Research in Autism Spectrum Disorders 19 (2015) 18–31 Contents lists available at ScienceDirect Research in Autism Spectrum Disorders Journal homepage: http://ees.elsevier.com/RASD/default.asp Exposure and response prevention therapy with cognitive defusion exercises to reduce repetitive and restrictive behaviors displayed by children with autism spectrum disorder Heidi J. Eilers *, Steven C. Hayes Department of Psychology, University of Nevada, Reno, NV, United States A R T I C L E I N F O A B S T R A C T Article history: Received 26 November 2014 Accepted 4 December 2014 Available online 20 July 2015 Repetitive and restrictive behaviors are disruptive in children with autism spectrum disorder. Exposure is an evidence-based approach to these problems and is hypothesized to work in part by enabling rule testing that undermines rule control. The present study undermined rule control more directly through cognitive defusion exercises. Experiment 1 used a multiple baseline across participants design to assess the effects of a cognitive defusion exercise in the form of word repetition and exposure on problem behavior associated with repetitive and restrictive behaviors observed in children with autism spectrum disorder. All 3 participants demonstrated a decrease in the percentage of problem behavior following the implementation of treatment which remained at near zero during a 3-month follow-up. Experiment 2 used an alternating treatments design to compare a cognitive defusion exercise and exposure to a control exercise and exposure. All except 1 of the participants displayed larger and quicker decreases in problem behavior during the cognitive defusion exercise condition compared to the control exercise condition. The results suggest that cognitive defusion exercises can enhance the treatment effects of exposure to decrease problem behavior associated with repetitive and restrictive behaviors. ß 2015 Elsevier Ltd. All rights reserved. Keywords: Cognitive defusion Experiential avoidance Exposure and response therapy Relational frame theory Repetitive and restrictive behaviors The presence of repetitive and restrictive behaviors is a core feature of Autism Spectrum Disorder (ASD; American Psychiatric Association, 2013). The level of complexity of behaviors included in the repetitive and restrictive category ranges from simple discrete repetitive motor and vocal stereotypy to complex rituals and routines. Complex or high-order repetitive and restrictive behaviors have been described as rigidities and rituals, insistence on sameness, difficulty with transitions, and circumscribed interests (Lewis & Bodfish, 1999; Turner, 1999). Despite being a core area of ASD, research in the area of assessment and intervention for high-order repetitive and restrictive behaviors is quite sparse compared to other features of ASD (Bodfish, 2004; Turner, 1999). High-order repetitive and restrictive behaviors have been referred to as ‘‘obsessive’’ and ‘‘compulsive,’’ but because many individuals with ASD are unable to describe private events reliably, it is difficult to assess the behavior as foreign or * Corresponding author. Present address: Easter Seals Southern California, 2121 Towne Centre Place, Ste 330, Anaheim, CA 92806, United States. Tel.: +1 7602243409. E-mail address: [email protected] (H.J. Eilers). http://dx.doi.org/10.1016/j.rasd.2014.12.014 1750-9467/ß 2015 Elsevier Ltd. All rights reserved. H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 19 unwanted, which is a core feature of obsessive–compulsive disorder (Zandt, Prior, & Kyrios, 2007). There are topographical differences between the high-order repetitive thoughts and behaviors reported with ASD and the thoughts and actions associated with obsessive–compulsive disorder. Those with OCD are more likely to report repetitive thoughts with aggressive, contamination, sexual, religious, symmetry, and somatic thematic content, and to engage in compulsive checking, cleaning, and counting. Those in the ASD group are more likely to report repetitive thoughts with a need to know or remember and hoarding content; repetitive behaviors are more likely than thoughts and consist of repetitive ordering, hoarding, touching, tapping, rubbing, self-damaging, self-mutilating, and telling or asking (Farrugia & Hudson, 2006). Nevertheless, repetitive and restrictive behaviors are associated with high levels of anxiety (Tantam, 2000; van Steensel, Bögels, Perrin, 2011), suggesting that they may serve similar functions as behaviors observed with OCD. Exposure with Response Prevention (ERP) has the largest overall effect size in the treatment of OCD (for a review see RosaAlcázar, Sánchez-Meca, Gómez-Conesa, & Marı́n-Martı́nez, 2008). Despite the success of ERP, there are often high rates of treatment refusal and dropout (Abramowitz, Taylor, McKay, 2009). Similarly, research examining the use of prevention or interruption of repetitive and restrictive responses with ASD has demonstrated that increases in aggression and disruption can occur and be reinforced by access to rituals and routines (Hausman, Kahng, Farrell, & Mongeon, 2009; Kuhn, Hardesty, & Sweeney, 2009). Thus, there is a need for additional treatment elements to be developed that reduce problematic behaviors (e.g., aggression, disruption, tantrums) during ERP in the ASD population while maintaining the impact of ERP. One possible element might be to find alternative ways to target the key processes of change in ERP. Exposure can be defined as ‘‘the organized presentation of previously repertoire-narrowing stimuli in a context designed to ensure repertoire expansion’’ (Hayes, Strosahl, & Wilson, 2012; p. 284). One means by which repertoire expansion may occur is through contact with contingencies that go beyond the rules derived by the individuals engaging in obsessive–compulsive behavior. For example, an individual may have derived the rule, ‘‘If I do not follow this routine, something horrible will happen.’’ By exposing the individual to activities alternative to the routine, the domination of this rule over action may be reduced. Another way of producing that effect is suggested by Relational Frame Theory (RFT; Hayes, Barnes-Holmes, & Roche, 2001). According to RFT, the behavioral impact of verbal rules can be changed either by altering the form or function of verbal networks through the use of relational cues. Cognitive defusion exercises such as those used in Acceptance and Commitment Therapy (ACT; Hayes et al., 2012) attempt to apply this idea to practical contexts by altering functional contextual cues so that verbal events are not taken literally, thus undermining the automatic behavior regulatory impact of verbal events. Consider the verbal rule: ‘‘In 60 seconds, this box will explode.’’ If that line is said by an actor on a stage, no one in the audience is likely to run from the theater. If it is said by a terrorist who stormed into a house, victims might indeed attempt to run or grab the box and throw it out the window. The difference is not to be found in what the words mean; the local relational networks they occasion are nearly identical. The difference is in functional cues (i.e., terrorist versus actor) that support a transformation of stimulus functions from the before–after relation and the term ‘‘explode.’’ These function in turn to motivate escape or avoidance actions. If we return to our parallel example of obsessive thinking, as in, ‘‘If I do not follow this routine, something horrible will happen,’’ we can apply RFT thinking by deliberately establishing cues that help the person notice their unfolding process of thinking in a nonthreatening context. Functionally speaking, this would be more like being a theater attendee than a terrorist victim. A person might sing, ‘‘Something horrible will happen,’’ as if in a musical production, or say it in a silly voice, such as that of Mickey Mouse, evoking the functions of songs or cartoons. The words might be said very slowly, evoking their auditory properties and diminishing the dominance of meaning that is entirely dependent on derived relational responding. The thought, ‘‘Something horrible will happen,’’ might be said in a child’s voice, evoking compassionate understanding, or it might be written on leaves and allowed to float by, evoking dispassionate observation. One of the simplest and most studied defusion methods was originally suggested by Titchener (1916, p. 425): repeating a single word aloud until the context needed for the word to have literal meaning diminishes. This procedure has been demonstrated to be effective at reducing emotional discomfort and believability with contamination-related thoughts (Watson, Burley, and Purdon, 2010) and other negative self-referential statements (Masuda, Hayes, Sackett, & Twohig, 2004; Masuda, Feinstein, Wendell, & Sheehan, 2010; Masuda, Twohig, et. al., 2010). All defusion techniques use functional cues to reduce problematic transformation of stimulus functions that are entirely dependent on a derived relational network and the beliefs they instantiate, and instead establish a broader set of functions such as observation, appreciation, humor, compassion, play, curiosity, and so forth. The present study examined the impact of ERP, when used with a combination of word repetition and saying thoughts in a silly voice, on problematic repetitive behaviors of children diagnosed with ASD. 1. Experiment 1 1.1. Participants and setting Three participants were included in this study. Participants were referred to the Center for Autism, Research, Evaluation, and Service (CARES) for behavioral intervention services regarding behavioral excesses and deficits associated with ASD. Eligibility for the study was determined by interviewing the participant’s primary caregiver and conducting direct observation. All participants were diagnosed with ASD and engaged in behavioral excesses associated with at least one of the following categories: (a) difficulty with transitions/change or the presence of novel stimuli (e.g., changing from one activity to another, presence or absence of a person, presence of particular objects), (b) insistence on performing tasks in a 20 H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 stereotyped manner (e.g., lining up blocks in a particular order), or (c) circumscribed interest (e.g., excessively talking about and playing with Superman paraphernalia). Participants were required to demonstrate the ability to repeat three to fiveword phrases and independently use at least two to three-word phrases to request at least three desired items or activities (e.g., ‘‘Red ball, please’’). All sessions were conducted by trained CARES therapists who completed at least 10 h of classroom education in autism, behavioral intervention procedures, and data collection procedures, in addition to at least 20 h of on-the-job training to implement behavioral intervention programs for children diagnosed with ASD. CARES therapists also received approximately 2 h per week of ongoing supervision and training in behavioral intervention and data collection procedures by licensed psychologists and/or board certified behavior analysts (BCBA). Michael was a 3-year-old boy diagnosed with autistic disorder. He was able to express his desire for at least 20 items using two to three-word phrases (e.g., ‘‘Juice, please’’) and regularly communicated using three to five-word phrases or sentences. Michael engaged in excessive crying when playing with his toys. He also cried whenever his toys broke, fell apart, or if he could not manipulate the toys in a particular manner. His mother reported that she was often unable to console him or calm him down for several hours when he cried as a response to his toys not working the way in which he wanted. She identified his train tracks as a primary problem. According to his mother, Michael would cry for several hours if his train track disconnected or fell apart while he was playing with it. His mother reported that he would not stop crying even if she immediately attempted to fix the train track. Rather, he would repeat statements about the train track breaking and continue to cry. Michael’s sessions were conducted approximately five days per week for 2 h on his living room floor. Sarah was a 4-year-old girl diagnosed with autistic disorder. Sarah expressed her desire for at least 30 items using four to five-word phrases (e.g., ‘‘I want to play dolls’’) and regularly communicated using five to six-word phrases or sentences. Sarah engaged in excessive crying, aggression, disruption, and perseverative speech. Sarah’s mother reported that Sarah was most likely to engage in problem behavior when she was disrupted from playing in a stereotyped manner (e.g., stacking blocks in a particular order) or when another person attempted to play with her and did not play according to her rules. For example, Sarah would engage in problem behavior if another person attempted to draw on the right side of the chalkboard. Sarah would draw a line through the middle of the chalkboard and would only draw on the right side of the chalkboard and instructed others that she was the only person allowed to draw on the right side. She would engage in problem behavior if another person attempted to draw on the right side of the chalkboard. Sarah was also reported to engage in problem behavior when she did not get to take the first turn when playing board games. Sarah’s sessions were conducted in a playroom in her house approximately three days per week for 2 h. Jacob was a 7-year-old boy diagnosed with autistic disorder. Jacob expressed his desire for at least 20 items using three to four-word phrases (e.g., ‘‘I want ball’’) and regularly communicated using four to five-word phrases. Jacob engaged in problem behavior (e.g., disruption, aggression, crying, yelling, perseverative speech) when he was presented with or instructed to eat novel foods. His mother reported that she no longer attempted to have unwanted or novel foods on the table when he ate because of his inappropriate behavior. Jacob’s sessions were conducted in the dining area of his home. Sessions were conducted approximately two days per week for 2 h. 1.2. Response measurement and interobserver agreement 1.2.1. Dependent variables Frequency of problem behavior per opportunity was recorded for all participants. For Michael, an opportunity was defined as any time a piece of the train track was removed or disconnected. For Sarah, opportunities were defined as any time she was: (a) playing a game and did not get to start first, (b) drawing on the chalkboard and another person attempted to take turns drawing on the right side of the chalkboard, and (c) stacking blocks and another person placed a block on the stack. For Jacob, opportunities were defined as any time he was presented with a novel food and instructed to touch it. All three participants engaged in crying (an instance was scored when it occurred for longer than 3 s and more than 5 s between occurrences), perseverative speech (defined as repeating a statement more than 3 times with less than 5 min between offset and onset of each statement), and attempts to escape (defined as any attempt to remove oneself from the room in the case of Sarah, or area designated by a chair in the case of Jacob, and a 10 12 ft area in the case of Michael). Sarah and Jacob additionally engaged in aggression (defined as any attempt to hit, bite, scratch, kick, push, or grab hair) and disruption (defined as any attempt to throw, knock down, destroy, or hide objects), and Jacob engaged in yelling (defined as any vocalizations above normal conversational levels and more than 5 s between occurrences). All