Decreasing excessive functional communication responses while treating destructive behavior using response restriction

Research in Developmental Disabilities 35 (2014) 2614–2623

Contents lists available at ScienceDirect

Research in Developmental Disabilities

Decreasing excessive functional communication responses while treating destructive behavior using response restriction Wayne W. Fisher, Brian D. Greer *, Angie C. Querim 1, Nicole DeRosa 2 University of Nebraska Medical Center’s Munroe-Meyer Institute, 985450 Nebraska Medical Center, Omaha, NE 68198, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 20 June 2014 Accepted 25 June 2014 Available online

Functional communication training (FCT) is an established treatment for destructive behavior that missucceeds in about 37% of cases when the reinforcement schedule for the functional communication response (FCR) is thinned using multiples schedules (mult FCT; Hagopian, Boelter, & Jarmolowicz, 2011). In this investigation, we evaluated the use of response restriction FCT (RR FCT) in a cohort of patients with poorly differentiated responding of the FCR during mult FCT. Results showed that (a) RR FCT maintained high rates of correct FCRs during the reinforcement component of RR FCT without increasing destructive behavior; (b) children displayed highly discriminated FCRs when an FCR card and a control card were simultaneously available during the reinforcement component of RR FCT; and (c) near-zero rates of destructive behavior were observed during the last five sessions of the terminal reinforcement schedule. Results are discussed relative to differences between mult FCT and RR FCT and successive and simultaneous discriminations. ß 2014 Elsevier Ltd. All rights reserved.

Keywords: Destructive behavior Functional analysis Multiple schedule Functional communication training Response restriction Successive discrimination Simultaneous discrimination

1. Introduction Approximately 4 million people in the U.S. have an intellectual disability, and 12.5% (½ million) display destructive behavior, such as aggression and self-injurious behavior (SIB; Emerson et al., 2001). The risk for destructive behavior increases with intellectual-disability severity, communication deficits, and co-occurring autism spectrum disorder (Holden & Gitlesen, 2006). Severe destructive behavior is a primary reason for institutionalization. In 1994, the annual costs of destructive behavior exceeded $3.5 billion in the U.S., and the costs are undoubtedly much higher today (Thompson & Gray, 1994). SIB produces health complications, including soft tissue damage, blindness, self-amputation of body parts, fractures, brain trauma, and even death (Hyman, Fisher, Mercugliano, & Cataldo, 1990). Individuals with aggressive behavior are at increased risk for institutionalization, social isolation, physical restraint, over-use of medication, denial of services, and physical abuse (Antonacci, Manuel, & Davis, 2008). One of the most important advancements in the treatment of severe destructive behavior has been the development of functional-analysis methods (Beavers, Iwata, & Lerman, 2013), which collectively represent a systematic approach to identifying the reinforcer for destructive behavior and a variety of other problem behaviors (e.g., Chapman, Fisher, Piazza, & Kurtz, 1993; Piazza et al., 2003). The identification of the consequence that reinforces destructive behavior facilitates the

* Corresponding author. Tel.: +1 402 559 3563. E-mail address: [email protected] (B.D. Greer). 1 Now at Ball State University. 2 Now at The Kelberman Center and Upstate Medical University. http://dx.doi.org/10.1016/j.ridd.2014.06.024 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

