Increasing Quiet Compliance by Detained Male Adolescents

Abstract

Some adjudicated adolescents receive treatment for their offenses in residential facilities. Detained adolescents’ engagement in either low levels of compliant behavior or excess behavior (e.g., swearing, gestures) while following commands from residential personnel may result in decreased opportunities for those youth to access preferred activities. The current study employed nonconcurrent multiple baseline across participants designs to evaluate the effects of a procedure to increase seven detained adolescents’ quiet compliance with academic and vocational demands. Results show that problem behavior decreased to zero or near-zero levels for each participant during simulated conditions and suggest that self-control, alone or in combination with a differential reinforcement of low rate behavior for omitting problem behavior, may have been responsible for the behavior changes. We discuss some clinical implications of the findings.

Keywords

detained adolescents problem behavior quiet compliance self-control differential reinforcement of low rate behavior DRL

Increasing Quiet Compliance by Detained Adolescents

According to the Office of Juvenile Justice and Delinquency Prevention, in 2013, the American legal system adjudicated and subsequently detained 123,655 juveniles for illegal behavior (Sickmund, Sladky, & Kang, 2015). Many detained adolescents engage in violent, aggressive, and other antisocial problem behaviors that can be disruptive to and stressful for the staff members within these facilities (Morrissey, 1997). To this end, research suggests that problem behavior of detained adolescents has become a high-level concern to administrators of these facilities (French & Gendreau, 2006). Although behavioral health services are available for detained adolescents in some residential facilities (e.g., Ford & Blaustein, 2013), there has been limited research on specific solutions for these problem behaviors in these settings (Fixsen, Blasé, Timbers, & Wolf, 2007). Specifically, the lack of a literature base that outlines a behaviorally based technology for practitioners may prevent detained juveniles from accessing effective treatment. Single-case experimental design (SCED) studies that directly evaluate interventions for problem behavior displayed by juvenile offenders may provide options for practitioners who treat individuals from this population (e.g., Lanovaz & Rapp, 2016). Over three decades ago, Hayes (1981) suggested that SCED methodology could give rise to practitioners publishing treatment effects they would otherwise not share with colleagues on a large scale.

Although there are relatively few studies that have focused on increasing compliant behavior by detained adolescents, there is well-developed literature for treating noncompliant behavior displayed by children with autism spectrum disorder (ASD: for example, see Kodak & Grow, 2011) and other neurodevelopmental disabilities. To this end, researchers have defined compliant and noncompliant behavior in several ways. For example, Cook, Rapp, and Schulze (2015) suggested that

compliance can be subcategorized as being active or passive; the former [active compliance] requires an individual to emit a specific response . . . passive compliance may involve teaching an individual to sit still, abstain from engaging in specific behavior, or otherwise tolerate an unpleasant event or stimulus. (p. 901)

Often, active compliance involves either initiating (or otherwise completing) a task within 5 to 10 s of instruction from a change agent (e.g., Wilder, Majdalany, Sturkie, & Smeltz, 2015) or engaging in behavior as specified in a task analysis (e.g., Conyers et al., 2004). Although detained juveniles may, at times, be required to display passive compliance (e.g., to sit quietly through a therapy session), the typical focus in academic and vocational settings is on active compliance with demands from teachers, dormitory staff, and security personnel (hereafter collectively referred to as residential personnel).

When treating noncompliant behavior for which a correct, active response is required, practitioners often distinguish between a skill deficit (i.e., the individual has not been trained to display the correct response) and a motivational deficit (i.e., the individual has the skill, but he is not sufficiently motivated to display it). Active compliance can involve an individual completing a task as indicated by a change agent. Nevertheless, it is possible for an individual to actively comply with a demand while simultaneously displaying problem behavior (e.g., swearing, eye rolling, or making inappropriate gestures while complying with a demand). We categorize this subtype of compliant behavior as “compliance with behavior excess” and, further, argue that this subtype may require different interventions than other subtypes of noncompliance.

In part, the intervention for this subtype of noncompliance could focus directly on the participant without altering the way demands are provided by residential personnel. For example, a child with ASD who displays compliance with excess may be exposed to differential consequences (e.g., see Cook et al., 2017) from a therapeutic change agent whereas a detained adolescent who displays compliance with excess may receive a brief verbal reprimand or temporarily lose privileges from residential personnel. Moreover, the youth’s receipt of verbal reprimands, loss of privileges, or both may then evoke or elicit other, more severe, problem behavior. Adolescents’ engagement in problem behavior may also strain relationships with the residential personnel. Toward this end, studies suggest that the exhibition of problem behavior by children with ASD substantially increases caregiver stress (e.g., Harrop, McBee, & Boyd, 2016). In this way, residential personnel who work with adolescents displaying high levels of problem behavior may also experience higher levels of stress.

