Skip to main content
NCBI home page
Journal of Applied Behavior Analysis logoLink to Journal of Applied Behavior Analysis
. 2012 Fall;45(3):549–564. doi: 10.1901/jaba.2012.45-549

THE EFFECTS OF MATCHED STIMULATION AND RESPONSE INTERRUPTION AND REDIRECTION ON VOCAL STEREOTYPY

Jessica J Love 1, Caio F Miguel 1, Jonathan K Fernand 1, Jillian K LaBrie 1
PMCID: PMC3469292  PMID: 23060668

Abstract

Stereotypy has been classified as repetitive behavior that does not serve any apparent function. Two procedures that have been found to reduce rates of vocal stereotypy effectively are response interruption and redirection (RIRD) and noncontingent access to matched stimulation (MS). The purpose of the current study was to evaluate the effects of RIRD alone, MS alone, and MS combined with RIRD. One participant's results suggested similar suppressive effects on vocal stereotypy across treatment conditions. For the second participant, a slightly greater suppression of stereotypy was associated with MS + RIRD. In addition, both participants emitted a greater frequency of appropriate vocalizations in conditions with RIRD. Data suggest that the addition of MS might facilitate the implementation of RIRD in applied settings.

Key words: autism, matched stimulation, response interruption and redirection, vocal stereotypy

The term stereotypy is often used to classify either motor or vocal repetitive behavior that does not appear to serve an adaptive function (Lovaas, Newsom, & Hickman, 1987). Although stereotypy is not displayed exclusively by individuals with disabilities, an increased rate can be observed in this population, especially in those diagnosed with autism (American Psychiatric Association, 2005; Lewis & Bodfish, 1998; Repp & Barton, 1980). Stereotypy is typically targeted for reduction given that it may interfere with learning, compete with more functional responses, and be socially stigmatizing (e.g., Durand & Carr, 1987; Koegel & Covert, 1972; Matson, Kiely, & Bamburg, 1997; Repp & Barton, 1980). Vocal stereotypy specifically may compete with more adaptive forms of communication (e.g., mands) and also create disruptions for other people in the social environment (e.g., Athens, Vollmer, Sloman, & St. Peter Pipkin, 2008).

Lovaas et al. (1987) suggested that stereotypical responding functions to provide sensory input to an individual. That is, it produces automatic reinforcement. Stereotypic behavior maintained by automatic reinforcement may be more difficult to treat because one does not have access to the exact source of reinforcement (Vollmer, 1994). Despite these challenges, interventions have emerged to treat automatically reinforced stereotypy. Two of the recently evaluated treatments are response interruption and redirection (RIRD; e.g., Ahearn, Clark, MacDonald, & Chung, 2007; Ahrens, Lerman, Kodak, Worsdell, & Keegan, 2011; Duffy-Cassella, Sidener, Sidener, & Progar, 2011; Liu-Gitz & Banda, 2010; Miguel, Clark, Tereshko, & Ahearn, 2009) and noncontingent access to matched stimulation (MS; e.g., Lanovaz, Fletcher, & Rapp, 2009; Rapp, 2007; Taylor, Hoch, & Weissman, 2005).

Ahearn et al. (2007) evaluated RIRD with four participants who displayed automatically reinforced vocal stereotypy. During RIRD, the experimenter interrupted the participants' vocal stereotypy and then redirected them to emit appropriate vocalizations such as answering social questions. Results showed a decrease in vocal stereotypy for all four participants and an increase in the frequency of appropriate vocalizations for three of four participants. In a more applied extension, Liu-Gitz and Banda (2010) evaluated teacher-implemented RIRD in a natural classroom setting. Results showed significant reductions in vocal stereotypy and provided further evidence for the efficacy of RIRD in school settings. Recently, Ahrens et al. (2011) sought to isolate the mechanism that is responsible for the effects of RIRD by implementing a treatment integrity fading procedure modeled after Lerman and Iwata (1996) and Smith, Russo, and Le (1999). Results indicated that reductions in vocal stereotypy followed a pattern of responding indicative of positive punishment.