2615

development of effective behavioral intervention because it allows the behavior analyst to discontinue the contingency between destructive behavior and its reinforcer (i.e., extinction [EXT]) and to deliver that reinforcer contingent on one or more appropriate responses (Fisher & Bouxsein, 2011; Vollmer & Athens, 2011). The function-based intervention for severe destructive behavior with the most research studies (and treated cases) supporting its effectiveness is called functional communication training (FCT; Tiger, Hanley, & Bruzek, 2008). FCT typically involves (a) training the individual to request the reinforcer for destructive behavior (identified by a functional analysis) using an appropriate communication response (i.e., a mand) and (b) discontinuing reinforcement for destructive behavior (i.e., EXT). Research results indicate that function-based interventions, like FCT, typically reduce destructive behavior by at least 90% and are more effective than similar behavioral treatments that are not informed by a functional analysis (Didden, Duker, & Korzilius, 1997; Didden, Korzilius, van Oorsouw, & Sturmey, 2006; Iwata, Pace, et al., 1994; Rooker, Jessel, Kurtz, & Hagopian, 2013). Although the effectiveness of FCT is well established when implemented in controlled environments by highly trained therapists using dense schedules of reinforcement (e.g., FR 1), treatment relapse often occurs when a caregiver is unable to accurately carry out the procedures in the natural environment. In the home, parents have multiple competing activities (e.g., meal preparation, cleaning, laundry, paying bills, phone calls, texts, emails), which make it highly impractical to carry out a treatment that requires them to deliver the requested reinforcer as often as once every thirty seconds (e.g., Betz, Fisher, Roane, Mintz, & Owen, 2013). As a result, a number of investigators have developed strategies for leaning the schedule of reinforcement for the functional communication response (FCR) in order to make the treatment more practical for use in the home, school, and other community settings (e.g., Fisher, Thompson, Bowman, Hagopian, & Krug, 2000). The goals of these schedule-thinning procedures are typically to (a) lower the rate of the FCR at times when reinforcement is unavailable, (b) have the FCR occur almost exclusively at times when it is reasonable for the parent to deliver the requested reinforcer, (c) maintain the FCR’s strength, and (d) mitigate or prevent resurgence of destructive behavior on those occasions when the FCR is emitted but not reinforced (Betz et al., 2013; Fisher, Kuhn, & Thompson, 1998; Fisher et al., 2000; Hagopian, Contrucci Kuhn, Long, & Rush, 2005; Hagopian et al., 2011; Hagopian, Toole, Long, Bowman, & Lieving, 2004; Hagopian, Fisher, Sullivan, Acquisto, & LeBlanc, 1998; Hanley, Iwata, & Thompson, 2001; Roane, Fisher, Sgro, Falcomata, & Pabico, 2004; Rooker et al., 2013; Tiger & Hanley, 2004; Tiger et al., 2008). One early approach to schedule thinning with FCT involved inserting and then progressively lengthening delays between the FCR and its reinforcer (e.g., a therapist telling the child that they have to wait immediately after emission of an FCR, followed a few seconds later by delivery of the reinforcer, and then progressively lengthening subsequent delays between the response and reinforcer; Fisher et al., 1993; Hagopian et al., 1998). A similar approach used with escape-reinforced destructive behavior (called ‘‘demand fading’’; Fisher et al., 1993; or ‘‘response chaining’’; Lalli, Casey, & Kates, 1995) consisted of (a) requiring the child to complete one or two demands prior to allowing escape for the FCR and then (b) gradually increasing the number of demands that had to be completed prior to reinforcing the FCR. Another early approach was to bring the FCR under discriminative control using multiple schedules (mult FCT) with quasi-randomly alternating components in which the FCR was reinforced (FR 1) or on EXT (Fisher et al., 1998; Hanley et al., 2001). A third approach involved providing the child with an alternative activity or reinforcer during periods when the FCR was not reinforced (Fisher et al., 2000; Hagopian et al., 2005), which was in part based on prior research with time-based (or noncontingent) reinforcement schedules (Hanley, Piazza, & Fisher, 1997) and basic research on self-control (e.g., Grosch and Neuringer, 1981). A fourth approach, called response restriction (RR FCT), involved removing the communication card or device for progressively longer durations (Fyffe, Kahng, Fittro, & Russel, 2004; Hagopian et al., 2004; Roane et al., 2004). Only a small number of investigations have directly compared different schedule-thinning procedures using withinsubject research methods (e.g., Hanley et al., 2001). Hanley et al. (2001) compared schedule thinning using mult FCT with several alternative schedules (progressively lengthened reinforcer delays, progressively lengthened fixed-interval schedules, and mixed schedules). They found that the discriminative stimuli used in mult FCT facilitated schedule thinning and low rates of destructive behavior by the end of treatment. Subsequent investigations have consistently replicated the finding that contingency-correlated stimuli facilitate (a) discriminative responding for the FCR and (b) schedule thinning (Betz et al., 2013; Tiger & Hanley, 2004). Hagopian et al. (2011) reviewed the extant literature on schedule thinning and found that mult FCT was successful in reaching the target terminal schedule for the FCR while maintaining low rates of destructive behavior in five of eight applications (62.5%). They concluded that mult FCT was the preferred method of schedule thinning when the goal was to have the individual tolerate periods of non-reinforcement for the FCR lasting longer than one min. Rooker et al. (2013) presented data on 14 applications of schedule thinning with mult FCT (without supplemental procedures) and similarly found that a 90% or greater reduction was observed with six applications (43%) and an 80% or greater reduction was observed with nine applications (64%). One reason that using mult FCT may not always achieve the goals of schedule thinning mentioned above is that some individuals have difficulty learning to fully discriminate between the reinforcement and EXT components of the multiple schedule (i.e., individuals may not display high and stable rates of the FCR in the presence of the Sd, low rates of the FCR in the D presence of the S , and low rates of destructive behavior throughout). In particular, individuals may have difficulty learning D to not display the FCR in the presence of the S (i.e., resistance to EXT). For example, in the Hanley et al. (2001) article, the subject named Karen displayed stable rates of the FCR in the presence of the Sd, but initially displayed 26.7 FCRs per min during the EXT component of the multiple schedule. In addition, even after 27 sessions of mult FCT (each lasting 10 to