One option for helping both the youth and facility personnel is to teach the youth to respond differently to the demands provided by some residential personnel. Thus, teaching detained adolescents to display actively compliant responses without excesses may enable them to avoid problematic behavioral cycles and, instead, contact more reinforcing events. That is, learning to display quiet compliance (i.e., compliant behavior without extra responses) in residential settings may be a behavioral cusp for detained adolescents (Bosch & Fuqua, 2001; Rosales-Ruiz & Baer, 1997), particularly if quiet compliance is learned by way of self-control. Self-control has been defined as a situation in which behavior is under the control of larger, more temporally distant consequences as opposed to under the control of smaller, more immediate consequences (Rachlin, 2016). In the current scenario, engaging in quiet compliance may be conceptualized as an act of self-control in that immediate reinforcement for the behavior may be relatively minor if present at all. However, the long-term consequences of engaging in quiet compliance may include better relationships with staff, increased opportunities to access new and enriched environments, and increased likelihood of accessing novel and preferred vocational opportunities. Given the issues outline above, studies on procedures that increase quiet compliance by detained juveniles are warranted.

The present study evaluated problem behavior emitted by seven detained adolescents. Each participant had a history of displaying behavioral excesses while following instructions from residential personnel. As a product of such behavior, these adolescents had lost privileges, delayed active engagement and progress in therapy sessions (a requirement for eventual release from the detention facility), and damaged relationships with multiple residential staff members. The intervention involved teaching the detained juveniles to comply with commands without excess behavior (e.g., swearing, eye rolling, etc.). Cooper, Heron, and Heward (2007) noted that researchers have defined self-control as either the ability to choose a delayed, larger reward instead of a smaller immediate reward or engaging in behavior at one point in time to change behavior at a later point in time. Drawing from a combination of the two meanings, the aim of the present study was to teach the youth to forgo the short-term reinforcer generated by excess behavior during demands in exchange for reinforcers that would later become available for having engaged in compliant behavior without excess behavior. Put differently, the goal of behavioral intervention was to increase each participant’s self-control such that he or she displayed quiet compliance during demands from residential personnel.

Method

Participants

Participants were seven adolescent males aged 15 to 18 years who were adjudicated for adolescent illegal sexual behavior and ordered to complete treatment in a secure residential facility. Todd, Frank, Lewis, and Devin were Black and Jeff, Terry, and Craig were White. Todd, Frank, Jeff, Craig, Lewis, and Devin each performed academically at or just slightly below grade level. Terry performed academically well below grade level (e.g., at approximately a third- or fourth-grade level for reading and spelling). Assessment and treatment services are provided through a university-based treatment program housed within the facility. Programming included weekly individual and group treatment, individual family and multi-family group treatment, educational and vocational services, and recreational activities. The primary treatment modality used in the program was cognitive-behavioral therapy. Clinical staff made referrals to a behavioral service unit for various problem behaviors that each participant emitted in response to change agents’ instructions in dormitory, classroom, and vocational settings. Each participant typically displayed compliant responding with additional behavior.

Setting

All sessions occurred in a group therapy room in the residential facility. Group therapy rooms were approximately 6 m by 8 m and contained a table, a couch, and approximately eight chairs. Applied behavior analysis (ABA) interns conducted one to four sessions each meeting, 1 to 2 times per week, and collected data using paper data sheets, pens, and a stopwatch.

Behavioral Measures

Dependent variables

Although the focus of this study was on increasing quiet compliance, we opted to collect data on each participant’s problem behavior (i.e., excess behavior) during each session, rather than for each demand, for three broad reasons. First, some students displayed one excess behavior during multiple demands, whereas other students displayed numerous excess behaviors after only one or two specific demands. Second, some residential personnel ignored low-level excess behavior, some provided consequences that were commensurate with the excess behavior, and others provided the same consequences (e.g., point reductions) for any excess behavior. Third, data showed that participants completed 95% of the demands in each session, suggesting that they were able to complete the tasks. Thus, we targeted quiet compliance, which we defined as complying with a direction without excess behavior (e.g., laughing, talking back, rolling eyes) that may be client specific. Given these broad concerns, we found that depicting the rate of problem behavior per simulated session best captured the problem for each participant. Observers collected data on the frequency of problem behavior displayed by each participant using immediate onset and 2-s offset criteria. Table 1 contains the definitions for each participant’s problem behavior.

Table 1.

Response Definitions for Each Participant’s Problem Behavior.