Because vocal stereotypy may occur quite frequently, a high number of RIRD implementations may be required to achieve considerable reductions in behavior, which may prevent clinicians from adopting the procedure (Miguel et al., 2009). This limitation may be circumvented by potentially decreasing the motivating operation (MO) for engaging in stereotypy through the presentation of stimuli that produce the same hypothesized sensory consequence as the problem behavior (Piazza, Adelinis, Hanley, Goh, & Delia, 2000; Piazza et al., 1998). The addition of these stimuli (matched stimulation) to a treatment with RIRD may be a more efficient approach than RIRD alone, because decreasing the MO for the products of stereotypy may result in fewer implementations of RIRD. Piazza et al. (1998, 2000) investigated the effects of MS on pica, saliva manipulation, hand mouthing, jumping, and climbing, which were all maintained by automatic reinforcement. Participants received continuous access to items that produced either the same (matched) or different (unmatched) forms of stimulation. Results indicated a greater reduction in the target behaviors for most participants when they were given continuous and noncontingent access to items that were hypothesized to match the target behavior's sensory consequences.

Fisher, Lindauer, Alterson, and Thompson (1998) also evaluated the effects of MS and unmatched stimulation on stereotypic object breaking plus tapping. Results indicated that matched toys alone decreased stereotypy for one of the two participants. Following the addition of response blocking to MS for one participant, rates of stereotypy further decreased and appropriate manipulation of the matched toys increased. Thus, previous literature on MS and noncontingent reinforcement highlights the potential benefits of supplementing a response-blocking procedure with MS.

In light of the aforementioned findings, the present investigation sought to evaluate whether the use of MS would increase the effectiveness of RIRD. By lowering the MO for the products of stereotypy, the addition of MS could increase the feasibility of implementing RIRD in school or community settings if the inclusion of MS decreases the number of times RIRD must be implemented. In addition, the relative efficacy of MS and RIRD treatment components were evaluated both alone and as a treatment package.

METHOD

Participants, Setting, and Materials

Participants were two children who had been diagnosed with an autism spectrum disorder by independent psychologists. Both participants communicated vocally, could sit for a minimum of 5 min in the absence of aggression (e.g., hitting, kicking, punching) or self-injury (e.g., scratching, hitting, biting him- or herself), and engaged in vocal stereotypy that was at least partially maintained by automatic reinforcement, as determined by a functional analysis (Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1994). Parents referred their children to participate in the study because their children's vocal stereotypy was considered to be intrusive in school or public settings.

Ivan was an 8-year-old boy who vocally communicated with others. He requested highly preferred items using “I want —.” Ivan could tact colors and numbers. His fine motor skills were consistent with those of his typical peers. We conducted sessions in his bedroom (2 m by 3 m), which contained a bed, a small table and two chairs, various leisure items, and mirrored closet doors. Ivan did not take any medications throughout the duration of the study.

Troy was a 9-year old boy who requested items using full sentences (e.g., “I want Elmo, please”). He could also tact colors and identify numbers up to 30. Troy's parents reported that his general academic abilities were comparable to his peers. Troy did not take any medication for the duration of the study. We conducted sessions in a treatment room (2 m by 3 m) equipped with a table and two chairs, a bulletin board, a video camera, and toys.

We conducted approximately three to four sessions per day, 2 to 3 days per week with both participants.

Response Measurement

Observers collected data on four dependent variables for each participant: vocal stereotypy, appropriate vocalizations, frequency of implementation of RIRD, and session length. For Ivan, vocal stereotypy was defined as any instance of noncontextual phrases or repetitions of noncontextual words, phrases, or sounds. It included noncontextual repetitions, defined as at least two occurrences of phonemes (two or more emissions of the same phoneme within a 5-s interval; e.g., “mm mm” or “na na na”) or phrases (e.g., “J is for jaguar”), delayed echolalia (e.g., “We go to the doctor first, then Blockbuster”), lip popping, and repetitious sounds with a closed mouth (e.g., “bbbbbb” sound with mouth closed and lips vibrating). Sound effects in the absence of appropriate toys (e.g., “shoo” or “vroom”), rhythmic breathing patterns (e.g., “huh huh huh” or “heh heh heh”), and noncontextual repetitive blowing of air (e.g., blowing in the absence of candles) were also included. Troy's vocal stereotypy was defined as any instance of noncontextual or nonfunctional speech, including rhythmic or patterned breathing (e.g., “huh huh huh,” “heh heh heh,” or breathing inward through pursed lips) and single phonemes (e.g., “mmm”). Repetitions of words, phonemes (e.g., “kkk,” “tuh tuh tuh”), or noncontextual high-pitched squeals or squeaks, and reciting of movie or videogame scripts were also included.

