rationales are effective, it is possible that these are the conditions which establish their effectiveness. Future research should examine rationales with described consequences (e.g., "Clean up; Dad will be home soon and he'll be very happy to see a clean house" or "Clean up; Dad will be home soon and he won't play with you until your room is clean").
In addition to a lack of a description of the consequences, rationales might also be ineffective because children may have no history of experiencing the consequences some caregivers describe in rationales. That is, a description of the consequences may be necessary but not sufficient. Some history of implementation of the consequences described in the rationale may be necessary to establish rationales as effective.
When (if) rationales are effective, it may be because the child and parent have a particular history of interaction. That is, a child may have historically received positive reinforcement contingent upon compliance with an instruction accompanied by a rationale. Alternatively, child compliance may increase due to negative reinforcement when rationales are used. By complying, children may avoid or escape the aversive consequence that has been applied when an instruction accompanied by a rationale has been delivered in the past. The delivery of positive or negative reinforcement may be more highly correlated with instructions accompanied by a rationale relative to instructions without a rationale. For example, a child might have a strong history of reinforcement, in the form of parent praise, for complying with "Do your homework; your teacher will be impressed that you finished it early", but a weaker history of reinforcement for "Do your homework". Future research could evaluate the role of history of reinforcement in the effectiveness of rationales.
The results of this study also highlight the need for assessment of the variables maintaining noncompliance (Rodriguez, Thompson, & Baynham, 2010). It is possible that rationales may be more or less effective for noncompliance, depending on the variables maintaining the behavior. For example, the use of rationales for attention-maintained noncompliance might be contra-indicated because additional attention is delivered contingent upon noncompliance as part of the procedure. In the current study, a functional analysis of noncompliance might have identified the variables maintaining noncompliance and provided a more detailed explanation as to why rationales were ineffective for these participants.
Contingent access to preferred edible items initially appeared to be effective to increase compliance for both participants in this study. However, the effects of this procedure were short-lived; compliance decreased to low levels after 2 to 3 sessions. One reason why the contingent access condition was ineffective could be that during each trial, the edible item was not as preferred as the toy participants interacted with. That is, in this condition, a schedule was arranged in which edible items and toys competed with each other. Momentary fluctuations in preference for one over the other could have affected compliance.
The ineffectiveness of contingent access in this study stands in contrast to other studies (e.g., Wilder et al, 2007) demonstrating that differential reinforcement can be effective in increasing compliance. One possible reason for this discrepancy is that in previous research, coupons or tokens were delivered contingent upon compliance. These coupons likely became conditioned reinforcers, as they were exchangeable for a variety of preferred items and activities. In the current study, specific edible items were delivered contingent upon compliance. The reinforcing effectiveness of the edible items may have decreased after a few sessions. The consumption of many edible items may have acted as an abolishing operation, decreasing compliance during subsequent sessions.
The guided compliance procedure consistently increased compliance for both participants. For Hank, the guided compliance procedure took some time to become effective, but once it was effective, it remained so. For Craig, the guided compliance procedure was effective more quickly. This is consistent with previous research on guided compliance (Wilder & Atwell, 2006); for some children the procedure takes some time to become effective, but for others it is effective immediately.
Data on problem behavior were collected as part of this study to examine any "side effects" that might occur with the interventions. Although problem behavior was generally low throughout the study, it was somewhat elevated in the guided compliance condition. Extinction was the mechanism likely responsible for the effects of the guided compliance intervention, as emotional behavior and aggression increased when the therapist physically guided participants to comply with the instruction (Iwata, Pace, Cowdery, & Miltenberger, 1994). To keep problem behavior to a minimum, practitioners might consider combining contingent access to preferred items (i.e., differential reinforcement) with a guided compliance procedure. That is, a therapist might deliver an edible following guided compliance; extinction-induced emotional behavior and aggression may be less likely to occur under these conditions (Lerman, Iwata, & Wallace, 1999).
Future research should also examine other rationales. For example, research could examine rationales that describe rules, such as "Get dressed; you're having guests over and you can't play with them in your pajamas". Future research should also examine the effects of rationales delivered by parents or teachers as opposed to therapists. It is possible that caregiver-delivered rationales are more effective than those delivered by a therapist, particularly if the caregiver has a history of including descriptions of consequences and implementing the described consequences along with the rationales.
Bernal, M. E., Klinnert, M. D., & Schultz, L. A. (1980). Outcome evaluation of behavioral parent training and client-centered parent counseling for children with conduct problems. Journal of Applied Behavior Analysis, 13, 677-691.
Cote, C. A., Thompson, R. H., & McKerchar, P. M. (2005). The effects of antecedent interventions and extinction on toddlers' compliance during transitions. Journal of Applied Behavior Analysis, 38, 235-238.