Participant	Problem behavior
Jeff	Off-topic questions: Any student question not related to the current topic.
Jeff	Disrespect: Laughing within 2 s of a directive or laughing while completing a previously delivered directive, loud audible sighs, shaking head, or eye rolling.
Todd	Verbal aggression: Any verbal behavior (including spoken, written, and gestures) that included threatening statements about others or self (e.g., “I’m going to hit you” or “I’m going to blow up”) or cursing (e.g., “F___ you”, “Oh s___”). These events could have occurred with or without eye gaze toward another person. Included sustained eye gaze (2 s or more) while mouthing words.
	Provoking others: Any verbal behavior (spoken, written, or gestures) directing others to injure the student (e.g., “come at me,” “swing”) or claiming that others (a peer or a staff member) had injured him. Also included in the definition was approaching to within 0.66 m of another person or making posturing threats while making sustained eye gaze toward another person.
	Inappropriate sexual behavior: Eye gaze directed toward a female’s chest or the pubic region or buttocks of a male or female that lasted 1-s or longer. This definition included verbal behavior (comments, gestures, or writing) that was sexual in nature.
	Talking out of turn: Any instance of the student talking when the therapist was talking or a student talking when there was an ongoing demand to remain silent.
	Bizarre speech: Any instance of student referring to his hypothetical child or non-contextual inappropriate topics such as “do you smoke crack?”
	Disrespect: As defined for Jeff.
Frank	Inappropriate vocalizations: Any instance of the student cursing or using derogatory language (e.g., “F____,” “This is bull____”)
	Talking back: Any instance in which the student emitted a comment or utterance within 2 s of the therapist’s delivery of a demand. Excluded instances wherein Frank asked a question for clarification (i.e., “Should I write number 2 next to or below number 1?”).
	Off-topic questions: As defined for Jeff.
Terry	Interruptions: Any instance of student behavior that resulted in the therapist ending or delaying the current activity for 5 s or more to attend to the student behavior.
	Unauthorized departure: Any instance of a student moving 5 m or more from the group perimeter without prior permission of the therapist.
	Talking out of turn: As defined for Todd.
	Talking back: As defined for Frank.
	Off-topic questions: As defined for Jeff.
Lewis	Interruptions: As defined for Terry.
	Talking out of turn: Any instance of the student talking when the therapist was talking or a student talking or laughing when there was an ongoing demand to remain silent.
	Talking back: As defined for Frank.
	Off-topic questions: As defined for Jeff.
	Disrespect: As defined for Jeff.
Craig	Interruptions: As defined for Terry.
	Talking out of turn: As defined for Lewis
	Talking back: As defined for Frank.
	Off-topic questions: As defined for Jeff.
	Disrespect: Any instance in which the student sighs audibly, rolls eyes, smack lips, laughs within 2 s of a directive being given, sustains eye contact with staff member for 2 s when the staff member is not talking, or shakes head. Immediate onset.
Devin	Interruptions: As defined for Terry.
	Talking out of turn: As defined for Lewis
	Talking back: As defined for Frank.
	Off-topic questions: As defined for Jeff.
	Disrespect: As defined for Jeff.

Interobserver agreement (IOA)

To assess IOA, ABA interns divided 10-min sessions into 30-s bins and then calculated agreement scores using the block-by-block or proportional method (Mudford, Martin, Hui, & Taylor, 2009). The ABA interns assessed IOA for 44%, 33%, 19%, 30%, and 17% of sessions for Todd, Terry, Craig, Lewis, and Devin, respectively; the mean IOA score for each participant’s behavior was 90% or higher. A formal index of IOA was not collected for either Jeff or Frank because (a) only one ABA intern had been approved for the facility when these two participants received treatment and (b) video recording (for post hoc observations) was not allowed in the facility. However, the data collector for Jeff and Frank also served as a reliable data collector for Terry, who received treatment after the facility approved additional ABA interns.

General Procedures

Preference assessment

Administrators in the treatment program had pre-approved a list of 17 items and activities that ABA interns (therapists) could use as reinforcers. Upon intake for ABA services, ABA interns interviewed the referring case manager about each participant’s preferred activities. Based on the procedures described by DeLeon and Iwata (1996), therapists then conducted a seven-item verbal multiple stimulus without replacement (V-MSWO) for each participant. Each preference assessment session consisted of seven note cards containing the type-written preferred items or activities available at the facility (e.g., snack, approved media, cards). The note cards were presented in the V-MSWO format as described by DeLeon and Iwata (1996). Specifically, seven note cards were presented to the participant in an arched array. Upon the first presentation of each note card (i.e., the first block of trials), therapists read the card aloud and asked the participant to provide an example of the item to ensure comprehension. Once all seven items were present in the arched array, the therapist instructed the participant to pick their favorite. A selection was determined by the participant touching the card, reading the card out loud, or both. Participants did not receive access to the items upon selecting the respective note card. Once a card was selected, the therapist removed the chosen card, moved the card on the left to the right, and presented a new trial. Trials continued until all cards were selected or the student vocally indicated he did not want the activities on the remaining cards (e.g., Terry never chose reading or puzzles). This process was repeated 3 times with each participant. At least one highly preferred item was identified for each participant. The top three or four items from the preference assessment were made available via a reward menu. Data from the preference assessments are available upon request from the second author.