Ivan's and Troy's appropriate vocalizations included mands for items, attention, or breaks using the “I want —” frame and tacts that included vocalizing the name of an item in its presence (e.g., “tiger” in the presence of a tiger figurine). For both participants, we did not score repetitive mands (e.g., “I want break, I want break”) or tacts (e.g., “tiger, tiger”) as vocal stereotypy unless the phrase was noncontextual (e.g., “crocodile, crocodile” when no crocodile was present). Furthermore, the repetitive tact was considered appropriate and contextual if the participant exhibited shared interaction with the experimenter. For example, if the participant looked at the tiger and looked at the experimenter while saying “tiger” during the second repetition of the word tiger, this behavior was scored as appropriate. We scored the above-mentioned examples of vocalizations as appropriate because it was assumed that joint attention with the experimenter during the vocalization might have served a social function. Moreover, if a participant repeated the same vocalization within 5 s (e.g., “I want break, I want break”), we scored only one instance of an appropriate vocalization. RIRD implementation was defined as the experimenter saying the participant's name, the participant complying with three vocal imitation demands or social questions, and the experimenter delivering praise (e.g., “Nice job talking”).

We measured vocal stereotypy using duration recording. To calculate the percentage of stereotypy that occurred in each session, we divided the total number of seconds the participant engaged in vocal stereotypy by the total number of seconds in the session (i.e., 300 s) and converted this number to a percentage. Observers did not record data on vocal stereotypy or appropriate vocalizations when the experimenter implemented RIRD. Furthermore, the experimenter stopped the session clock each time he or she implemented the procedure to ensure that time spent implementing RIRD was taken out of the 5-min total session time.

Data on appropriate vocalizations and implementation of RIRD were collected using frequency recording and expressed as cumulative frequency across sessions. At the conclusion of the study, we calculated and compared the average number of implementations of RIRD per session during the conditions of MS + RIRD and RIRD alone for each participant. In addition, averages for each condition were compared across participants.

Interobserver Agreement

A second trained observer scored data during at least 33% of all conditions by watching the video footage. We calculated agreement on occurrences of stereotypy using a time-window analysis method (Mudford, Martin, Hui, & Taylor, 2009), which involved a second-by-second comparison across the two observers for both occurrence and nonoccurrence of stereotypy. We scored an agreement if both observers agreed that stereotypy occurred or did not occur for a 1-s interval. We divided the number of agreements by the number of agreements plus disagreements and converted this number to a percentage. We also calculated interobserver agreement for appropriate vocalizations using the total agreement method by comparing the cumulative frequencies recorded by observers. The smaller total from one observer was divided by the larger total from the second observer and this number was converted to a percentage. Mean agreement for vocal stereotypy and appropriate vocalizations was 96% (range, 89% to 100%) and 95% (range, 84% to 100%), respectively. Mean agreement for appropriate vocalizations was 100%.

Treatment Integrity

We collected treatment integrity on implementation of RIRD and MS during a portion of sessions in each condition of the treatment evaluation. Correct implementation of RIRD was scored if the experimenter delivered RIRD within 2 s of vocal stereotypy and ensured that the participant complied with three consecutive demands in the absence of stereotypy. We calculated integrity during RIRD by dividing the total number of correct implementations of RIRD by the total number of opportunities to implement RIRD in the session. The quotient was converted to a percentage of overall treatment integrity for the implementation of RIRD.

Treatment integrity was also calculated during conditions that included the delivery of MS using 5-s partial-interval recording. An interval was scored as containing MS if the toys were making sounds at any time during the 5-s interval. We divided the number of intervals containing MS by the total number of intervals for each session (i.e., 60) and converted the result to a percentage to obtain an overall percentage of integrity. Session percentages were then added and averaged to yield a percentage of overall integrity of MS for each participant.

Across both participants, we evaluated treatment integrity for implementation of RIRD during 41% of sessions and for implementation of MS during 36% of sessions. Treatment integrity averaged 96% (range, 92% to 100%) for RIRD and 99% (range, 93% to 100%) for MS.

Preexperimental Procedures

We conducted a stimulus preference assessment, a functional analysis, RIRD probes, and a matched stimuli assessment prior to the experimental evaluation.