definitions were written down and each data collector reviewed the written description of the behaviors prior to participating in data collection training. Using a paper and pencil data recording system, frequency of problem behavior per opportunity was recorded and converted to a percentage occurrence by dividing the number of opportunities with problem behavior by the total number of opportunities presented and multiplying by 100%. Data collectors were trained by simultaneously but independently collecting data with someone already trained in data collection and were required to have 90% agreement during a 2-h observation period. Interobserver agreement (IOA) was assessed by having two observers simultaneously but independently collect data. Agreement was calculated using the exact count-per-interval method (Cooper, Heron, & Heward, 2007). Sessions were divided into 5-min intervals and an agreement was scored during each interval if the same number was scored for the frequency of target behavior and as a disagreement if the numbers were different. Percentage agreement was calculated by H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 21 dividing the number of agreements by agreements plus disagreements and multiplying by 100%. IOA was collected during 77%, 33%, and 36% of sessions and averaged 97%, 99%, and 99% (range, 93% to 100%) for Michael, Sarah, and Jacob, respectively. 1.2.2. Treatment integrity Treatment integrity was evaluated by having an independent observer record yes or no to indicate if the therapist had the participant engage in the cognitive defusion exercise for at least 30 s prior to exposure. Then, during the exposure to target stimuli condition, the frequency of exposures was recorded as opportunities during each 5-min interval. Therapists were required to provide a specified number of exposure opportunities during each session as outlined in the procedures below. Treatment integrity was 100% for all therapists. IOA was assessed by having a second observer record the occurrence of therapist behaviors for 70% of Michael’s sessions, 29% of Sarah’s sessions, and 50% of Jacob’s sessions. Percentage of agreement was calculated for participation in the defusion exercise by dividing the number of agreements by agreements plus disagreements and multiplying by 100%. For exposure opportunities, an agreement was scored during each 5-min interval if the same number was scored for the frequency of opportunities and as a disagreement if the numbers were different. IOA was calculated by dividing the agreements by agreements plus disagreements and multiplying by 100%. Agreement remained at 100% across sessions for all participants. 1.3. Mini preference assessment Participants were asked to name five highly preferred items/activities prior to the start of the study. Prior to each exposure, the items/activities were placed in front of the participants (activities were represented pictorially or in written form) and participants were asked to choose one of the five items/activities (e.g., ‘‘Which one do you want?’’). Participants were provided noncontingent access to the item following exposure to the target situation/stimuli. 1.4. Experimental design and sequence A nonconcurrent multiple baseline across participants design was used to evaluate the effectiveness of cognitive defusion exercises and exposure to reduce the percentage of problem behaviors displayed by individuals diagnosed with ASD when exposed to the target situation/stimuli. 1.4.1. Baseline Participants were presented with target situations/stimuli throughout the 2-h session. Michael was presented with 30 opportunities to engage in problem behavior per session by being instructed to play trains for 5 min and the therapist disconnecting the train track approximately every 30 s. The train track was placed back together contingent on problem behavior. Sarah was presented with four to seven opportunities to engage in problem behavior per session during one of three target situations wherein the therapist provided an opportunity to engage in problem behavior. In one situation, Sarah was instructed to play a board game, and the therapist took the first turn of the game. Contingent on problem behavior, the therapist started the game over and allowed Sarah to go first. In the second situation, Sarah was instructed to play blocks, and the therapist attempted to take turns stacking the blocks. Contingent on problem behavior, the therapist stopped attempting to place blocks on the stack. Last, Sarah was instructed to draw pictures on the chalkboard. The therapist attempted to take turns drawing and drew a picture on the right side of the board. Contingent on Sarah engaging in problem behavior, the therapist stopped drawing. Jacob was presented with 10 opportunities to engage in problem behavior per session by being instructed to touch a novel food item. The food item was removed contingent on problem behavior. 1.4.2. Treatment The therapist implemented a defusion exercise for 30 s prior to exposure to the target situation/stimuli. The statements targeted for cognitive defusion were those hypothesized to have a controlling relation over problematic repetitive behavior for each participant when exposed to situations that normally occasioned the problem behavior. Specifically, the therapist instructed the participant to repeat the phrase in ‘‘silly voices’’ by saying, ‘‘You’re going to say (the phrase) in silly voices like me.’’ The therapist and participant repeated the phrase in silly voices for 30 s. Immediately following the defusion exercise, the therapist presented the participant with the target situation/stimuli. All participants were prompted to continue engaging in the activities using a 3-step graduated prompting hierarchy (verbal instruction, model or gestural prompt, and physical guidance). After participants completed the exposure session, they were provided with the preferred item chosen from the mini preference assessment. Michael was instructed to repeat the phrase, ‘‘The train track broke,’’ and was then exposed to 5 min of playing trains while the therapist disconnected the track every 30 s. Sarah was instructed to repeat one of the following three statements: (a) ‘‘I want to go first,’’ then she was instructed to play a board game, and the therapist took the first turn, (b) ‘‘I don’t want to share,’’ then she was instructed to play blocks while the therapist took a turn stacking a block after each block Sarah stacked, or (c) ‘‘I don’t want to take turns,’’ then she was instructed to draw pictures on the chalkboard, and the therapist took a turn H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 22 drawing a picture on the chalkboard following each drawing Sarah completed. Jacob was instructed to say, ‘‘I don’t like (name of food),’’ then he was presented with the item and instructed to touch the item. 1.4.3. Follow-up The 3-month posttreatment follow-up sessions were identical to baseline except that the exposure situation was not removed contingent on the occurrence of problem behavior. Following sessions, each participant was given access to the preferred stimulus chosen prior to the session regardless of how the participant performed. 