Experimental design

A four-tiered nonconcurrent multiple baseline (NMBL) across participants design was used to evaluate the effects of the intervention on problem behavior displayed by Jeff, Todd, Frank, and Terry. Subsequently, a three-tiered NMBL across participants design with an embedded reversal design for the first tier was used to evaluate the effects of the intervention on problem behavior displayed by Lewis, Craig, and Devin. Data paths for each participant’s behavior were visually inspected for the effects of the intervention in a targeted and nontargeted condition (except Craig).

Procedures

As previously noted, ABA interns conducted two different types of demand conditions: academic and vocational (the only condition for Craig). Each session was 10 min in duration. The order in which therapists conducted each condition was counterbalanced across sessions. The academic demand condition consisted of a therapist sitting across from the participant at a table. An academically appropriate workbook was present along with a notebook with lined paper and a pencil for the participant. During academic baseline and treatment conditions, the therapist delivered continuous academic demands that were tailored for each participant (e.g., “open the book to page 256,” “write down the definition,” “read number two out loud,” etc.). This condition was intended to simulate the stimulus events in the participants’ classrooms.

The vocational demand condition consisted of the therapist standing on one side of the room, in which clothing (i.e., one pair of jeans and three t-shirts), a puzzle in a box, a box of playing cards, crayons in a crayon box, approximately eight stacking chairs, and three magazines were present. During vocational baseline and treatment conditions, the therapist delivered continuous vocational demands (e.g., “fold the shirt,” “open the crayon box,” “pick up a hand-full of puzzle pieces”) that were related to the present items. The therapist delivered demands in a random, nonlinear fashion. That is, the tasks were intended to be unpredictable. This condition was intended to simulate stimulus events in the participants’ dormitory and vocational settings.

Baseline

During the first baseline session, the participant’s lead ABA therapist told the participant that they would be working on following directions. The therapist informed each participant that she was going to deliver different directions, either academic or vocational, and then instructed the participant to follow the directions. The therapist also told each participant that she would not deliver any feedback on his performance for the first few sessions. The therapist then solicited questions from the participant before beginning the session.

Throughout each session, the lead therapist delivered simple one-step directions specific to the condition (i.e., vocational or academic). If the participant complied with a direction, the therapist immediately (e.g., within 3 s) presented a new direction. Compliance was defined as any instance in which the participant followed a directive within 5 s. Each directive provided a new opportunity for compliance. If the participant did not emit a compliant response within 5 s of a directive, the therapist refrained from commenting and, instead, immediately presented a new direction. The therapist ignored problem behavior. That is, a participant could have been compliant with a directive while simultaneously emitting problem behavior and no differential consequences would have been provided. After 10 min elapsed, a secondary therapist signaled that the session had ended. Depending on the day and the session, the lead therapist either initiated a new session after a 1-min intertrial interval or told the participant that they were done for the day.

After the last session of each day, the lead therapist told the participant to pick one preferred activity that they could engage in for about 10 min for “participating in the session.” If the participant asked specific questions about his performance or treatment, the therapist re-stated that they were working on following directions together, but provided no further information. With the exception of Craig, who received only vocational sessions, therapists conducted an equal number of academic and vocational baseline sessions for a given participant.

Treatment

Following baseline conditions, therapists only provided treatment in one condition; this allowed us to test for generalization of behavior change to the non-treated condition. Jeff received treatment in the vocational condition because it was the only condition in which he exhibited problem behavior. Therapists continued to conduct sessions in the academic condition to evaluate whether Jeff’s problem behavior increased when treatment was delivered in the academic condition. Todd and Terry received treatment in the academic condition because their problem behavior was stable across sessions in that condition. Frank received treatment in the academic condition because his teachers reported that he displayed problem behavior on a daily basis. Craig only participated in vocational conditions because staff reported that he only displayed problem behavior in vocational settings. Lewis and Devin received treatment in the academic condition because both displayed higher baseline rates of problem behavior in that condition.

Treatment sessions were identical to baseline sessions except for the combination of antecedent instructions and feedback that was delivered after the session was completed. The condition that did not contain treatment was conducted in the same manner as described in baseline. To deliver antecedent instructions to the participant, the lead therapist met with the participant before the first treatment session and showed the participant a graph with his problem behavior. The therapist explained the definition of problem behavior and why the participant needed to decrease his problem behavior (i.e., explaining how their behavior made staff behave toward them). The therapist further explained that if the participant was able to meet a certain target (i.e., 80% reduction from individualized baseline rates for each participant), then he would be allowed to choose one preferred item from the previously mentioned reward menu. The reward menu included a list of approved activities and snacks. These items had been identified via interviews from students and approved by the treatment team administration. In addition, researchers had conducted a V-MSWO for each student upon intake to ensure the menu included items that were highly preferred. The therapist also explained that if the participant was able to refrain from all instances of problem behavior, he would be allowed to choose two preferred items. The therapist further noted that access to preferred items or activities would only be provided for meeting the target level of problem behavior (see the appendix for a sample script). In this way, the intervention was intended to function as a differential reinforcement of low rate behavior (DRL) procedure (e.g., Champagne, Ike, McLaughlin, & Williams, 1990). The therapist received verbal consent from each participant and answered questions prior to conducting the first treatment session. This entire process required approximately 10 min.

