569509 TECXXX10.1177/0271121415569509Topics in Early Childhood Special EducationDurand and Moskowitz research-article2015

Article

Functional Communication Training: Thirty Years of Treating Challenging Behavior

Topics in Early Childhood Special Education 2015, Vol. 35(2) 116–126 © Hammill Institute on Disabilities 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0271121415569509 tecse.sagepub.com

V. Mark Durand, PhD1 and Lauren Moskowitz, PhD2

Abstract Thirty years ago, the first experimental demonstration was published showing that educators could improve significant challenging behavior in children with disabilities by replacing these behaviors with forms of communication that served the same purpose, a procedure called functional communication training (FCT). Since the publication of that paper, hundreds of demonstrations of the effectiveness of this approach have been published. This article describes the original study, the basic steps involved in using FCT, and how this approach can be used to reduce challenging behavior among toddlers and preschoolers.

Keywords functional communication training, challenging behavior, preschoolers, toddlers, positive behavior support

The concept is a deceptively simple one: Much of the mis- behavior seen in children can be viewed as a form of com- munication. Philosophers and others have for thousands of years used this metaphor to explain why young children cry and scream (e.g., Plato, 1960; Rousseau, 1762/1979). However, it wasn’t until relatively recently that serious research was conducted to discover if this concept could be used to actually improve the behavior of children. Three decades ago, a study published by one of us (V.M.D.) and the late Edward (Ted) Carr was the first to document how challenging behavior could be significantly improved by teaching children to use alternative communicative responses to these problem behaviors (Carr & Durand, 1985). The rationale used in the study was that if we could determine the “messages” of the children’s behavior prob- lems and teach them to communicate the same message more appropriately, their problem behavior would not be necessary and would be reduced.

The Study

In this first demonstration, four children participated who were between the ages of 7 and 14, who had a variety of developmental disorders, and who had chronic and serious challenging behaviors (e.g., aggression, tantrums, self- injury). The first step was to determine the function(s) of their behavior problems—one of a set of procedures that are now generally referred to as functional behavioral assess- ment (FBA). Research that predated our study showed that some children would engage in problem behavior to seek attention from others (e.g., Lovaas, Freitag, Gold, & Kassorla, 1965) or to escape from difficult tasks (e.g., Carr,

Newsom, & Binkoff, 1980). We therefore designed differ- ent task settings to determine if teacher attention or the type of tasks they were presented would elicit behavior prob- lems. (It should be noted that other research identified addi- tional common messages such as a desire for favorite objects or activities [e.g., Durand & Crimmins, 1988] or to gain access to various forms of physical stimulation [e.g., Rincover, Cook, Peoples, & Packard, 1979].) The children in our study were observed in typical classroom situations, but we manipulated the amount of teacher attention they received (high vs. low levels of attention). In a second series of conditions, we kept the level of teacher attention the same but manipulated the difficulty of the tasks they were presented with (easy vs. difficult tasks).

What we observed in this experiment was that the chil- dren responded in different ways to these manipulations. Two children primarily became seriously disruptive when they were presented with difficult tasks, although levels of teacher attention had no measurable influence on their behavior. Using the communication metaphor, we inter- preted this to suggest that they were using their challenging behaviors to ask to escape from the task (“I don’t want to do this!”). One student was unaffected by changes in task dif- ficulty but engaged in more challenging behaviors when the amount of teacher attention was reduced. Again, this pattern

1University of South Florida, St. Petersburg, USA 2St. John’s University, Jamaica, NY, USA

Corresponding Author: V. Mark Durand, University of South Florida St. Petersburg, 140 Seventh Avenue S., St. Petersburg, FL 33701, USA. Email: [email protected]

117 Durand and Moskowitz

of responding could be interpreted to be a request for atten- tion (“Would you come here?”). Finally, the fourth student engaged in more problem behavior when either exposed to difficult tasks or to lower levels of teacher attention, sug- gesting that this participant used challenging behavior to communicate for different things at different times.