2616

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

12 min), she displayed 0.9 FCRs per min during the last five sessions. It is not difficult to envision how inconvenient it would be to a parent if their child requested the functional reinforcer about once every minute at times when the parent displayed D the S to signal the unavailability of reinforcement. D For the subject named Jake in the Hanley et al. (2001) study, the rates of the FCR in the presence of the S were considerably lower but reached near-zero rates only after about 38 sessions of exposure to mult FCT. Results of other investigations have suggested that providing individuals with spoken rules that specify the response-consequence relations D correlated with the Sd and S can produce more rapid and complete discrimination between the components of mult FCT (Betz et al., 2013; Tiger & Hanley, 2004). However, such rules are only helpful for individuals who are verbally competent and who can readily follow such rules. An alternative approach to schedule thinning that could be used with individuals who do not have the verbal competencies needed to readily follow contingency-specifying rules is called RR FCT (e.g., Roane et al., 2004). With RR FCT, the individual is taught to request the reinforcer for destructive behavior by touching a picture card or handing the card to the therapist. Initially, the card is available continuously. Once low and stable rates of destructive behavior have been observed for multiple sessions and the individual is consistently requesting reinforcement via the FCR, the card is removed from view and made unavailable for a brief period (usually a few seconds). Next, the duration of time that the card is unavailable and out of sight is gradually lengthened (as long as destructive behavior remains low). Two potential advantages of RR FCT are (a) the FCR never contacts the response-weakening effects of EXT and (b) it can help to ensure (within reasonable limits) that the child will only request the function reinforcer via the FCR at times when it is practical for the therapist (or caregiver) to deliver it because the therapist controls the availability of the FCR card. However, one potential disadvantage of switching to RR FCT when mult FCT results in poorly differentiated responding of the FCR, is that RR FCT could potentially evoke or occasion an increase in destructive behavior during periods in which reinforcement (and the FCR) is unavailable. Basic research has shown that when multiple concurrent responses are available and one of them is restricted, it typically produces an increase in responding to one or more of the remaining alternatives (Crosbie, 1993; Dunham & Grantmyre, 1982; Green & Streifel, 1988). Thus, additional research is needed to determine whether RR FCT represents a viable option for completing reinforcer-schedule thinning without evoking increased problem behavior when a child has not made sufficient progress in schedule thinning using mult FCT. Only a small number of studies (Fyffe et al., 2004; Hagopian et al., 2004; Roane et al., 2004) with a small number of subjects (i.e., a total of 4 subjects) have evaluated the effects of RR FCT. In addition, none of these studies has examined whether RR FCT would be an effective alternative when subjects fail to show fully discriminated responding of the FCR during mult FCT (e.g., Hanley et al., 2001). Therefore, the current investigation targeted individuals who failed to fully discriminate responding during mult FCT in order to determine whether eliminating the opportunity to emit an FCR during the reinforcement-unavailable component of RR FCT would produce a substantial increase in destructive behavior. An additional goal of the study was to determine whether subjects would show fully discriminated responding if the D discrimination was simplified from a successive discrimination (i.e., Sd during the reinforcement component; S during the D d EXT component) to a simultaneous discrimination (i.e., S and S both present during the reinforcement component and both absent during the reinforcement-unavailable component). Basic research comparing successive and simultaneous discriminations using comparable stimuli have generally found that successive discriminations are more difficult than simultaneous discriminations (see Mackintosh, 1974 for a discussion). Finally, we present summary outcome data on rates of destructive behavior for the final five treatment data points at the terminal schedule for each child as a potential replication of prior studies using RR FCT to conduct schedule thinning. 2. Method 2.1. Study overview A functional analysis was completed first to identify the reinforcer responsible for each child’s destructive behavior. Functional analysis results were then used to design an FCT treatment for each child during which a card-touch response was taught that resulted in access to the reinforcer previously maintaining destructive behavior. Next, a mult-FCT condition was arranged during which FCRs produced the functional reinforcer according to an FR 1 schedule of reinforcement during reinforcement components and which resulted in no programmed consequence during EXT components. Across both components, destructive behavior was placed on EXT. All children were observed to make frequent FCRs during the EXT component of mult FCT, despite destructive behavior often remaining low or decreasing across sessions. Next, subjects were taught to discriminate between two cards, one card associated with reinforcement (an FCR card) and another card associated with EXT (the control card). Response restriction FCT began once children reliably emitted discriminated card touches. The contingencies for each card were later reversed, and the previous FCR card was removed to demonstrate that FCRs for each D child were fully discriminated between the Sd and S present within the simultaneous discrimination. The original RR FCT preparation and contingencies were subsequently reinstated to replicate discriminated FCRs. Last, we thinned the schedule of reinforcement using RR FCT to a terminal schedule of 60 s during the reinforcement component and 240 s during the reinforcement-unavailable component.