Treatment sessions were then conducted as in baseline. Following the session, the primary data collector told the lead therapist and participant how many instances of problem behavior were emitted by the participant in the 10-min session. If there were any instances of problem behavior, the primary data collector asked the participant if he recalled what the problem behavior was. If the participant gave the correct response, the primary data collector neutrally confirmed that the participant was correct. If the participant gave the incorrect response, the primary data collector gave neutral verbal feedback about what specific behavior they tracked as problem behavior. After the primary data collector delivered feedback, the lead therapist told the student what he had earned (i.e., one or two reward choices) and allowed him to choose from the menu. If a student chose activity-based items (e.g., approved media, basketball), they were allowed 10 min of access per reward. If a student chose a snack, they were allowed to choose one packaged snack per reward from the snack closet on facility grounds. Snack options included individual bags of chips, cookies, and other popular snack items. Jeff typically chose a snack, access to music, or both. Todd typically chose a snack, access to playing cards, or both. Frank and Terry typically chose a snack, access to approved media, or both. Lewis, Craig, and Devin typically chose access to approved media.

Lewis’s case manager reported that Lewis often engaged in problem behavior (e.g., talking out of turn) when other students talked to Lewis during independent work periods. As such, Lewis participated in additional sessions in which he was taught to emit quiet compliance even when distractions occurred. Baseline distractor sessions were identical to previous baseline sessions except a therapist, who simulated a student’s behavior in Lewis’s classroom, emitted a distracting vocal statement (e.g., “Lewis, let’s play tic tac toe,” “look at this cool drawing”) on a fixed-time, 30-s schedule. The DRL treatment for the distractor sessions was provided in a manner that was identical to previous DRL sessions in the academic condition.

Results

Figure 1 shows the rate of problem behavior across sessions for Jeff, Todd, Frank, and Terry. During the baseline phase, Jeff (first panel) displayed low to moderate rates of problem behavior in the vocational condition (M = 0.3 rpm) and no problem behavior in the academic condition. Following the introduction of DRL in the vocational condition, Jeff’s problem behavior decreased to zero in the vocational condition (M = 0.0 rpm) and remained at zero in the academic condition. Across baseline and DRL sessions in the vocational and academic conditions, the therapist delivered demands to Jeff at mean rates of 6.8 rpm (range, 4.9-8.4 rpm) and 2.0 rpm (range, 1.2-2.7 rpm), respectively. During the baseline phase, Todd (second panel) displayed low to moderate rates of problem behavior in both the vocational (M = 0.5 rpm) and academic (M = 0.6 rpm) conditions. Following introduction of DRL in the academic condition, Todd’s problem behavior decreased in both the academic (M = 0.0 rpm) and vocational (M = 0.2 rpm) conditions. Across baseline and DRL sessions in the vocational and academic conditions, the therapist delivered demands to Todd at mean rates of 10.6 rpm (range, 8.7-12.1 rpm) and 6.3 rpm (range, 5.1-7.5 rpm), respectively. During the baseline phase, Frank (third panel) displayed low to moderate rates of problem behavior in the vocational (M = 0.6 rpm) and academic (M = 0.2 rpm) conditions. As with Todd, following the introduction of DRL in the academic condition, Frank’s problem behavior decreased in both the academic (M = 0.0 rpm) and the vocational (M = 0.1 rpm) conditions. Across baseline and DRL sessions in the vocational and academic conditions, the therapist delivered demands to Frank at mean rates of 6.7 rpm (range, 3.9-7.5 rpm) and 2.0 rpm (range, 1.3-2.5 rpm), respectively. During the baseline phase, Terry (fourth panel) displayed moderate to high rates of problem behavior in the vocational (M = 1.4 rpm) and academic (M = 1.0 rpm) conditions. As with Frank and Todd, after the introduction of DRL in the academic condition, Terry’s problem behavior decreased in the academic (M = 0.0 rpm) and vocational (M = 0.3 rpm) conditions. Across baseline and DRL sessions in the vocational and academic conditions, therapist delivered demands to Terry at mean rates of 10.3 rpm (range, 7.0-11.5 rpm) and 5.4 rpm (range, 1.3-7.3 rpm), respectively.

Figure 1.

Rate of problem behavior across baseline and DRL phases for Jeff (first panel), Todd (second panel), Frank (third panel), and Terry (fourth panel).