After these students participated in the FBA and the potential messages were identified for their behaviors, we tested the notion that teaching them to communicate these messages in a more appropriate way (in this study respond- ing verbally) would result in their behavior problems being reduced in frequency. We made the decision that, for the message to escape from the difficult task, we would not teach them to say “I don’t want to do this!” as this would interfere with their educational progress. Instead, we taught the phrase “I don’t understand,” because in a typical educa- tional situation this would be followed by help or prompts from a teacher. This type of help would presumably make the task easier to complete and therefore the student would not seek to escape the situation. For the attention-getting phrase, we chose to teach the response “Am I doing good work?” One of us (V.M.D.) informally observed students in several regular education classes and determined that teach- ers would be more likely to attend to students who were seeking feedback on their work than if students just requested the teacher’s attention. Knowing how others are likely to respond to a child’s communication is an important aspect of the successful implementation of functional com- munication training (FCT).

In the second phase of this study, we placed each of these same four students in both difficult task conditions and alternated that with reduced adult attention conditions. In some of these settings, we taught and prompted them to communicate the phrase that matched the situation that resulted in problem behavior (e.g., Prompting “I don’t understand” to students who misbehaved in the difficult task situation and “Am I doing good work?” for students seeking attention). In other conditions, we prompted them to communicate the phrase that did not match the function of their behavior problem (e.g., Prompting them to say “Am I doing good work?” for students trying to escape from the task and “I don’t understand” for students seeking atten- tion).1 We wanted to demonstrate that simply communicat- ing with others would not be sufficient to reduce behavior problems. Instead, we needed to show that FCT is success- ful only when the student successfully communicated the phrases that would match the function of their behavior problems. For example, if a student becomes upset during a difficult task and asks for and gets attention but not help on the task, this should not reduce behavior problems. In other words, it is not communication per se that is the key to reducing problem behavior but communicating for the things that they were requesting using their behavior problems.

The results of this second phase of our study supported our predictions. Only when each child made a request for attention and/or help that matched the function of his or her behavior problems did these behaviors substantially reduce in frequency. So by teaching children to request help, we were able to improve their chronic behavior problems that were being used to escape from tasks. Similarly, when we taught children to appropriately ask for adult attention, behavior problems that were being used to get attention improved significantly. Importantly, this first study showed that we could improve the severe and chronic behavior problems exhibited by these children in their classrooms with a strictly positive technique (Carr & Durand, 1985).

Mechanisms of Change

The mechanism of behavior change that underlies the reduction of problem behavior using FCT is the concept of “functional equivalence” (Carr, 1988; Durand, 1987). The assumption is that problem behaviors are maintained by a particular reinforcer or reinforcers (e.g., attention from oth- ers, escape from work). Theoretically, then, these behaviors can be replaced by other behaviors if these new behaviors serve the same function and are more efficient at gaining the desired reinforcers. As mentioned, FCT is presumed to reduce problem behavior because it involves teaching and reinforcing a replacement behavior that serves the same function.

The concept of “recruiting natural communities of rein- forcement” (Stokes, Fowler, & Baer, 1978) is central to the success of FCT (Durand, 1990b). In other words, the unique aspect of FCT over other interventions is that the student initiates the treatment rather than having another person observe a behavior problem and begin an intervention. This is a particularly important aspect of FCT when comparing the outcomes with FCT to other intervention strategies. For example, we conducted an experimental analysis of this aspect of FCT in a study comparing this treatment approach with another, common approach to reacting to challenging behavior—time-out from positive reinforcement (TO; Durand & Carr, 1992). In this study, we selected 12 students 4 to 5 years of age, all of whom engaged in a variety of chal- lenging behaviors (e.g., tantrums, self-injury, and aggres- sion). In addition, we screened for students whose challenging behaviors were being maintained by attention from others. The rationale for selecting students who only had attention-maintained problem behavior was to ensure that they would be appropriate for using TO.