Preference Assessment

The experimenter conducted three sessions of a multiple-stimulus without replacement (MSWO) preference assessment for both matched and unmatched toys based on procedures described by DeLeon and Iwata (1996). Toys classified as “matched” were those that made sound, and toys classified as “unmatched” were toys that did not make sound. The presence or absence of these toys was planned for each condition, because vocal stereotypy was hypothesized to be maintained by auditory stimulation. Parents completed a modified Reinforcer Assessment for Individuals with Severe Disabilities (RAISD; Fisher, Piazza, Bowman, & Amari, 1996) to identify some of the items included in the preference assessments. Along with the items identified via the RAISD, the experimenter conducted an MSWO to identify additional items. Two preference assessments were conducted, one for 10 toys that produced sound and one for 10 toys that did not produce sound. Thereafter, one MSWO per day was conducted with the seven top-ranked items. Depending on the type of sessions conducted during a day, sometimes the experimenter conducted two preference assessments (one for matched and one for unmatched). The toys that produced sound were later tested (MS assessment) to see whether they would compete with the auditory product (i.e., automatic reinforcer) of vocal stereotypy.

Items identified for inclusion in functional analysis sessions were highly and moderately preferred toys that did not produce sound. The use of these toys served to control for potential suppressive effects that sound may have exerted on stereotypy levels (e.g., Rapp, 2006, 2007). Anecdotal observations indicated that the highly preferred items used during sessions of the functional analysis did not suppress stereotypy. The experimenter randomly selected four items identified as moderately and highly preferred from the daily MSWO for baseline and treatment sessions to prevent potential satiation effects.

Functional Analysis

We conducted a functional analysis of vocal stereotypy according to procedures described by Roscoe, Carreau, MacDonald, and Pence (2008) and Athens et al. (2008). We rotated among seven highly and moderately preferred toys from the initial MSWO during sessions, with three toys per functional analysis session (that required toys). Toys were rotated to prevent satiation. The toys included in the previous session were not made available in subsequent sessions unless the participant requested an item or indicated sustained preference (e.g., continually looking at the item or reaching for it), in which case he received access to it during the next session.

We conducted three conditions in the functional analysis in a fixed order: no consequence, attention, and demand. We conducted twice as many no-consequence sessions as demand and attention sessions, because we hypothesized that stereotypy was maintained by automatic reinforcement. An extended no-consequence condition was conducted for Ivan only, given that his performance was low and undifferentiated during the last few sessions of the multielement functional analysis. Sessions lasted 5 min.

No consequence

Participants had continuous access to two toys identified as moderately preferred from the initial MSWO preference assessment. We included moderately preferred toys so that they would not compete with stereotypy (Ahearn, Clark, DeBar, & Florentino, 2005). The experimenter withheld social interactions by sitting in a chair and looking away from the participant. The person filming the session was across from the participant and also did not interact with him. No programmed consequences were provided for appropriate vocalizations.

Attention

The participant and experimenter sat in a room in which two moderately preferred toys (that did not produce sound) were available on a table. Contingent on vocal stereotypy, the experimenter responded by lightly touching the participant on the shoulder and saying, “Don't do that; it's loud” or “Please be quiet.”

Demand

The experimenter rotated the presentation of four tasks similar to those presented during the participant's school or day-treatment program. We selected unmastered (e.g., less than 80% correct) and nonvocal tasks based on caregiver report. Demands for Ivan included hanging a shirt on a coat hanger, tying a shoe, folding a shirt, and buttoning a shirt. Demands for Troy included solving three-digit multiplication problems, tying a shoe, catching a ball, and putting pennies into a box with one hand. The experimenter used a least-to-most prompting hierarchy for all task presentations. Contingent on stereotypy, the experimenter said, “Okay, you don't have to” and removed the task for 15 s. No programmed consequences were provided for appropriate vocalizations.

MS Assessment

The purpose of this assessment was to demonstrate that the items used in conditions containing MS would suppress vocal stereotypy. We conducted the assessment using a multielement design and rapidly alternated among three conditions: baseline, toys with sound, and toys without sound. We alternated each baseline session with one of the toy conditions to avoid carryover effects. During Phase 1, we alternated between baseline sessions and toys with sound (i.e., matched stimulation). During Phase 2, we alternated between baseline and toys without sound. Phase 1 was then reinstated to replicate the effects of matched toys. Each session lasted 5 min.