1.5. Results and discussion Percentage of problem behaviors are depicted in Fig. 1. All three participants displayed a reduction in the percentage of opportunities with problem behavior following the implementation of a cognitive defusion exercise and exposure therapy and maintained near zero levels at a posttreatment 3-month follow-up. The mean percentage of problem behavior for Michael during baseline was 34% of opportunities (range, 0% to 94%). The mean percentage of problem behavior during treatment was 24% (range, 0% to 44%). Michael maintained zero levels of responding at a 3-month posttreatment follow-up. Defusion plus Exposure Percentage of Opportunities with Problem Behavior Baseline Follow-Up 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Michael 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Sarah 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Jacob 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Sessions Fig. 1. Percentage of opportunities with problem behavior across Michael (top panel), Sarah (middle panel), and Jacob (bottom panel). H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 23 The mean percentage of problem behavior for Sarah during baseline was 69% (range, 20% to 100%). The mean percentage of problem behavior during treatment was 5% (range, 0% to 17%). Sarah maintained zero levels of responding at a 3-month posttreatment follow-up. The mean percentage of problem behavior for Jacob during baseline was 69% (range, 38% to 100%). The mean percentage of problem behavior during treatment was 6% (range, 0% to 12%). The mean percentage of problem behavior for Jacob at a 3month posttreatment follow-up was 7% (range, 0% to 13%). The results support the hypothesis that repetitive and restrictive behaviors are at least partially maintained by the functional impact of private verbal events. Exposure to the target situation/stimuli and participation in cognitive defusion exercises resulted in reductions in problem behavior across all participants, and reductions remained at near zero levels at a 3-month posttreatment follow-up indicating that the effects of treatment were also maintained over time. At the 3-month posttreatment follow-up, Michael’s mother reported that she didn’t observe problem behavior when he played with the trains, and he occasionally repeated the phrase used during the study when a piece of the train broke. Michael’s mother reported that he was more willing to experience novel stimuli/situations, but he required verbal reminders to use the exercise with new situations. Sarah’s mother reported that Sarah didn’t engage in problem behavior when exposed to small changes or novel stimuli/ situations but continued to engage in problem behavior when major changes to her schedule occurred, such as going on vacation or starting/stopping school. Sarah’s mother reported that she did not observe Sarah engage in the word repetition exercise independently; however, she noticed a decrease in intensity and duration of problem behavior when Sarah was verbally prompted to say repetitive phrases in song form prior to exposure. Jacob’s mother reported that he still engaged in problem behavior when exposed to novel food items, but the intensity and duration have decreased. Jacob’s mother reported that he had increased the number of foods he was willing to consume since he participated in the treatment and was willing to tolerate novel foods present on the table. Jacob’s mother reported that she had not observed him engaging in the word repetition exercise outside of treatment sessions. One weakness of Experiment 1 was that cognitive defusion exercises were used in combination with exposure therapy across all participants, making it unclear whether the exercises were responsible for any of the treatment effects. Further investigation is warranted to examine the benefits of the cognitive defusion exercises in decreasing problem behavior associated with repetitive and restrictive behaviors. Experiment 2 was conducted to compare cognitive defusion exercises to highly similar control exercises so as to examine the effects of word repetition exercises on the reduction of problem behavior. 2. Experiment 2 2.1. Participants and setting Four participants referred to CARES for behavioral intervention services for behavioral excesses associated with ASD were included in this study. The same criteria used in Experiment 1 were used to determine eligibility for Experiment 2. Jacob was a participant in Experiment 1 and also participated in Experiment 2. Jacob engaged in problem behavior when he was playing with chalk. Jacob lined up the chalk in patterns and smelled each chalk piece before lining them up. Jacob would yell, cry, and engage in aggression, disruption, and perseverative speech if another person attempted to prevent him from smelling the chalk, moved any of the pieces of chalk, or if any of the chalk pieces broke. Sessions were conducted in Jacob’s family room on the floor approximately two to three days per week. John was a 5-year-old boy diagnosed with pervasive developmental disorder, not otherwise specified (PDD-NOS). John regularly communicated using three to four-word phrases, including when making requests, and was able to request at least 15 items (e.g., ‘‘I want candy’’). John engaged in problem behaviors (e.g., crying, perseverative speech, running away) when he was presented with any type of statue. This was a significant problem for John and his family because his father collected movie memorabilia and had several statues that he had to hide from John because John would cry when he saw them. Sessions were conducted in John’s bedroom two to three days per week. Owen was a 5-year-old boy diagnosed with ASD. Owen regularly communicated using three to four-word phrases, including when making requests, and was able to request at least 10 items (e.g., ‘‘I want ball’’). Owen engaged in problem behavior when his routine was changed. He regularly attended CARES behavioral intervention sessions in his room and began engaging in problem behaviors (e.g., crying, whining, and verbal protests) when sessions were moved to another location to work on generalization of newly acquired skills. He refused to go downstairs to work on skills in another setting, and his progress was being hindered by his resistance to move to another location. Sessions were conducted two to three days per week in Owen’s house. Sessions began upstairs in his play room and continued with a transition downstairs to the living room and/or dining room. Cameron was a 6-year-old boy diagnosed with ASD. Cameron regularly communicated using four to five-word phrases, including when making requests, and was able to request at least 50 items (e.g., ‘‘I want to play cards’’). Cameron engaged in problem behaviors (e.g., aggression, disruption, crying, yelling, perseverative speech) when instructed to play games or activities in new ways. Cameron insisted on playing games and activities according to his own rules. Any attempts to change the way he played games or activities resulted in problem behavior. Sessions were conducted two to three days per week in Cameron’s bedroom. 24 H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 2.2. Response measurement and interobserver agreement 2.2.1. Dependent variables Data were collected both on problem behaviors and appropriate play using a 10-s partial interval method. In terms of problem behaviors, all participants engaged in crying and perseverative speech as defined in Experiment 1. Jacob and Cameron engaged in aggression, disruption, and yelling as defined in Experiment 1. John and Owen engaged in elopement (defined as attempts to leave the room, attempts to hide under his blankets or on the side of his bed, in the case of John and as moving further than 1 ft away from the therapist when walking downstairs, in the case of Owen). Jacob and Owen engaged in verbal protests (defined as saying, ‘‘No,’’ ‘‘Don’t do that,’’ ‘‘Stop,’’ in the case of John, and verbal statements refusing to go downstairs, in the case of Owen). Jacob’s appropriate play behavior was defined as sitting within reach of the chalk and manipulating the chalk at least once per 10-s interval. John’s appropriate play was defined as sitting within 2 ft of the statue and touching one of his toys at least once per 10-s interval. Owen’s appropriate play was scored if he was compliant walking downstairs (defined as following instruction to go downstairs within 5 s of