Six students were randomly assigned to one of the two treatment conditions—FCT or TO. Just prior to introducing the treatments, the students were individually placed with teachers who had no knowledge of the study and who were instructed to work with the student on a task. No instruc- tions were given about how to react to behavior problems.

118 Topics in Early Childhood Special Education 35(2)

Treatment was then introduced by other teachers and was successful in reducing challenging behaviors for each child. Thus, the first finding was that both FCT and TO could suc- cessfully reduce challenging behavior. Then, the students were placed back with the teachers who were naïve to the treatment program. For the students who had received TO as a treatment, the students quickly found that TO as a con- sequence would not occur in this setting and they resumed their behavior problems. In contrast, the students who received FCT as a treatment used their communicative responses to request attention from the naïve teachers and the naïve teachers responded appropriately, maintaining the reduction of the challenging behavior. In short, the advan- tage of FCT was not solely in the initial reduction in chal- lenging behavior—which both treatments could produce—but also in its ability to prompt the teacher to reinforce them with what they were requesting (recruit nat- ural communities of reinforcement; Shoham-Vardi et al., 1996). In other words, FCT generalized to the untrained condition, whereas TO did not.

The Implications and Impact of FCT

FCT was developed in the early 1980s at a time when con- troversies erupted over the use of painful or humiliating interventions with persons with disabilities. The sometimes acrimonious nature of these discussions did little to further the solution to this issue, which was to provide an evidence base for non-aversive interventions that could be used to successfully treat severe challenging behaviors (Durand, 1990a). FCT fit with the zeitgeist of the time to develop research on interventions for even the most severe challeng- ing behaviors that would avoid relying on procedures that might be unpleasant or dehumanize the individual (Durand, 1987). This first study documented that a positive approach consistent with today’s efforts at positive behavior support (PBS) could effectively treat chronic and severe problem behavior.

Since its publication, this study has been cited more than 1,500 times and replicated in hundreds of other studies on a wide range of populations and types of problem behavior (for reviews, see Durand, 2012; Kurtz, Boelter, Jarmolowicz, Chin, & Hagopian, 2011; Tiger, Hanley, & Bruzek, 2008). There is a wide range of behaviors that have been targeted with FCT, including aggression and self-injury to elope- ment (i.e., running away; Lang et al., 2009) and inappropri- ate sexual behavior (Fyffe, Kahng, Fittro, & Russell, 2004). FCT is used in homes, schools and in the community by family members as well as a range of professionals (Dunlap, Ester, Langhans, & Fox, 2006; Durand, 1999; Kemp & Carr, 1995). Research on FCT has been successfully used with individuals across all ages (from infants/toddlers to older adults), developmental levels (from those with perva- sive needs for support to those with average or above

average cognitive abilities), language (Mancil, 2006; Petscher, Rey, & Bailey, 2009; Snell, Chen, & Hoover, 2006), and diagnoses such as autism spectrum disorder (ASD; for example, Brown et al., 2000), attention-deficit hyperactivity disorder (ADHD; for example, Flood & Wilder, 2002), and traumatic brain injury (Gardner, Bird, Maguire, Carreiro, & Abenaim, 2003). Our first study included children who could say a few words, but other studies have now shown that children who are non-verbal can use a variety of augmentative systems to succeed with FCT (Durand, 1999; Mirenda, 1997; Walker & Snell, 2013). FCT exceeds the American Psychological Association’s cri- teria for empirically supported treatments to be designated as a well-established treatment for challenging behavior for children with intellectual and developmental disabilities, including ASD (Chambless et al., 1996; Task Force Promoting Dissemination of Psychological Procedures, 1995; Wong et al., 2013).