Figure 2 shows the rate of problem behavior across sessions for Lewis, Craig, and Devin. During the baseline phase, Lewis (first panel) displayed moderate to high rates of problem behavior in the academic condition (M = 1.2 rpm) and low rates of problem behavior in the vocational condition (M = 0.2 rpm). Following the introduction of DRL in the academic condition, Lewis’s problem behavior decreased to low rates in the academic condition (M = 0.2 rpm) and also decreased to near-zero levels in the vocational condition (M = 0.04 rpm). Across baseline and DRL sessions in the academic and vocational conditions, the therapist delivered demands to Lewis at mean rates of 7.6 rpm (range, 6.5-9.0 rpm) and 9.7 rpm (range, 8.0-11.0 rpm), respectively. During the baseline with distractor phase, Lewis displayed moderate rates of problem behavior in both the academic (M = 0.6 rpm) and vocational (M = 0.5 rpm) conditions. Following the introduction of DRL in the academic plus distractor condition, Lewis’s problem behavior decreased to low rates in both the academic (M = 0.1 rpm) and vocational (M = 0.1 rpm) conditions. Across baseline and DRL distractor sessions in the academic and vocational conditions, the therapist delivered demands to Lewis at mean rates of 7.2 rpm (range, 6.2-7.7 rpm) and 10.1 rpm (range, 8.7-11.4 rpm), respectively. During the baseline phase, Craig (second panel) displayed low to moderate rates of problem behavior (M = 0.4 rpm) in the vocational condition. Following the introduction of DRL in the vocational condition, Craig’s problem behavior decreased to near-zero levels (M = 0.03 rpm). Across baseline and DRL sessions, the therapist delivered demands to Craig at a mean rate of 9.3 rpm (range, 7.7-10.9 rpm). During the baseline phase, Devin (third panel) displayed high rates of problem behavior in the academic condition (M = 2.1 rpm) and moderate to low rates of problem behavior in the vocational condition (M = 0.7 rpm). Following the introduction of DRL in the academic condition, Devin’s problem behavior decreased to low rates in both the academic (M = 0.14 rpm) and vocational (M = 0.12 rpm) conditions. Across baseline and DRL sessions in the academic and vocational conditions, the therapist delivered demands to Devin at a mean rate of 6.5 rpm (range, 5.5-7.4 rpm) and 9.1 rpm (range, 8.0-9.9 rpm), respectively.

Figure 2.

Rate of problem behavior across baseline and DRL phases for Lewis (first panel), Craig (second panel), and Devin (third panel).

Discussion

Results of this study indicate that providing instructions plus differentially reinforcing compliant behavior without behavior excesses (i.e., quiet compliance) decreased problem behavior for seven detained male adolescents. Upon implementation of treatment in the targeted condition, problem behavior rapidly decreased to zero or near-zero levels for each participant. In addition, problem behavior in the nontargeted condition decreased to zero levels across three to five sessions for Todd, Frank, Terry, Lewis, and Devin, and did not increase for Jeff.

Although the behavioral mechanism responsible for the change in each participant’s behavior is not clear, there are two characteristics of the data sets that suggest instructional or self-control (e.g., Rachlin, 2016) may have been responsible for treatment effects. First, levels of problem behavior decreased to zero levels in the first treatment session for four of the seven participants. That is, four participants displayed quiet compliance prior to contacting the programmed contingency. This outcome suggests that the therapist’s delivery of antecedent instructions, at least in part, contributed to the treatment effects. Nevertheless, because each participant contacted programmed reinforcers shortly after meeting the stated response requirement, we cannot conclude that instructions alone accounted for each participant’s onging quiet compliance.

Second, and not unrelated to the first, the rapid spread of treatment effects to non-treatment conditions provides additional evidence that self-control (based on instructions delivered by therapists), as opposed to the actual contingency, decreased problem behavior. Notably, participants never contacted the programmed reinforcement for displaying quiet compliance during the non-treatment conditions. Nevertheless, it is possible that the participants did not readily discriminate between the two conditions. That is, both conditions involved a therapist’s delivery of continuous demands with which participants often complied. In this sense, because therapists conducted sessions in an alternating fashion, the participants’ quiet compliance could have been reinforced on an intermittent schedule.

Third, the therapists did not provide consequences for engaging in quiet compliance until the end of the 10-min DRL sessions. As such, the consequence (access to a preferred item or activity) for low- or zero-rate problem behaviors during the treatment sessions was delayed by as much as 15 min. The fact that participants displayed quiet compliance so rapidly, and with the delayed consequence, further suggest that self-control in the form of rule-following contributed to the participants’ behavior change (e.g., see Kissi et al., 2017).