the delivery of the instruction) and played hide-and-seek with the therapist in the downstairs living room and/or dining room (defined as counting, looking for the therapist, or hiding at least once during the 10-s interval). Cameron’s appropriate play was defined as sitting within reach of the game and touching the game piece during his turn at least once per 10-s interval. All definitions were written down and each data collector reviewed the written description of the behaviors prior to participating in data collection training. Using a paper and pencil data recording system, each topography of problem behavior was scored separately per 10-s interval. Percentage of intervals with problem behaviors as well as percentage of intervals with appropriate play was calculated separately by dividing the number of intervals with problem behavior or appropriate play by the total number of intervals and multiplying by 100%. IOA was assessed by having two observers simultaneously but independently collect data. Percentage agreement was calculated using the interval-by-interval method (Cooper et al., 2007) by dividing the number of 10-s intervals with agreement by the total number of intervals and multiplying by 100%. Agreement was calculated for 81%, 31%, 90%, and 27% of sessions and averaged 97%, 99%, 98%, and 98% (range, 90% to 100%) for Jacob, John, Owen, and Cameron, respectively. 2.2.2. Treatment integrity Treatment integrity was evaluated by having an independent observer record yes or no to indicate if the participant was exposed to the cognitive defusion exercise using relevant phrases or using the control irrelevant phrases, for at least 30 s, prior to being exposed to the target situation/stimuli. Whether or not the therapist provided exposure to the target situation/ stimuli was also recorded using a whole interval 10-s recording system. Treatment integrity was 100% for all therapists. IOA was assessed by having a second observer record the occurrence of IV behaviors for at least 30% of the sessions. IOA for participation in the cognitive defusion exercise was calculated using the same procedure as Experiment 1. IOA for exposure was calculated by dividing the number of intervals with agreements by the number of intervals with agreements plus disagreements and multiplying by 100%. Agreement was assessed for 81% of Jacob’s sessions, 31% of John’s sessions, 90% of Owen’s session, and 27% of Cameron’s sessions. Agreement was 100% across all sessions for all participants. 2.3. Social validity A 5-point Likert Scale (strongly agree to strongly disagree) with 12 statements about the treatment (see Table 1) was administered to the primary caregivers and behavioral therapists working with the participants. There were a total of 6 raters (3 caregivers and 3 behavioral therapists). 2.4. Mini preference assessment Procedures for the mini preference assessments were identical to Experiment 1. 2.5. Experimental design and sequence An alternating treatments design was used to compare the effects of exposure plus a control exercise and exposure plus a cognitive defusion exercise on the percentage of problem behaviors and appropriate play displayed by participants when exposed to the target situation/stimuli. The order of conditions was determined by a flip of a coin, with which heads resulted in the exposure and cognitive defusion condition and tails resulted in the exposure and control exercise being conducted. Conditions were presented in a semi-random order in which neither condition was presented more than two consecutive sessions. 2.5.1. Exposure and cognitive defusion exercise Prior to the exposure session, the participant engaged in the 30-s defusion exercise. The therapist instructed the child to repeat the phrase in ‘‘silly voices’’ for 30 s prior to being exposed to the target situation/stimuli. The therapist gave the H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 25 Table 1 Treatment acceptability rating scale. Statement Number of responses SA A N D SD 1. I believe it is important to change the problem behaviors targeted with this treatment 2. The treatment is an acceptable way to change the child’s problem behavior 3. I am willing to use this treatment to change the child’s problem behavior 4. I believe it is acceptable to use the treatment, even though the child is unable to consent to the treatment 5. I like the procedures used in this treatment 6. I believe the treatment is an effective way to reduce the child’s problem behavior 7. I believe the treatment is likely to result in permanent improvement 8. I believe it is acceptable to use this treatment with children who cannot choose treatments for themselves 9. I believe the child experienced discomfort during the treatment 10. I believe the child benefited from this treatment 11. I believe the procedures used in this treatment are reasonable 12. Overall, I have a positive reaction to this treatment 4 4 4 4 4 5 2 4 0 3 4 4 2 2 2 2 1 1 3 2 2 3 2 2 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Note. Statements are located in the left column. The number of responses to each statement are located in the right column. SA (strongly agree), A (agree), N (neutral), D (disagree), and SD (strongly disagree). instruction, ‘‘You’re going to say (the phrase) in silly voices like me.’’ Then, the therapist modeled the phrase in a silly voice. The therapist repeated the model if the child stopped saying the statement at any time. Following completion of the exercises, participants were immediately exposed to the target situation/stimuli for 5 min. Participants were given access to the preferred item identified in the mini preference assessment following exposure to the target situation/stimuli. Cognitive defusion exercises for Jacob consisted of repeating the phrase, ‘‘smell chalk,’’ for 30 s using silly voices. Exposure for Jacob consisted of the presentation of chalk, with at least one broken piece. He was prevented from smelling the chalk by physically guiding his hands back down, and approximately every 30 s, the therapist moved one of the pieces of chalk and placed it in another position on the floor. Cognitive defusion exercises for John consisted of repeating the phrase, ‘‘I don’t like the statue,’’ in silly voices. Exposure for John consisted of placing a statue on the table located in his room. John was instructed to play with his toys and remain at least 2 ft from the statue. Cognitive defusion exercises for Owen consisted of repeating the phrase, ‘‘I am going downstairs,’’ in silly voices. Exposure for Owen consisted of transitioning from his bedroom, walking downstairs, and entering the living room. Cognitive defusion exercises for Cameron consisted of repeating the phrase, ‘‘I don’t like to play the wrong way,’’ in silly voices. Exposure was defined as playing a board game using rules that differed in at least one way from the written instructions that came with the game. Changes to the game included one of the following: (a) going around the board backwards, (b) changing the value of points earned, (c) or doing the opposite of what was stated in to do in the instructions on cards chosen. 2.5.2. Exposure and control exercise The exposure and control exercises were identical to the exposure and cognitive defusion exercises, except that each participant repeated a control statement instead of silly phrases prior to the exposure session. The control statements consisted of the following: (a) ‘‘The sky is blue’’ (for Jacob and Cameron), (b) ‘‘The cow says moo’’ (for John), and (c) ‘‘The truck is red’’ (for Owen). 