Research With Toddlers and Preschoolers

Children’s problem behavior is often rated as the chief con- cern for early childhood educators, with the prevalence of problem behavior estimated at approximately 10% to 25% of young children (e.g., Joseph & Strain, 2003). Increasingly, researchers and policymakers have been paying more atten- tion to the problem behavior exhibited by young children, and there is an increasing concern regarding the long-term implications of such behavior (e.g., Dunlap, Strain, et al., 2006; New Freedom Commission on Mental Health, 2003). Importantly, there is a poor prognosis for children who exhibit early-onset behavioral problems. Without effective interventions to address problem behaviors and teach more appropriate replacement behaviors, toddlers and preschool- ers who exhibit such behavioral challenges may be at risk for long-term problems with socialization and academic success (e.g., Patterson, Reid, & Dishion, 1992). Evidence suggests that problem behaviors that begin in early child- hood persist from toddlerhood through adolescence and even adulthood. In fact, recent findings suggest that behav- ior problems are present in a recognizable form as early as 8 months of age (Lorber, Del Vecchio, & Smith Slep, 2014).

There is increasing demand for effective interventions for children at younger ages than have traditionally been targeted in earlier years (e.g., Dunlap, Ester, et al., 2006). Given the fact that FCT can be implemented by natural intervention agents such as parents, teachers, and teaching assistants (e.g., Schindler & Horner, 2005), and given that a child who is taught a functional communication skill can then recruit reinforcers without an interventionist needing to train other people how to respond, this makes it an ideal intervention approach for use with very young children in their homes and preschools.

119 Durand and Moskowitz

Research suggests that FCT is an effective strategy for addressing the behavior problems of young children. Studies using FCT with young children have targeted tod- dlers from 16 to 36 months of age (Dunlap, Ester, et al., 2006; Harding et al., 2009; Peck et al., 1996; Wacker et al., 2005; Winborn, Wacker, Richman, Asmus, & Geier, 2002) as well as preschoolers from 3 to 5 years of age (e.g., Olive, Lang, & Davis, 2008; Schindler & Horner, 2005), with most studies targeting children who span across both tod- dler and preschool years. Not only have researchers and therapists been able to effectively implement FCT with young children (Durand & Carr, 1992; Falcomata, Roane, Feeney, & Stephenson, 2010; Mildon, Moore, & Dixon, 2004; Peck et al., 1996; Winborn et al., 2002), but parents have been successful in using FCT with their toddlers and preschoolers in home settings (Berg, Wacker, Harding, Ganzer, & Barretto, 2007; Derby et al., 1997; Dunlap, Ester, et al., 2006; Harding et al., 2009; Mancil, Conroy, Nakao, & Alter, 2006; Moes & Frea, 2002; Olive et al., 2008; Rispoli, Camargo, Machalicek, Lang, & Sigafoos, 2014; Schieltz et al., 2011; Wacker, Harding, & Berg, 2008; Wacker et al., 2011) and teachers have had success using FCT in early childhood settings, preschools, and schools (Gibson, Pennington, Stenhoff, & Hopper, 2010; Lambert, Bloom, & Irvin, 2012; Schindler & Horner, 2005). In addition, although the majority of research on FCT has used verbal speech or sign language, there are several studies of FCT in young children that use “communication cards” or picture cards (e.g., Berg et al., 2007; Falcomata et al., 2010; Lambert et al., 2012; Mancil et al., 2006; Peck et al., 1996; Rispoli et al., 2014; Schieltz et al., 2011; Schindler & Horner, 2005; Wacker et al., 2005; Wacker et al., 2011; Winborn et al., 2002) or assistive technology such as a microswitch (e.g., Harding et al., 2009; Olive et al., 2008; Peck et al., 1996; Schieltz et al., 2011; Wacker et al., 2005; Wacker et al., 2011; Winborn et al., 2002). Such alternative forms of communication might be particularly important for young children with developmental delays.

The breadth of studies using FCT with young children is shown in Table 1, which includes 21 studies that used FCT exclusively with young children (ages 1–6 years old). Other studies include mixed-age groups with younger children and older children, but these studies are not included in this review. The table summarizes the participants’ age and diagnoses, the problem behavior targeted with FCT, the communicative response taught, and the interventionist who implemented FCT. Although FCT is efficacious, there are certain considerations to be taken into account when implementing FCT that are impor- tant to its effectiveness in natural settings. Two such consider- ations are the issues of schedule thinning and contextual fit.