The present study extends behavior-analytic research with detained male adolescents in at least three ways. First, it illustrates how a NMBL design may be used to evaluate the effects of a specific intervention for detained adolescents. Toward this end, Novotny et al. (2014) found that NMBL designs with three or more tiers provided substantial protection against false positives. More broadly, practitioners are being asked to serve more diverse populations for which there may be extensive practice gaps. One way to fill these practice gaps is for practitioners to conduct applied studies (Brogan, Falligant, & Rapp, 2017; Kelley et al., 2015). Second, the data patterns of four participants (i.e., Todd, Frank, Terry, and Craig) provide evidence that instructional control may reduce problem behavior of some detained male adolescents. This outcome is noteworthy given that each participant was incarcerated as a result of an offense that could be characterized by impulsivity (i.e., low self-control). Finally, although the multiple stimulus without replacement preference assessment (DeLeon & Iwata, 1996) was designed for application for individuals with developmental disabilities, the exhibition of desirable behavior change by all participants suggests that the V-MSWO format may be an appropriate variation for detained male adolescents.

There are several limitations of the present study. First, we did not evaluate the operant function of each participant’s problem behavior during demands. Given that participants generally displayed actively compliant responses, it seems unlikely that their behavior produced social negative reinforcement (i.e., escape from instruction). Alternatively, it is possible that behavior excesses produced social positive reinforcement in the form of peer attention or automatic reinforcement in the form of counter control (e.g., delivering aversive stimulation in kind to a change agent who provides an aversive demand). Despite the absence of a functional analysis, the intervention was effective for each participant. Second, the lack of secondary observations for both Jeff and Frank detracts somewhat from the believability of the findings. Nevertheless, the sole observer for Jeff and Frank subsequently produced high IOA scores for subsequent participants when secondary observers became available. By extension, it seems likely that the sole observer engaged in accurate behavior measurement for Jeff and Frank. Third, it is unclear as to what specific participant characteristics (e.g., age, academic grade level, or behavioral health diagnoses) may account for the success of the intervention in this study. Interestingly, Jeff, Todd, Frank, Craig, Lewis, and Devin, each performed at or just below their academic grade level whereas Terry performed at approximately a third- or fourth-grade level. In light of the broad differences in participants’ academic abilities, Terry’s behavior decreased in both conditions in the same manner as the other three participants. Finally, we did not directly evaluate the extent to which quiet compliance generalized to untrained settings with other staff members. Nevertheless, multiple staff members reported varying degrees of improvement for all seven participants.

Future research should replicate the procedures described in this study. In addition, future research may aim to identify what particular participant characteristics are associated with decreases in problem behavior from instructional control only and determine the extent to which contingencies are needed to support quiet compliance. Furthermore, future research may consider incorporating social validity measures to determine the acceptability of the treatment procedures by detained adolescents and the detectability of the outcomes by clinical staff and other residential staff members. Toward this end, future research should directly evaluate the extent to which decreases in problem behavior during treatment sessions generalize to periods of time in class or with various staff persons.

Footnotes

Appendix

Acknowledgements

The authors would like to thank the Department of Youth Services for their support, and the students who participated in the study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Kristen M. Brogan, BCaBA, is currently a student in the master’s of Applied Behavior Analysis Program at Auburn University. Her research interests include the assessment and treatment of problem behavior emitted by detained adolescents, the assessment and treatment of stereotypy and applications of single-subject experimental designs.

John T. Rapp, PhD, BCBA-D, is a professor in the Department of Psychology at Auburn University. His research interests include the assessment and treatment of stereotypy and applications of single-subject experimental designs.

Amanda M. Niedfeld is currently a student in the master’s of Applied Behavior Analysis Program at Auburn University. Her research interests include the assessment and treatment of problem behavior emitted by detained adolescents, the assessment and treatment of automatically reinforced behavior, and applications of single-subject experimental designs.

Jodi C. Coon is currently a student in the master’s of Applied Behavior Analysis Program at Auburn University. His research interests include the assessment and treatment of problem behavior emitted by detained adolescents, the assessment and treatment of automatically reinforced behavior, and applications of single-subject experimental designs.

Jan L. Everhart Newman, JD, PhD is currently an assistant clinical professor in the Department of Psychology at Auburn University. Her research is broadly focused on the relationships between child and adolescent psychopathology, trauma, and delinquency in court-involved youth.

Barry R. Burkhart, PhD in a professor in the Department of Psychology at Auburn University. He has broad recent interests in assessment and treatment of problems resulting from violence and victimization. Currently, he is co-director of a treatment program for juvenile sex offenders incarcerated in a state training school. He is a fellow of Divisions 12 and 29 of the American Psychological Association.

References

Bosch

Fuqua

R. W.

(2001). Behavioral cusps: A model for selecting targeting behaviors. Journal of Applied Behavior Analysis, 34, 123-125.