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

2617

2.2. Subjects and setting Four children who were referred for the assessment and treatment of destructive behavior and who exhibited poor discrimination during the reinforcement and EXT components of mult FCT participated. Evan was a 7-year-old boy diagnosed with an autism spectrum disorder, stereotypic movement disorder with self-injurious behavior, and disruptive behavior disorder with aggression who engaged in self-injury and aggression. Evan communicated using a combination of single words, sign language, gestures, and leading caregivers to preferred items. Dan was a 7-year-old boy diagnosed with intellectual disability and disruptive behavior disorder. Dan engaged in aggression and property destruction, and his functional analysis assessed both topographies of destructive behavior. However, a later functional analysis revealed Dan’s property destruction to be maintained by automatic sources of reinforcement. Therefore, only Dan’s aggression data are presented in the current investigation. Dan’s primary form of communication consisted of two-word phrases and some short sentences. John was a 4-year-old boy diagnosed with an autism spectrum disorder and disruptive behavior disorder who engaged in aggression and property destruction. John communicated using single words and two-word phrases. In a separate evaluation, DeRosa, Fisher, and Steege (under review) taught and measured John’s use of two forms of FCRs (i.e., a card touch and a vocal response) while measuring John’s destructive behavior during each FCT type. John was appropriate for inclusion in the current evaluation because John’s destructive behavior remained lower and his FCRs were more consistent when a card-touch FCR was used. Casey was a 4-year-old boy diagnosed with an autism spectrum disorder, intermittent explosive disorder, and wandering who engaged in aggression, self-injury, property destruction, tantrums, and elopement. Casey’s aggression, self-injury, and property destruction were targeted for assessment and treatment. However, Casey engaged in only three instances of self-injury across functional analysis and treatment-evaluation sessions. Therefore, Casey’s self-injury data were excluded from analysis. Casey communicated using single words and two-word phrases. All sessions were conducted in therapy rooms (3 m by 3 m) equipped with one-way observation windows. Evan’s sessions were conducted in a similar-sized, padded therapy room for safety purposes. Therapy rooms contained tables, chairs, and any materials necessary for session. 2.3. Response measurement and interobserver agreement Aggression was defined as any response that could harm another person and included hitting, kicking, biting, pushing, pinching, scratching, and spitting at the therapist. Self-injury for Ethan was defined as any response that could harm himself and included head banging, wrist or hand hitting, self-pinching, and self-scratching. Property destruction was defined as any response that could harm materials or the environment and included hitting, kicking, and throwing materials not intended for that purpose, as well as overturning furniture. A card-touch response was used as the FCR with each child to ensure the therapist could quickly prompt the communication response and then immediately present the functional reinforcer, thus minimizing exposure to the establishing operation for destructive behavior (DeRosa et al., under review). Correct functional communication response was defined as the child touching the FCR card during mult-FCT pretraining, mult FCT, RR FCT and as touching the control card during RR-FCT control. Incorrect functional communication response was defined as touching the FCR card during the EXT component (EXT FCRs) of mult-FCT pretraining and mult FCT and as touching the control card (control FCRs) during RR-FCT pretraining and RR FCT. The terms EXT FCRs and control FCRs were adopted to better distinguish how incorrect FCRs differed across treatment phases. Across all phases, however, children emitted few FCRs during reinforcer-access periods. Therefore, we omitted these data. Interobserver agreement was assessed by having a second observer simultaneously yet independently collect data on 32% (range, 13% to 50%) of functional-analysis sessions and on 33% (range, 17% to 46%) of treatment sessions. An agreement was defined as both observers recording the same number of responses in a 10-s interval. Interobserver-agreement coefficients were calculated by summing the number of agreement intervals and dividing by the number of agreement plus disagreement intervals then converting the proportion to a percentage. Interobserver agreement for aggression averaged 98% (range, 87% to 100%) for Evan, 98% (range, 88% to 100%) for Dan, 99% (range, 95% to 100%) for John, and 100% for Casey. Interobserver agreement for self-injury averaged 93% (range, 57% to 100%) for Evan. Interobserver agreement for property destruction averaged 98% (range, 83% to 100%) for John and 99% (range, 90% to 100%) for Casey. Interobserver agreement for correct FCRs averaged 97% (range, 92% to 100%) for Evan, 94% (range, 81% to 100%) for Dan, 98% (range, 92% to 100%) for John, and 92% (range, 80% to 100%) for Casey. Interobserver agreement for incorrect FCRs averaged 98% (range, 93% to 100%) for Evan, 95% (range, 81% to 100%) for Dan, 98% (range, 90% to 100%) for John, and 97% (range, 87% to 100%) for Casey. 2.4. Functional analysis Functional-analysis conditions were modeled after Iwata, Dorsey, Slifer, Bauman, and Richman (1994) with modifications recommended by Fisher, Piazza, and Chiang (1996). Ignore, attention, play, escape, and tangible conditions were conducted with all children with one exception: a tangible test condition was not conducted with Dan, as indirect-assessment results indicated his destructive behavior was unlikely to be maintained by this source of reinforcement. Preferred materials were identified for each child by conducting paired-choice preference assessments (Fisher et al., 1992). All sessions lasted 5 min with the exception that Dan’s sessions were 10 min in duration.

2618

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

During the ignore condition, a therapist sat with the child and ignored destructive behavior. Children did not have access to preferred materials. High-quality attention from a therapist was provided briefly (approximately 1 min) prior to the start of the attention condition. Children were then provided moderately preferred toys while the therapist read a magazine or completed work and ignored all destructive behavior. Therapists provided statements of concern for 20 s following destructive behavior. Children and therapists played together with highly preferred toys during the play condition. Therapists avoided issuing demands, and destructive behavior produced no programmed consequence. A three-step (verbal, model, physical) prompting sequence was used to guide compliance with pre-academic and academic instructions during the escape condition. Destructive behavior produced a 20-s break from demands. Highly preferred toys were provided briefly (approximately 1 min) prior to the start of the tangible condition. Preferred toys were then removed and were provided for 20 s following destructive behavior. 2.5. Mult-FCT treatment evaluation Mult FCT was evaluated following each child’s functional analysis and following a mult-FCT pretraining condition in which children were taught to emit the FCR. During mult FCT, FCRs in the presence of the Sd resulted in access to the D functional reinforcer identified during the functional analysis, whereas FCRs in the presence of the S resulted in EXT. Destructive behavior produced no programmed consequence throughout mult FCT. 2.5.1. Mult-FCT pretraining Prior to the start of mult FCT, children were taught to emit the FCR. All sessions consisted of 10 trials. A progressiveprompt delay was used in which the establishing operation for destructive behavior was presented (e.g., ‘‘Touch your nose’’ for escape-reinforced destructive behavior) while immediately guiding the child to emit the FCR and then terminating the establishing operation (e.g., providing a 20-s break from instructions). After two sessions, the prompt delay was increased from 0 s to 2 s and later to 5 s and then 10 s provided destructive behavior remained low. Additional teaching procedures were implemented as needed to teach each child to emit the FCR. 2.5.2. Mult FCT Mult FCT began once children were reliably emitting the FCR. Sessions were 10 min in duration for all children with the exception that Casey’s sessions lasted 5 min. Mult FCT consisted of quasi-randomly alternating periods in which FCRs were reinforced according to an FR 1 schedule and periods in which FCRs were placed on EXT. The schedule-correlated stimuli D used to signal each component of mult FCT differed slightly across children. For Evan, a green (Sd) or red (S ) card (7.6 cm by 12.7 cm) attached to a lanyard worn by the therapist was used to signal the reinforcement and EXT components, respectively. For Dan and John, a green wristband was worn by the therapist to signal reinforcement components, and the wristband was removed and hidden from view during EXT components. For Casey, a green wristband was also worn by the therapist to signal reinforcement components; however, EXT components were signaled by replacing the green wristband with a red wristband. For all children, FCRs during reinforcement components produced 20-s access to the functional reinforcer. There were no programmed consequences for FCRs during EXT components. During both components, destructive behavior resulted in no programmed consequence. For all children except Casey, reinforcement and EXT components of mult FCT were conducted for 60 s each (i.e., mult FCT 60/60). Reinforcement components were 60 s in duration for Casey, and EXT components lasted 30 s. For three children (Dan, John, and Casey), the therapist delivered pre-session rules (e.g., ‘‘Dan, we are going to do some work. When the green bracelet is on, you can ask for a break by touching the card. When the green bracelet is off, you can touch the card, but you will not get a break.’’). Contingency-specifying rules were used to facilitate discrimination of the FCR between reinforcement and EXT components and to minimize the occurrence of destructive behavior. 2.6. RR-FCT treatment evaluation Because children consistently emitted FCRs during the EXT component of mult FCT, children were exposed to RR-FCT pretraining followed by RR FCT. During RR-FCT pretraining, children were taught to discriminate the FCR card from a control card using stimulus shaping and stimulus prompts. We then evaluated children’s correct and incorrect FCRs and as well as rates of destructive behavior during quasi-randomly alternating periods in which reinforcement was available for 60 s and periods in which reinforcement and both response cards were unavailable for 60 s. 2.6.1. RR-FCT pretraining During RR-FCT pretraining, FCR cards and control cards differed by color or by the presence of text (e.g., ‘‘Break’’ on the FCR card). In addition, the control cards were initially smaller than the FCR card. Initial control-card sizes differed across children, but in all cases, the size of the control card was gradually increased over pretraining sessions until each child’s control card was equal in size to his FCR card. Control cards ranged in size from 3.3 cm by 4.8 cm (smallest starting control card) to 14 cm by 21.6 cm (largest ending control card). All sessions consisted of 10 trials, and cards were quasi-randomly alternated between left and right positions across trials. Pretraining began by simultaneously presenting the FCR card with the smallest control card. The therapist then presented the establishing operation for destructive behavior and immediately guided the FCR. Following two sessions at a 0-s prompt