2.6. Results and discussion 2.6.1. Jacob The percentage of intervals with problem behavior displayed by Jacob during cognitive defusion exercises plus exposure compared to control exercises plus exposure is depicted in the upper panel of Fig. 2. Jacob displayed a lower percentage of problem behavior during the cognitive defusion exercises plus exposure condition. The mean percentage of problem behavior during the treatment condition was 10% (range, 0 to 27%). The mean percentage of problem behavior during the control condition was 21% (range, 7% to 37%). Percentage of appropriate play during the treatment condition compared to the control condition is displayed in the lower panel of Fig. 2. Jacob engaged in a higher percentage of appropriate play during the treatment condition. The mean percentage of appropriate play during the treatment condition was 90% (range, 74% to 100%). The mean percentage of appropriate play during the control condition was 79% (range 64%, to 90%). 2.6.2. John The percentage of problem behavior displayed by John during cognitive defusion exercises plus exposure compared to control exercises plus exposure is depicted in the upper panel of Fig. 3. John displayed a lower percentage of problem behavior during the cognitive defusion exercises plus exposure condition. The mean percentage of problem behavior during the treatment condition was 2% (range, 0% to 7%). The mean percentage of problem behavior during the control condition was 16% H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 26 50% Percentage of Intervals (Problem Behavior) 45% 40% Control 35% Defusion + Exposure 30% 25% 20% 15% 10% 5% 0% 1 2 3 4 5 6 7 8 9 Sessions 10 11 12 13 14 15 16 100% Percentage of Intervals (Appropriate Play) 90% 80% 70% 60% Defusion + Exposure 50% Control 40% 30% 20% Jacob 10% 0% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Sessions Fig. 2. Jacob’s percentage of intervals with problem behavior (top panel) and appropriate play (bottom panel). (range, 0% to 50%). Percentage of appropriate play during the treatment condition compared to the control condition is displayed in the lower panel of Fig. 3. John engaged in a higher percentage of appropriate play during the treatment condition. The mean percentage of appropriate play during the treatment condition was 98% (range, 94% to 100%). The mean percentage of appropriate play during the control condition was 84% (range, 50% to 100%). 2.6.3. Owen The percentage of problem behavior displayed by Owen during cognitive defusion exercises plus exposure compared to control exercises plus exposure is depicted in the upper panel of Fig. 4. Owen displayed a lower percentage of problem behavior during the cognitive defusion exercises plus exposure condition. The mean percentage of problem behavior during the treatment condition was 3% (range, 0 to 10%). The mean percentage of problem behavior during the control condition was 19% (range, 0% to 47%). The percentage of appropriate play during the treatment condition compared to the control condition is displayed in the lower panel of Fig. 4. Owen engaged in a higher percentage of appropriate play during the treatment condition. The mean percentage of appropriate play during the treatment condition was 98% (range, 90% to 100%). The mean percentage of appropriate play during the control condition was 82% (range, 54% to 100%). H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 Percentage of Intervals (Problem Behavior) 50% 27 Control 40% 30% 20% Defusion + Exposure 10% 0% 1 2 3 4 5 6 7 Sessions 8 9 10 11 12 13 100% Percentage of Intervals (Appropriate Play) 90% Defusion + Exposure 80% 70% 60% 50% 40% Control 30% 20% John 10% 0% 1 2 3 4 5 6 7 8 Sessions 9 10 11 12 13 Fig. 3. John’s percentage of intervals with problem behavior (top panel) and appropriate play (bottom panel). 2.6.4. Cameron The percentage of problem behavior displayed by Cameron during cognitive defusion exercises plus exposure compared to control exercises plus exposure is depicted in the upper panel of Fig. 5. Cameron displayed a higher percentage of problem behavior during the cognitive defusion exercises plus exposure condition. The mean percentage of problem behavior during the treatment condition was 13% (range, 0% to 30%). The mean percentage of problem behavior during the control condition was 4% (range, 0% to 20%). Percentage of appropriate play during the treatment condition compared to the control condition is displayed in the lower panel of Fig. 5. Cameron engaged in a lower percentage of appropriate play during the treatment condition. The mean percentage of appropriate play during the treatment condition was 88% (range, 70% to 100%). The mean percentage of appropriate play during the control condition was 96% (range, 80% to 100%). Three out of the four participants displayed lower percentages of problem behavior and an increase in the percentage of appropriate play with the cognitive defusion exercises compared to the control exercise. The results suggest that cognitive defusion exercises can enhance the effects of exposure therapy with some individuals. H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 28 50% Percentage of Intervals (Problem Behavior) 45% 40% 35% Control 30% 25% 20% 15% 10% Defusion + Exposure 5% 0% 1 2 3 4 5 6 7 Sessions 8 9 10 11 100% Percentage of Intervals (Appropriate Play) 90% Defusion + Exposure 80% 70% 60% Control 50% 40% 30% 20% Owen 10% 0% 1 2 3 4 5 6 7 Sessions 8 9 10 11 Fig. 4. Owen’s percentage of intervals with problem behavior (top panel) and appropriate play (bottom panel). 2.6.5. Statistical comparison The level of problem behavior was compared statistically with paired-associate t tests, in which the first value for the defusion plus exposure condition was entered in comparison to the first value in the exposure condition, and so on. If one condition had more data points than the other due to random assignment, the mean for the less frequent condition was used as the comparison data point. Overall, defusion lead to significantly lower levels of disruptive behavior in the four participants as a whole (t(29) = 3.16, p = .004, two tailed). When conducted participant by participant, there was no significant difference between the conditions for Cameron (t(6) = 1.36, p = .22, two tailed), but the effects were at least marginally significant in favor of defusion for all other participants: Jacob: t(8) = 3.91, p = .004, two tailed; John: t(7) = 2.88, p = .024, two tailed; Owen: t(5) = 2.05, p = .095, two tailed. It will be important going forward to identify the circumstances under which cognitive defusion exercises may result in increased problem behavior as was the case with Cameron. Although previous research (e.g., Masuda et al., 2009) suggests that exposure to defusion exercises for a duration as small as 3 s can be successful in observing treatment effects, children with ASD might be prone to focus on irrelevant features of the environment, and as such, successful treatment may require H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 29 50% Percentage of Intervals (Problem Behavior) 45% 40% 35% Defusion + Exposure 30% Control 25% 20% 15% 10% 5% 0% 1 2 3 4 5 6 Sessions 7 8 9 10 11 100% 90% Percentage of Intervals (Appropriate Play) 80% 70% Control 60% Defusion + Exposure 50% 40% 30% 20% 10% Cameron 0% 1 2 3 4 5 6 Sessions 7 8 9 10 11 Fig. 5. Cameron’s percentage of intervals with problem behavior (top panel) and appropriate play (bottom panel). multiple examples of stimuli to be present in avoided settings and prolonged treatment exposure. Unfortunately, Cameron’s intervention services were changed to a different location and it was not possible to continue to work with him to identify possible variables influencing the increase in problem behaviors. Further investigation will be needed to identify the length of treatment or changes in variables required to demonstrate positive results with individuals who do not initially respond to treatment. For example, while we selected verbal targets that seemed to make sense given the target behavior, more sophisticated functional analytic procedures may have been needed to ensure that such choices were on target. 