Brogan

K. M.

Falligant

J. M.

Rapp

J. T.

(2017). Interdependent group contingencies decrease adolescents’ disruptive behaviors during group therapy: A practitioner’s demonstration. Behavior Modification, 41, 405-421.

Champagne

J. M.

Ike

E. C.

McLaughlin

T. F.

Williams

R. L.

(1990). Use of differential reinforcement of lower rates of behavior and self-monitoring with a delinquent adolescent in a residential setting. Journal of Instructional Psychology, 17, 123-132.

Conyers

Miltenberger

R. G.

Peterson

Gubin

Jurgens

Selders

. . . Barenz

(2004). An evaluation of in vivo desensitization and video modeling to increase compliance with dental procedures in persons with mental retardation. Journal of Applied Behavior Analysis, 37, 233-238.

Cook

J. L.

Rapp

J. T.

Mann

K. R.

McHugh

Burji

Nuta

(2017). A practitioner model for increasing eye contact in children with autism. Behavior Modification, 41, 382-404.

Cook

J. L.

Rapp

J. T.

Schulze

K. A.

(2015). Differential negative reinforcement of other behavior to increase wearing of a medical bracelet. Journal of Applied Behavior Analysis, 48, 901-906.

Cooper

J. O.

Heron

T. E.

Heward

W. L.

(2007). Applied behavior analysis (2nd ed.). Upper Saddle River, NJ: Prentice Hall.

DeLeon

I. G.

Iwata

B. A.

(1996). Evaluation of a multiple-stimulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29, 519-533.

French

S. A.

Gendreau

. (2006). Reducing prison misconducts: What works! Criminal Justice and Behavior, 33, 185-21

10.

Fixsen

D. L.

Blasé

K. A.

Timbers

G. D.

Wolf

M. M.

(2007). In search of program implementation: 792 replications of the teaching-family model. The Behavior Analyst Today, 8, 96-110.

11.

Ford

J. D.

Blaustein

M. E.

(2013). Systematic self-regulation: A framework for trauma-informed services in residential juvenile justice program. Journal of Family Violence, 28, 665-677.

12.

Harrop

McBee

Boyd

B. A.

(2016). How are child restricted and repetitive behaviors associated with caregiver stress over time? A parallel process multilevel growth model. Journal of Autism and Developmental Disorders, 5, 1773-1783.

13.

Hayes

S. C.

(1981). Single case experimental design and empirical clinical practice. Journal of Consulting and Clinical Psychology, 49, 193-211.

14.

Kelley

D. P.

III Wilder

D. A.

Carr

J. E.

Rey

Green

Lipschultz

(2015). Research productivity among practitioners in behavior analysis: Recommendations from the prolific. Behavior Analysis in Practice, 8, 201-206.

15.

Kissi

Hughes

Mertens

Barnes-Holmes

De Houwer

Crombez

(2017). A systematic review of pliance, tracking, and augmenting. Behavior Modification. 41, 683-707.

16.

Kodak

Grow

L. L.

(2011). Behavioral treatment of autism. In Fisher

W. W.

Piazza

C. C.

Roane

H. S.

(Eds.), Handbook of applied behavior analysis (pp. 402-416). New York, NY: Guilford Press.

17.

Lanovaz

M. J.

Rapp

J. T.

(2016). Using single-case experiments to support evidence-based decisions: How much is enough? Behavior Modification, 40, 377-395.

18.

Morrissey

(1997). A multimodal approach to controlling inpatient assaultiveness among incarcerated juveniles. Journal of Offender Rehabilitation, 25, 31-42.

19.

Mudford

O. C.

Martin

N. T.

Hui

J. K.

Taylor

S. A.

(2009). Assessing observer accuracy in continuous recording of rate and duration: Three algorithms compared. Journal of Applied Behavior Analysis, 42, 527-539.

20.

Novotny

M. A.

Sharp

K. J.

Rapp

J. T.

Jelinski

J. D.

Lood

E. A.

Steffes

A. K.

(2014). False positives with visual analysis for nonconcurrent multiple baseline designs and ABAB designs: Preliminary findings. Research in Autism Spectrum Disorders, 8, 933-943.

21.

Rachlin

(2016). Self-control based on soft commitment. The Behavior Analyst, 39, 259-268.

22.

Rosales-Ruiz

Baer

D. M.

(1997). Behavioral cusps: A developmental and pragmatic concept for behavior analysis. Journal of Applied Behavior Analysis, 30, 533-544.

23.

Sickmund

Sladky

Kang

(2015). Easy Access to Juvenile Court Statistics: 1985-2013. Retrieved from https://www.ojjdp.gov/ojstatbb/ezajcs/

24.

Wilder

D. A.

Majdalany

Sturkie

Smeltz

(2015). Further evaluation of the high-probability instructional sequence with and without programmed reinforcement. Journal of Applied Behavior Analysis, 48, 511-522.