Schedule Thinning

When teaching a child to use a communicative response, it is important that we initially reinforce him by giving him

what he is asking for every time he asks for it (i.e., provid- ing him with a “rich” or “dense” schedule of reinforcement, in which the reinforcer is easily obtained) so that he quickly acquires the skill (Durand, 1990b). If the reinforcer was too difficult to obtain in the early stages of attempting to use a new communication skill, the child might resort to using problem behavior to gain his reinforcer rather than using his new communication skill. However, although continuously reinforcing the new communicative skill at high rates will help the child learn the skill more quickly, in practice, it can be challenging for parents or teachers to have to respond to the child’s request every time he makes it. For example, if a child is requesting attention every minute, it can be irritat- ing or stressful for family members or teachers and likely impossible to sustain. Likewise, if a child is repeatedly requesting a break every 1 or 2 min, or requesting his toy every few minutes, it can keep him from completing tasks and result in reduced learning opportunities. Therefore, to address this problem, several studies have examined reduc- ing or “thinning” a schedule of reinforcement for a com- munication skill from a continuous schedule to a more intermittent schedule.

In one example, Rispoli and colleagues (2014) examined the effects of FCT plus extinction and schedule thinning for three preschool children with ASD (ages 3–4 years old) who displayed problem behavior when their rituals were interrupted. After a functional analysis (FA) showed that problem behavior was maintained by reinstatement of the interrupted routine, each child was taught the communica- tive response of saying, “I don’t want that” (or handing the interventionist a card with the “no” symbol) if they did not want their routine to be interrupted. Use of the communica- tive response resulted in temporary reinstatement of the ritual, whereas problem behavior did not result in attention or reinstatement of the ritual. The researchers gradually increased the time between the communicative response (e.g., “I don’t want that”) and the reinforcement (i.e., rein- statement of ritual) using a digital kitchen timer, such that the child could receive reinforcement for using his commu- nicative response only when the timer sounded. Reinforcement was available after 5 s of ritual interruption in the first phase, with the duration of ritual interruption gradually increased to 2 min. Results revealed that, follow- ing FCT plus extinction, problem behavior decreased and appropriate communication increased for all three children across each of the schedule-thinning phases. Although the children initially used the communicative response at high levels, their use of the response decreased (while problem behavior remained near zero) as schedule thinning pro- gressed (as the duration of the ritual interruption increased).

The researchers conceptualized the schedule thinning as possibly being similar to a “signaled demand fading” in that there are limited demands in the initial FCT plus extinction condition and the demands (i.e., duration of ritual interruption) are systematically increased before providing reinforcement

Table 1. Studies Using FCT With Young Children.

Citation n Participant ages (years) Diagnoses Problem behavior

Communicative response form Interventionist

Arndorfer, Miltenberger, Woster, Rortvedt, and Gaffaney (1994)

5 (n = 2 FCT) 2–3 Developmental speech delay, fetal alcohol syndrome

Aggression, SIB, disruptive behaviors

Verbal Parent, researcher

Berg, Wacker, Harding, Ganzer, and Barretto, (2007)

4 4–5 DD, speech and language disorders

Aggression, property destruction, non- compliance

Verbal, picture card (touching card)

Parent

Blair, Fox, and Lentini (2010) 3 (n = 1 FCT) 3 Language developmental delay

Tantrums, aggression Did not specify (multicomponent PBS plan including “teaching skills”)

Teacher, teaching assistant

Derby et al. (1997) 4 2–5 DD, ID, language delay, visual impairment, asthma, cerebral palsy, epilepsy

SIB, property destruction, non- compliance, tantrums

Verbal, sign language Parent

Duda, Dunlap, Fox, Lentini, and Clarke (2004)

2 (n = 1 FCT) 3 Down syndrome Aggression, disruptive behavior

Verbal, gestural (tap arm) Teacher, paraprofessional

Dunlap, Ester, et al. (2006) 2 2 Speech and language delay Aggression, disruptive behaviors