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

2619

delay, subsequent delays were increased to 2 s and then to 5 s, provided destructive behavior remained low. Once children consistently emitted correct FCRs independently at a 5-s prompt delay, the size of the control card was gradually increased until it was equal in size to the FCR card. To enhance discrimination between cards, the FCR card for three children (Evan, Dan, and Casey) was placed initially 10 cm to 15 cm closer to the child than the control card. The last step of RR-FCT pretraining was to place the cards equidistant from the child. 2.6.2. RR FCT RR FCT started once children were reliably discriminating between the FCR card and the control card. The RR-FCT condition was similar to mult FCT in that the reinforcement component and the reinforcement-unavailable component were quasi-randomly alternated in 60-s periods (60/30 used with Casey). The RR-FCT condition differed from mult FCT in that the FCR card and control card were simultaneously presented when reinforcement was available, and both cards were removed when reinforcement was unavailable. Stimuli signaling each component of mult FCT (e.g., the colored wristbands) were also removed during RR FCT. All sessions were 10 min in duration with the exception that Casey’s sessions were 5 min. 2.6.3. RR-FCT control The RR-FCT control condition was identical to RR FCT, except the original FCR card was removed, and reinforcement was made contingent on control-card touches. The original FCR card was removed to prevent the response-weakening effects of pairing the original FCR card with EXT, as the original FCR card and its associated contingencies were later reinstated during a return to the RR-FCT condition. This control condition was included to demonstrate that FCRs for each child were fully D discriminated between the Sd and S present within the simultaneous discrimination. 2.6.4. RR-FCT schedule thinning Following reintroduction of RR FCT, schedule thinning was implemented by gradually lengthening the duration of the reinforcement-unavailable component with each child and continued until a terminal schedule of 240 s was reached. The duration of the reinforcement component remained 60 s throughout schedule thinning. 3. Results 3.1. Functional analysis Functional-analysis results for all children are depicted in Fig. 1. Evan engaged in consistently higher rates of destructive behavior during only the escape condition relative to play. Therefore, Evan’s destructive behavior was maintained by escape. Dan engaged in aggression during escape and ignore conditions of the functional analysis. Based on the sequencing of conditions, Dan’s aggression during the escape condition appeared to carry over to ignore sessions. An additional analysis was conducted (data available from the author) that suggested Dan’s aggression was maintained only by escape. John engaged in elevated rates of destructive behavior during the escape condition relative to play, indicating that his destructive behavior was also maintained by escape. Casey engaged in destructive behavior during all test conditions of the functional analysis. Therefore, a pairwise design was used to isolate each potential function of Casey’s destructive behavior. Elevated rates of destructive behavior occurred only during the escape condition relative to play. Therefore, Casey’s destructive behavior was also maintained by escape. 3.2. Treatment Fig. 2 displays treatment data for each child. Evan engaged in variable but high rates of correct FCRs and FCRs during the EXT component of mult FCT, and Evan displayed moderate to high rates destructive behavior. Following RR-FCT pretraining, Evan engaged in increased rates of correct FCRs, and his destructive behavior decreased to low rates across the reinforcement D available and reinforcement-unavailable components of RR FCT, despite the unavailability of the S and the FCR during the reinforcement-unavailable components. Evan engaged in no instances of FCRs to the control card during RR FCT, indicating that Evan effectively discriminated between the FCR cards. When the FCR card was removed and reinforcement was made available for control-card touches during RR-FCT control, Evan emitted high rates of FCRs to the control card, and he engaged in low to moderate rates of destructive behavior. Reinstating RR FCT initially produced an increase in the rate of Evan’s destructive behavior as well as poorer discrimination between the FCR cards. Therefore, we replicated previously observed patterns of responding to the FCR card(s) during a subsequent reversal to RR-FCT control followed by RR FCT. Evan’s destructive behavior remained low throughout the remainder of his evaluation. During mult FCT, Dan displayed decreasing rates of correct FCRs, FCRs during the EXT component, as well as decreasing rates of aggression. Dan continued to engage in low levels of aggression throughout the remainder of the evaluation. However, Dan displayed high rates of correct FCRs and low rates of FCRs to the control card during RR FCT. Dan’s rate of FCRs to the control card increased during RR-FCT control, and previously observed levels of correct FCRs and FCRs to the control card were replicated during Dan’s final RR-FCT phase.