Crowther, J. H., Bond, L. A., & Rolf, J. E. (1981). The incidence, prevalence, and severity of behavior disorders among preschool-aged children in day care. Journal of Abnormal Child Psychology, 9, 23-42.
Fisher, W., Piazza, C. C, Bowman, L. G., Hagopian, L. P., Owens, J. C, & Slevin, I. (1992). A comparison of two approaches for identifying reinforcers for persons with severe and profound disabilities. Journal of Applied Behavior Analysis, 25, 491-498.
Forehand, R., Gardner, H., & Roberts, M. (1978). Maternal response to child compliance and noncompliance: Some normative data. Journal of Clinical Child Psychology, 7,121-124.
Goetz, E. Holmberg, M., & LeBlanc, J. (1975). Differential reinforcement of other behavior and noncontingent reinforcement as control procedures during the modification of a preschooler's compliance. Journal of Applied Behavior Analysis, 8, 77-82.
Iwata, B. A., Pace, G. M., Cowdery, C. E., & Miltenberger, R. G. (1994). What makes extinction work: An analysis of procedural form and function. Journal of Applied Behavior Analysis, 27, 131-144.
Kalb, L. M. & Loeber, R. (2003). Child disobedience and noncompliance: A Review. Pediatrics, 111, 641-652.
Kapalka, G. M. (2007). Parenting your out-of-control child: An effective, easy-to-use program for teaching self-control Oakland, CA: New Harbinger Publications.
Keenan, K., Shaw, D., Delliquadri, E., Giovannelli, J., & Walsh, B. (1998). Evidence for the continuity of early problem behaviors: Application of a developmental model. Journal of Abnormal Child Psychology, 26, 441-454.
Lerman, D. C, Iwata, B. A., & Wallace, M. D. (1999). Side effects of extinction: Prevalence of bursting and aggression during treatment of self-injurious behavior. Journal of Applied Behavior Analysis, 32, 1-8.
Mace, F. C, Hock, M. L., Lalli, J. S., West, B. J., Belfiore, P., Pinter, E., et al. (1988). Behavioral momentum in the treatment of noncompliance. Journal of Applied Behavior Analysis, 21,123-141.
Malott, R. W., Whaley, D. L., & Malott, M. E. (1997). Elementary principles of behavior (3rd Ed.), Upper Saddle River, NJ: Prentice-Hall.
Marcus, B. & Vollmer, T. (1995). Effects of differential negative reinforcement on disruption and compliance. Journal of Applied Behavior Analysis, 28, 229-230.
McMahon, R. J., & Forehand, R. (2003). Helping the noncompliant child: Family-based treatment for oppositional behavior (2nd ed.). New York: Guilford Press.
Nelsen, J. (2006). Positive discipline. New York: Random House
Rodriguez, N., Thompson, R., & Baynham, T. (2010). Assessment of the relative effects of attention and escape on noncompliance. Journal of Applied Behavior Analysis, 43, 143-147.
Wilder, D., & Atwell, J. (2006). The effects of a guided compliance procedure to reduce noncompliance among preschool children. Behavioral Interventions, 21, 265-272.
Wilder, D. A., Allison, J. A., Nicholson, K., Abellon, O. E., & Saulnier, R. (2010). Further evaluation of antecedent interventions on compliance: The effects of "rationales" to increase compliance among preschoolers. Journal of Applied Behavior Analysis, 43, 601-613.
Wilder, D., Nicholson, K., & Allison, J. (2010). An evaluation of advance notice to increase compliance among preschoolers. Journal of Applied Behavior Analysis, 43, 751-755.
Wilder, D., Harris, C, Reagan, R., & Rasey, A. (2007). Functional analysis and treatment of noncompliance by preschool children. Journal of Applied Behavior Analysis, 40,173-177.
Zeilberger, J., Sampen, S. & Sloane, H. N. (1968). Modification of a child's problem behaviors in the home with the mother as therapist. Journal of Applied Behavior Analysis, 1, 47-53.
David A. Wilder, Kristin Myers, Katie Nicholson, Janelle Allison, and Anthony T. Fischetti Florida Institute of Technology
Correspondence to David A. Wilder, Florida Institute of Technology, School of Psychology, 150 W. University Blvd., Melbourne, FL 32901. E-mail: dawUder@fit.edu.
|Printer friendly Cite/link Email Feedback|
|Author:||Wilder, David A.; Myers, Kristin; Nicholson, Katie; Allison, Janelle; Fischetti, Anthony T.|
|Publication:||Education & Treatment of Children|
|Date:||Feb 1, 2012|
|Previous Article:||Utilizing antecedent manipulations and reinforcement in the treatment of food selectivity by texture.|
|Next Article:||Evidence-based practice: a framework for making effective decisions.|