2.6.6. Social validity The results of the social validity assessment are depicted in Table 1. Overall, the primary caregivers and behavioral therapists rated the treatment as useful, acceptable, and important with the most common responses being strongly agree (statements 1–6, 8, 10–12). Statement 7 (‘‘I believe the treatment is likely to result in permanent improvement’’) was administered to assess the raters’ belief about the long-term benefits of the treatment. Three of the raters said they agreed with the statement, two raters said they strongly agreed, and one of the raters said they were neutral. Question 9 assessed the raters’ belief regarding the child experiencing discomfort during the treatment (‘‘I believe the child experienced 30 H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 discomfort during the treatment’’). Three of the raters said they disagree with the statement, two said they agree with the statement, and one said they were neutral. It would be expected to have some of the raters agree with statement 9, as it is likely that the participants did experience some discomfort during the treatment procedure. It may be that in comparison to the level of discomfort observed in the natural setting, the level of discomfort appeared to be less during treatment, which led to the raters disagreeing with the statement. Statement 12 assessed the raters overall reaction to the treatment (‘‘Overall, I have a positive reaction to this treatment’’). Four raters said they strongly agree with the statement and two raters said they agree with the statement. The rating scale suggests that the primary caregivers and behavioral therapists believed the treatment was acceptable, useful, important, long-lasting, and did not result in the child experiencing discomfort. 3. General discussion Exposure therapy has been a longstanding treatment to extinguish conditioned emotional responses to stimuli. Repetitive and restrictive behaviors appear to be at least partially maintained by escape/avoidance of aversive psychological experiences; therefore, exposure therapy alone should be effective in reducing problem behavior. ERP has been demonstrated to be an effective treatment for OCD (Foa, Franklin, & Kozak, 1998), but additional treatment development is needed since there are often high rates of treatment refusal and drop-out. Exposure is thought to work in part by reducing the impact of repertoire narrowing stimuli, including inaccurate verbal rules. This is part of the basis for a cognitive theory of exposure, but recent research suggests that cognitive defusion is also an important process of change in exposure even in traditional cognitive behavioral methods of exposure (Wolitzky-Taylor, Arch, Rosenfield, & Craske, 2012). Cognitive defusion exercises appear to work by reducing the dominance of rule-governed behavior and increasing response variability. From an RFT perspective, participating in defusion exercises prior to exposure sessions should enhance the exposure effects by altering the function of the associated verbal phrases, which are thought to be repertoire narrowing and lead to avoidance. For example, John ran away upon seeing any type of statue. The phrase picked for defusion was, ‘‘I don’t like the statue’’ based on the idea that the statues may have participated in evaluative verbal networks (e.g., ‘‘I don’t like that,’’ ‘‘I hate that’’) which increased their aversive properties and capacity to foster avoidance and escape responses. Exposure to statues occurred in either condition, but reductions in avoidance were more rapid after saying, ‘‘I don’t like the statue,’’ in silly voices, as compared to saying, ‘‘The cow says moo,’’ in silly voices. This provides incremental support for the idea that the verbal functions of avoided events can be altered by deliberately changing the functional context of verbal events—in John’s case, from one of literal evaluation, to one of playful expression. The current study demonstrated that defusion exercises did appear to enhance the exposure effects of the participants, with the exception of Cameron. Word repetition in silly voices, per se, is not defusion as a process just as praise is not necessarily reinforcement. If word repetition did not reduce the literal quality of words, then other forms of defusion exercises may be better suited for the individual. Had there been more time, the logical next step with Cameron would have been to try other defusion methods, such as, Think the Opposite or Monsters on the Bus (for examples of defusion exercises see Hayes & Smith, 2005, pp. 83–85). Future research should consider the use of individually-based tests of verbal functions and the transformation of stimulus functions by various alternative defusion methods in a similar fashion as has been done with functional analysis on the consequential end. It may also be that Cameron was not truly exposed to the aversive situation. Exposure therapy alone has also failed to produce results in some cases, in particular with panic disorder, resulting in criticisms and rejections of classical conditioning theory (for a detailed review see Thorn, Chosak, Baker, & Barlow, 1999). Bouton, Barlow, and Mineka (2001) point out that the strength of a particular conditioned response can be controlled by contextual stimuli, which may be a particularly relevant point when working with children with ASD, who are more likely to focus on irrelevant features of the environment (Lovaas, Schreibman, Koegel, & Rehm, 1971; Schreibman, 1997). A limitation of the current experiments is that they rely on circumstantial evidence to hypothesize about the topography of the cognitions that may be related to repetitive behavior. Research thus far using cognitive defusion exercises have often relied on the participants to describe the verbal statements or thoughts that evoke avoidance behavior. That approach can be challenging with developmentally delayed populations, and the current experiments instead determined verbal targets based on the circumstances under which the target behaviors were most likely to occur and whatever overt statements had been made by the participants. These methods alone, may not have adequately targeted the verbal events that evoked avoidant behavior. In future research, the use of implicit assessment approaches (e.g., Scanlon, McEnteggart, Barnes-Holmes, & Barnes-Holmes, 2014) or other functional tests conducted prior to treatment in order to analyze the role of problematic verbal events controlling problem behaviors may provide ways around these difficulties. Another limitation to the current experiments is the inability to identify the mechanisms responsible for behavior change with certainty, since it is not possible to extract cognitive defusion exercises from exposure therapy when assessing the effects of cognitive defusion exercises. This difficulty comes from the necessity of exposing the participants to the novel stimuli/situation that occasions the target behaviors in order to assess the effects of treatment on target behaviors. There tended to be gradual improvement across the study. One possible conclusion that could be argued is that behavior would have improved on its own over time. Another, more likely possibility, is that extinction was partly responsible for the behavior change. H.J. Eilers, S.C. Hayes / Research in Autism Spectrum Disorders 19 (2015) 18–31 31 Repetitive and restrictive behaviors have been cited as core features of ASD. As such, identifying an effective treatment to reduce problem behaviors associated with repetitive and restrictive behaviors contributes to the current empiricallyvalidated treatments to reduce symptoms of ASD. The current experiments provide a treatment procedure that enhances the effects of exposure-based