[(Fig._1)TD$IG]

2620

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

Fig. 1. Responses per minute of destructive behavior during ignore, attention, play, escape, and tangible conditions of the functional analysis for Evan, Dan, John, and Casey.

John engaged in moderate rates of correct FCRs and FCRs during the EXT component when mult FCT was implemented. John displayed low rates of destructive behavior that decreased across treatment phases. During RR-FCT phases, John emitted a high rate of correct FCRs and few FCRs to the control card. Higher rates of FCRs to the control card were observed when card touches to the control card resulted in reinforcement during RR-FCT control. D Casey emitted a moderate rate of correct FCRs and a high rate of FCRs in the presence of the S during mult FCT. Casey engaged in a variable but generally low rate of destructive behavior throughout the evaluation. Similar to the other children, Casey emitted a higher rate of correct FCRs than FCRs to the control card during both RR-FCT phases and produced an elevated rate FCRs to the control card during RR-FCT control. Schedule thinning was later completed (data available from the author) with each child. Rates of destructive behavior during the last five sessions at the terminal schedule of 60 s of reinforcement alternated with 240 s during which reinforcement was unavailable were 0.04, 0.02, 0.04, and 0 for Evan, Dan, John, and Casey, respectively. 4. Discussion Four children with autism spectrum disorder and/or intellectual disability who displayed severe destructive behavior were exposed to FCT and schedule thinning using two types of compound schedules, mult FCT and RR FCT. All four children showed poorly discriminated responding of the FCR during mult FCT in that a sizable percentage of their FCRs occurred in the D presence of the S . By contrast, RR FCT obtained all of the goals for FCT schedule thinning previously mentioned. That is, RR FCT eliminated FCRs when reinforcement was unavailable without increasing destructive behavior. Subjects displayed highly discriminated FCRs when an FCR card and a control card were simultaneously available during the reinforcement component of RR FCT. Response restriction FCT prevented the FCR from contacting the response-weakening effects of EXT. Finally, the FCR card was absent for 80% of session time during the terminal schedule, and near-zero rates of destructive behavior were observed during the last five sessions for all children.

[(Fig._2)TD$IG]

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

2621

Fig. 2. Responses per minute of destructive behavior, correct FCRs, FCRs during EXT, and FCRs to a control card during mult FCT, RR FCT, and RR-FCT control for Evan, Dan, John, and Casey.

Perhaps the most interesting and important finding of this investigation was that the initial introduction of RR FCT did not result in a substantial increase in destructive behavior for any of the children. This is somewhat surprising because RR FCT differs from mult FCT in two salient ways that might affect destructive behavior. During the reinforcement-unavailable D component of RR FCT, neither the S nor the FCR response card was present, whereas in the EXT component of mult FCT, both D were present. One might expect that the removal of the S from the reinforcement-unavailable component would result in some increase in destructive behavior if that stimulus exerted some suppressive control over the response (cf. Terrace, 1971). D The current results suggest that the S during mult FCT exerted little or no suppressive control over destructive behavior. D Future researchers should test whether mult FCT results in subjects learning to avoid the S using a preparation similar to the one used by Terrace to further examine this issue. Similarly, given that FCRs persisted at relatively high rates during the EXT component of mult FCT, one might expect that its restriction might evoke destructive behavior, because this latter response had a history of reinforcement in the same context (similar to what occurs with response class hierarchies, resurgence, or other response-restriction paradigms; cf. Crosbie, 1993; Lalli, Mace, Wohn, & Livezey, 1995; Mace et al., 2010). One possible reason for the absence of a substantial increase in destructive behavior is that this response had been exposed to EXT for multiple sessions prior the introduction of RR FCT. For example, behavioral momentum theory predicts that longer exposures to differential reinforcement of alternative behavior should lower the probability of resurgence of destructive behavior relative to shorter exposures (see Nevin & Shahan, 2011 for a discussion). Another potentially interesting finding was that all four subjects showed highly discriminated responding when RR FCT D was introduced and the Sd and S were present simultaneously during the reinforcement component of this compound schedule. One factor that probably contributed to this level of discriminated responding was that errorless-training