Verbal Parent

Durand and Carr (1992) 12 (n = 6 in FCT group)

3–5 ASD, ID, ADHD, developmental language disorder

Aggression, tantrums, property destruction,

Verbal Researcher

opposition Falcomata, Roane, Feeney, and

Stephenson (2010) 1 5 ASD Elopement Picture card (touching

card) Therapist

Gibson, Pennington, Stenhoff, and Hopper (2010)

1 4 ASD Elopement Verbal, hand raising Teacher

Harding et al. (2009) 3 1–3 DD, language delay, disruptive behavior disorder

SIB, aggression, property destruction

Verbal, picture card (touching card), sign language, assistive technology

Parent

Lambert, Bloom, and Irvin (2012) 3 3–4 DD Aggression, tantrums Picture card (touched teacher with card)

Teacher

Mancil, Conroy, Nakao, and Alter (2006)

1 4 PDD Tantrums Verbal, picture card (hand to researcher)

Researcher, parent

Mildon, Moore, and Dixon (2004) 1 4 ASD Aggression, disruptive behaviors, non-

Verbal Researcher

compliance Moes and Frea (2002) 3 3 ASD Aggression, disruptive

behavior Verbal, sign language Parent, sibling

Olive, Lang, and Davis (2008) 1 4 ASD Elopement, disruptive behavior

Assistive technology Parent

(continued)

120

Table 1. (continued)

Participant Communicative response Citation n ages (years) Diagnoses Problem behavior form Interventionist

Peck et al. (1996) 5 1–4 DD, language delay, short- bowel syndrome, severe pulmonary hypertension, atrial septal defect, bronchopulmonary dysplasia, failure to thrive

SIB, aggression, property destruction, non-compliance, screaming

Verbal, picture card (touching card), assistive technology, choice- making procedures (i.e., choosing an appropriate object for getting attention, eating)

Parent, researcher, nurse

Rispoli, Camargo, Machalicek, Lang, and Sigafoos (2014)

Schieltz et al. (2011)

3 3–4 ASD, PDD-NOS Aggression, tantrums Verbal, picture card (exchange of card)

Parent, therapist

10 1–4 ASD, DD, FXS, ID, Peter’s anomaly, Mecke’s diverticulum reflux, and Viral-induced asthma

SIB, aggression, property destruction, and disruptive behavior

Verbal, picture card, sign language, assistive technology

Parent

Schindler and Horner (2005) 3 4–5 ASD, charge syndrome Aggression and Picture card (pointing to Parent, teacher, disruptive behavior picture) teaching assistant

Wacker et al. (2005) 25 1–6 ASD, DD, PDD, speech and language delay, Rett syndrome, cerebral palsy, seizure disorder, Soto syndrome, behavior disorder, ADHD, expressive communication disorder, Lesch-Nyhan syndrome, bronchial pulmonary dysplasia, asthma, microcephaly, vision and hearing impairment, articulation disorder

SIB, aggression, property destruction

Verbal, picture card (touching card), sign language, assistive technology

Parent

Wacker, Harding, and Berg (2008) 3 5 ASD, FXS, PDD, ID, ODD SIB, aggression, property destruction

Verbal Parent

Wacker et al. (2011) 8 2–4 ASD, FXS, DD, mild/ moderate ID

SIB, aggression, property destruction

Verbal, picture card (touching card), sign language, assistive technology

Parent

Winborn, Wacker, Richman, Asmus, and Geier (2002)

2 2.5 Developmental delays and seizures

SIB, aggression, property

Destruction, tantrums, non-compliance

Verbal, picture card, assistive technology, gestural (shake head “no”)

Therapist

Note. FCT = functional communication training; SIB = self-injurious behavior; DD = developmental disability; PBS = positive behavior support; ID = intellectual disability; ASD = autism spectrum disorder (might be referred to as “autism” or “autistic disorder” in the study); PDD-NOS = pervasi