treatments by reducing more quickly the aggressive and disruptive behavior typically observed when children with ASD are exposed to novel stimuli/situations. Furthermore, the current experiments provide an alternative theory regarding the possible function of repetitive and restrictive behaviors. Many of the studies examining automatically reinforced behaviors have hypothesized positive reinforcement in the form of sensory stimulation or negative reinforcement in the form of pain attenuation (Leblanc, Patel, & Carr, 2000). Few studies have hypothesized the function to be negative reinforcement in the form of avoiding aversive private events such as difficult verbal events and their emotional impact (what has been called ‘‘experiential avoidance’’ in the ACT literature). The current experiments suggest that repetitive and restrictive behavior treatment procedures could benefit from borrowing from existing literature and treatment procedures used with anxiety-based disorders, such as OCD. Very few interventions have been used in ASD treatment to target rule-governed behavior and even fewer to target behaviors that may be maintained by avoidance of private events. A well-developed research base now exists in the area of language and cognition within the context of RFT that can be incorporated as a translational extension into existing applied behavioral interventions. The current study is a first step toward identifying procedures derived from the RFT literature that may be used to enhance treatment outcomes addressing rule-governed behavior and experiential avoidance in children with autism. We thank Larry Williams, Ramona Houmanfar, Barbara Kohlenberg, and Thomas Nickles for their helpful comments on an earlier version of the manuscript and the staff at Center for Autism Research, Evaluation, and Services (CARES) for their assistance with data collection. These studies are based on a dissertation submitted by the first author, under the supervision of the second author, to the Department of Psychology at University of Nevada for the Ph.D. degree. References Abramowitz, J. S., Taylor, S., & McKay, D. (2009). Obsessive–compulsive disorder. The Lancet, 374, 491–499. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. Bodfish, J. W. (2004). Treating the core features of autism: Are we there yet? Mental Retardation and Developmental Disabilities Research Review, 10, 318–326. Bouton, M. E., Barlow, D. H., & Mineka, S. (2001). A modern learning theory perspective on the etiology of panic disorder. Psychological Review, 108, 4–32. Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis (2nd ed.). Upper Saddle River, NJ: Prentice Hall. Farrugia, S., & Hudson, J. (2006). Anxiety in adolescents with Asperger syndrome: Negative thoughts, behavioral problems, and life interference. Focus on Autism and other Developmental Disabilities, 21, 22–25. Foa, F. B., Franklin, M. E., & Kozak, M. J. (1998). Psychosocial treatments for obsessive–compulsive disorder: Literature review. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.), Obsessive–Compulsive Disorder: Theory, research and treatment (pp. 258–276). New York, NY: Guilford. Hausman, N., Kahng, S., Farrell, E., & Mongeon, C. (2009). Idiosyncratic functions: Severe problem behavior maintained by access to ritualistic behaviors. Education and Treatment of Children, 32, 77–87. Hayes, S. C., Barnes-Holmes, D., & Roche, B. (2001). Relational Frame Theory: A Post-Skinnerian account of human language and cognition. New York: Plenum Press. Hayes, S. C., & Smith, S. (2005). Get out of your mind and into your life: The new Acceptance and Commitment Therapy. Oakland, CA: New Harbinger Publications. Hayes, S. C., Strosahl, K., & Wilson, K. G. (2012). Acceptance and Commitment Therapy: The process and practice of mindful change (2nd ed.). New York: Guilford Press. Kuhn, D. E., Hardesty, S. L., & Sweeney, N. M. (2009). Assessment and treatment of excessive straightening and destructive behavior in an adolescent diagnosed with autism. Journal of Applied Behavior Analysis, 42, 355–360. Leblanc, L. A., Patel, M. R., & Carr, J. E. (2000). Recent advances in the assessment of aberrant behavior maintained by automatic reinforcement. Journal of Behavior Therapy and Experimental Psychiatry, 31, 137–154. Lewis, M. H., & Bodfish, J. W. (1999). Repetitive behavior disorders in autism. Mental Retardation Developmental Disabilities Research Review, 4, 80–89. Lovaas, I. O., Schreibman, L., Koegel, R., & Rehm, R. (1971). Selective responding by autistic children to multiple sensory input. Journal of Abnormal Psychology, 77, 211–222. Masuda, A., Feinstein, A. B., Wendell, J. W., & Sheehan, S. T. (2010). Cognitive defusion versus thought distraction: A clinical rational, training, and experiential exercise in altering psychological impact of negative self-referential thoughts. Behavior Modification, 34, 520–538. Masuda, A., Hayes, S. C., Sackett, C. F., & Twohig, M. P. (2004). Cognitive defusion and self-relevant negative thoughts: Examining the impact of a ninety year old technique. Behavior Research and Therapy, 42, 477–485. Masuda, A., Hayes, S. C., Twohig, M. P., Drossel, C., Lillias, J., & Washio, Y. (2009). A parametric study of cognitive defusion and the believability and discomfort of negative self-relevant thoughts. Behavior Modification, 33, 250–262. Masuda, A., Twohig, M. P., Stormo, A. R., Feinstein, A. B., Chou, Y., & Wendell, J. W. (2010). The effects of cognitive defusion and thought distraction on emotional discomfort and believability of negative self-referential thoughts. Journal of Behavior Therapy and Experimental Psychiatry, 41, 11–17. Rosa-Alcázar, A., Sánchez-Meca, J., Gómez-Conesa, A., & Marı́n-Martı́nez, F. (2008). Psychological treatment of obsessive–compulsive disorder: A meta-analysis. Clinical Psychology Review, 28, 1310–1325. Scanlon, G., McEnterggart, C., Barnes-Holmes, Y., & Barnes-Holmes, D. (2014). Using the implicit relational assessment procedure (IRAP) to assess implicit gender bias and self-esteem in typically developing children and children with ADHD and dyslexia. Behavioral Development Bulletin, 19, 48–59. Schreibman, L. (1997). The study of stimulus control in autism. In D. M. Baer & E. M. Pinkson (Eds.), Environment and behavior (pp. 203–209). Boulder, CO: Westview Press. Tantam, D. (2000). Psychological disorder in adolescents and adults with Asperger Syndrome. Autism, 4, 47–62. Thorn, G. R., Chosak, A., Baker, S. L., & Barlow, D. H. (1999). Psychological theories of panic disorder. In D. J. Nutt, J. C. Ballenger, & J. P. Lepine (Eds.), Panic disorder: Clinical diagnosis, management, and mechanisms (pp. 93–108). London: Martin Dunitz. Titchener, E. B. (1916). A text-book of psychology. New York, NY: MacMillan. Turner, M. (1999). Annotation: Repetitive behavior in autism: A review of psychological research. Journal of Psychological Psychiatry, 40, 839–849. van Steensel, F. J. A., Bögels, S. M., & Perrin, S. (2011). Anxiety disorders in children with adolescents with autistic spectrum disorders: A meta-analysis. Clinical Child and family Psychological Review, 14, 302–317. Watson, C., Burley, M. C., & Purdon, C. (2010). Verbal repetition in the reappraisal of contamination-related thoughts. Behavioral and Cognitive Psychotherapy, 38, 337–353. Wolitzky-Taylor, K. B., Arch, J. J., Rosenfield, D., & Craske, M. G. (2012). Moderators and non-specific predictors of treatment outcome for anxiety disorders: A comparison of cognitive behavioral therapy to acceptance and commitment therapy. Journal of Consulting and Clinical Psychology, 80, 786–799. Zandt, F., Prior, M., & Kyrios, M. (2007). Repetitive behavior in children with high functioning autism and obsessive compulsive disorder. Journal of Autism and Developmental Disorders, 37, 251–259. All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