2622

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

procedures (e.g., gradual stimulus fading) were used to teach the discrimination. A second factor is that RR FCT involved a D simultaneous discrimination between the Sd and S whereas mult FCT involved a successive discrimination. Simultaneous discriminations are typically acquired more easily than successive discriminations using comparable stimuli, although there are exceptions (see Mackintosh, 1974 for a discussion). Nevertheless, one limitation of the current investigation is that errorless procedures were used to train the simultaneous discrimination in RR FCT but not to train the successive discrimination in mult FCT. It is possible that if comparable gradual stimulus fading procedures were used to train the successive discrimination in mult FCT that more fully discriminated responding would have been observed. Future researchers should directly compare how rapidly discriminated FCRs are acquired under simultaneous and successive discrimination paradigms using equivalent stimulus fading methods. The current investigation specifically targeted children who showed poorly discriminated responding during mult FCT. That is, in our treatment program we typically implement mult FCT first and then switch to RR FCT only in those cases where mult FCT does not produce highly discriminated responding for the FCR. An alternative and perhaps preferable method of determining whether to use mult FCT or RR FCT would be to base such treatment decisions on a brief and accurate empirical assessment. Therefore, future research should be directed toward the development of an efficient assessment that would accurately predict whether mult FCT or RR FCT would be more effective for producing highly discriminated responding of the FCR with individual subjects. Finally, Hagopian et al. (2011) identified a number of potential limitations of RR FCT. One limitation was that RR FCT does not permit the individual to contact periods of non-reinforcement. This limitation was addressed (at least to some extent) in this study by including a control card in the reinforcement component of RR FCT. That is, control-card touches contacted EXT in the RR-FCT phases and contacted reinforcement in the RR-FCT-control phases, and each child showed highly discriminated responding for this control card (i.e., low rates when it was correlated with EXT in RR FCT and high rates when it was correlated with reinforcement in RR-FCT control). A second limitation of RR FCT mentioned by Hagopian et al. is that it may not be appropriate for individuals who use an augmentative-communication device as one would not want to restrict access to the individual’s communication system. However, most such devices allow caregivers, teachers, and/or therapists to make specific communication icons available at certain times and not others. In such cases, it would be possible to make the FCR visible on the display of the device during the reinforcement component of RR FCT and not visible during the reinforcement-unavailable component. Future research should specifically examine the effectiveness of RR FCT with individuals who use augmentative-communication devices.

References Antonacci, D. J., Manuel, C., & Davis, E. (2008). Diagnosis and treatment of aggression in individuals with developmental disabilities. Psychiatric Quarterly, 79, 225–247. Beavers, G. A., Iwata, B. A., & Lerman, D. C. (2013). Thirty years of research on the functional analysis of problem behavior. Journal of Applied Behavior Analysis, 46, 1–21. Betz, A. M., Fisher, W. W., Roane, H. S., Mintz, J. C., & Owen, T. M. (2013). A component analysis of reinforcer-schedule thinning during functional communication training. Journal of Applied Behavior Analysis, 46, 219–241. Chapman, S., Fisher, W. W., Piazza, C. C., & Kurtz, P. F. (1993). Functional assessment and treatment of life-threatening drug ingestion in a dually-diagnosed youth. Journal of Applied Behavior Analysis, 26, 255–256. Crosbie, J. (1993). The effects of response cost and response restriction on a multiple-response repertoire with humans. Journal of the Experimental Analysis of Behavior, 59, 173–192. DeRosa, N., Fisher, W.W., & Steege, M.W., An evaluation of time in establishing operation on the effectiveness of functional communication training. Journal of Applied Behavior Analysis, (under review).. Didden, R., Duker, P., & Korzilius, H. (1997). Meta-analytic study of treatment effectiveness for problem behaviors with individuals who have mental retardation. American Journal of Mental Retardation, 101, 387–399. Didden, R., Duker, P., & Korzilius, H. (1997). Meta-analytic study of treatment effectiveness for problem behaviors with individuals who have mental retardation. American Journal of Mental Retardation, 101, 387–399. Didden, R., Korzilius, H., van Oorsouw, W., & Sturmey, P. (2006). Behavioral treatment of challenging behaviors in individuals with mild mental retardation: Metaanalysis of single-subject research. American Journal of Mental Retardation, 111, 290–298. Dunham, P. J., & Grantmyre, J. (1982). Changes in a multiple-response repertoire during response-contingent punishment and response restriction: Sequential relationships. Journal of the Experimental Analysis of Behavior, 37, 123–133. Emerson, E., Kiernan, C., Alborz, A., Reeves, D., Mason, H., Swarbrick, R., et al. (2001). The prevalence of challenging behaviors: A total population study. Research in Developmental Disabilities, 22, 77–93. Fisher, W. W., & Bouxsein, K. (2011). Developing function-based reinforcement procedures for problem behavior. In W. W. Fisher, C. C. Piazza, & H. S. Roane (Eds.), Handbook of behavior analysis. New York: Guilford Publishing. Fisher, W. W., Kuhn, D. E., & Thompson, R. H. (1998). Establishing discriminative control of responding using functional and alternative reinforcers during functional communication training. Journal of Applied Behavior Analysis, 31, 543–560. Fisher, W. 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. Fisher, W., Piazza, C., Cataldo, M., Harrell, R., Jefferson, G., & Conner, R. (1993). Functional communication training with and without extinction and punishment. Journal of Applied Behavior Analysis, 26, 23–36. Fisher, W. W., Piazza, C. C., & Chiang, C. L. (1996). Effects of equal and unequal reinforcement durations during a functional analysis. Journal of Applied Behavior Analysis, 29, 117–120. Fisher, W. W., Thompson, R. H., Hagopian, L. P., Bowman, L. G., & Krug, A. (2000). Facilitating tolerance of delayed reinforcement during functional communication training. Behavior Modification, 24, 3–9. Fyffe, C. E., Kahng, S. W., Frittro, E., & Russell, D. (2004). Functional analysis and treatment of inappropriate sexual behaviors. Journal of Applied Behavior Analysis, 37, 401–404. Green, G., & Streifel, S. (1988). Response restriction and substitution with autistic children. Journal of the Experimental Analysis of Behavior, 50, 21–32. Grosch, J., & Neuringer, A. (1981). Self-control in pigeons under the Mischel paradigm. Journal of the Experimental Analysis of Behavior, 35, 3–21.

W.W. Fisher et al. / Research in Developmental Disabilities 35 (2014) 2614–2623

2623

Hagopian, L. P., Boelter, E. W., & Jarmolowicz, D. P. (2011). Reinforcement schedule thinning following functional communication training: Review and recommendations. Behavior Analysis in Practice, 4, 4–16. Hagopian, L. P., Contrucci Kuhn, S. A., Long, E. S., & Rush, K. S. (2005). Schedule thinning following communication training: Using competing stimuli to enhance tolerance to decrements in reinforcer density. Journal of Applied Behavior Analysis, 38, 177–193. Hagopian, L. P., Fisher, W. W., Sullivan, M. T., Acquisto, J., & LeBlanc, L. A. (1998). Effectiveness of functional communication training with and without extinction and punishment: A summary of 21 inpatient cases. Journal of Applied Behavior Analysis, 31, 211–235. Hagopian, L. P., Toole, L. M., Long, E. S., Bowman, L. G., & Lieving, G. A. (2004). A comparison of dense-to-lean and fixed-lean schedules of alternative reinforcement and extinction. Journal of Applied Behavior Analysis, 37, 323–338. Hanley, G. P., Iwata, B. A., & Thompson, R. H. (2001). Reinforcement schedule thinning following treatment with functional communication training. Journal of Applied Behavior Analysis, 34, 17–38. Hanley, G. P., Piazza, C. C., & Fisher, W. W. (1997). Noncontingent presentation of attention and alternative stimuli in the treatment of attention-maintained destructive behavior. Journal of Applied Behavior Analysis, 30, 229–237. Holden, B., & Gitlesen, J. P. (2006). A total population study of challenging behaviour in the county of Hedmark Norway: Prevalence, and risk markers. Research in Developmental Disabilities, 27, 456–465. Hyman, S. L., Fisher, W., Mercugliano, M., & Cataldo, M. F. (1990). Children with self-injurious behavior. Pediatrics, 85, 437–441. Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27, 197–209. Iwata, B. A., Pace, G. M., Dorsey, M. F., Zarcone, J. R., Vollmer, T. R., Smith, R. G., et al. (1994). The functions of self-injurious behavior: An experimentalepidemiological analysis. Journal of Applied Behavior Analysis, 27, 215–240. Lalli, J. S., Casey, S., & Kates, K. (1995). Reducing escape behavior and increasing task completion with functional communication training, extinction, and response chaining. Journal of Applied Behavior Analysis, 28, 261–268. Lalli, J. S., Mace, F. C., Wohn, T., & Livezey, K. (1995). Identification and modification of a response-class hierarchy. Journal of Applied Behavior Analysis, 28, 551–559. Mace, F. C., McComas, J. J., Mauro, B. C., Progar, P. R., Taylor, B. A., Ervin, R., et al. (2010). Differential reinforcement of alternative behavior increases resistance to extinction: Clinical demonstration, animal modeling, and clinical test of one solution. Journal of the Experimental Analysis of Behavior, 93, 349–367. Mackintosh, N. J. (1974). The psychology of animal learning. New York: Academic Press. Nevin, J. A., & Shahan, T. A. (2011). Behavioral momentum theory: Equations and applications. Journal of Applied Behavior Analysis, 44, 877–895. Piazza, C. C., Fisher, W. W., Brown, K. A., Shore, B. A., Patel, M. R., Katz, R. M., et al. (2003). Functional analysis of inappropriate mealtime behaviors. Journal of Applied Behavior Analysis, 36, 187–204. Roane, H. S., Fisher, W. W., Sgro, G. M., Falcomata, T. S., & Pabico, R. R. (2004). An alternative method of thinning reinforcer delivery during differential reinforcement. Journal of Applied Behavior Analysis, 37, 213–218. Rooker, G. W., Jessel, J., Kurtz, P. F., & Hagopian, L. P. (2013). Functional communication training with and without alternative reinforcement and punishment: An analysis of 58 applications. Journal of Applied Behavior Analysis, 46, 708–722. Terrace, H. S. (1971). Escape from S-. Learning and Motivation, 2, 148–163. Thompson, T., & Gray, D. B. (1994). Destructive behavior in developmental disabilities. Thousand Oaks, CA: Sage. Tiger, J. H., & Hanley, G. P. (2004). Developing stimulus control of preschooler mands: An analysis of schedule-correlated and contingency-specifying stimuli. Journal of Applied Behavior Analysis, 37, 517–521. Tiger, J. H., Hanley, G. P., & Bruzek, J. (2008). Functional communication training: A review and practical guide. Behavior Analysis in Practice, 1, 16–23. Vollmer, T. R., & Athens, E. (2011). Developing function-based extinction procedures. In W. W. Fisher, C. C. Piazza, & H. S. Roane (Eds.), Handbook of behavior analysis. New York: Guilford Publishing.