Baseline

The purpose of this condition was to evaluate levels of vocal stereotypy in the absence of toys and social interaction. The experimenter sat in the room with the participant but did not interact with him in any way. No leisure items or materials were present. Although sessions began with the participant seated in a chair, he was free to move about the room, and the experimenter did not redirect him to sit.

Toys with sound

The purpose of this condition was to measure levels of vocal stereotypy in the presence of toys that produced sound, because we wanted to determine whether the auditory stimulation matched or effectively substituted for the sensory consequences typically produced by engaging in vocal stereotypy (Rapp, 2007). Sessions included two of four preferred items (identified in an MSWO). Thus, we evaluated two sets of toys. During the session, the participant sat at a table or on the floor and had access to a set of toys. Following 2 s without toy activation, the experimenter manipulated the item to produce auditory stimulation. This ensured consistent delivery of MS throughout the condition. After demonstrating that the presence of both sets of toys reduced stereotypy, the experimenter rotated the two sets during treatment conditions that included MS.

Toys without sound

The purpose of this condition was to evaluate vocal stereotypy levels in the presence of toys that did not produce sound. We used the same toys as those in the toys-with-sound condition but removed the batteries (Taylor et al., 2005). This condition sought to isolate the effects of the sound produced by the toys on vocal stereotypy.

RIRD Probes

Prior to treatment, an assessment probe was conducted to identify appropriate vocal demands for use during conditions containing RIRD. Prior to the study, we conducted probes of tasks that required vocalizations across two different experimenters and two different settings for a minimum criterion of 89% accuracy. The results indicated that both participants were most consistent and accurate with vocal imitation of two-syllable words.

From a master list of 15 words, we created three lists that contained 10 words and randomly rotated these lists across all sessions (lists available from the second author). We used the same word lists for both participants because of similar performances during the RIRD probes. The experimenter presented all words from the lists created during the probe session. The purpose of these lists was to ensure demand consistency across all experimenters. The experimenter randomly determined the order of words from each list by assigning numbers to each item and then drawing numbers from a hat. The list was placed out of the participant's view. A different list was used each day. Approximately three to four sessions were conducted per day.

Procedure

During all conditions, the participant had access to two toys (matched or unmatched) and had the opportunity to interact with the experimenter. The experimenter smiled and visually attended to the participant while he played, but did not say anything unless the participant emitted an appropriate vocalization. For example, if the participant said, “Dinosaur!” while playing with a dinosaur magnet, the experimenter responded by saying, “You're right, that is a dinosaur! Nice talking!”

Baseline and MS-alone conditions lasted exactly 5 min. RIRD and MS + RIRD conditions were longer than 5 min depending on the frequency of implementation of RIRD. There was no session cap for conditions with RIRD.

If the participant left his chair at any point during a session in any condition, the experimenter provided a gestural prompt by pointing towards or tapping the seat of the chair. If the participant complied with sitting in the chair, the experimenter delivered praise. If he did not sit within 5 s of the prompt, the experimenter provided a vocal prompt, “[Name], sit down please.” Compliance resulted in praise. If the participant did not comply with the vocal prompt, the experimenter physically guided the participant back to his chair by touching his back and walking with him back to the chair.

The two toys included in baseline and RIRD-alone sessions were identified as moderately preferred from the preference assessment and did not produce sound. We selected moderately preferred toys because previous research has shown that highly preferred items (although unmatched in stimulation) could occasion reductions in vocal stereotypy (Ahearn et al., 2005). The participant had the opportunity to manipulate only one toy at a time. He could gain access to the other toy by requesting it. All conditions, with the exception of baseline, included prompts to engage in a mand if a participant was looking at a toy (indicating interest) but did not ask for it.

To facilitate discrimination between MS + RIRD and RIRD-alone conditions, the experimenter wore a blue shirt during sessions in which matched stimulation was available (paired with the MS + RIRD and MS-alone condition) and a black shirt during sessions in which MS was not available (paired with the RIRD-alone condition). The experimenter also paired the instruction “We're going to play with toys that make noise” with the blue shirt. With the black shirt, the experimenter paired the instruction, “We're going to play with quiet toys.” During baseline, the experimenter wore any color shirt other than blue or black and did not provide any instructions.

Baseline

The participant had access to two unmatched toys in a room with the experimenter. No programmed consequences were provided contingent on vocal stereotypy. We conducted at least three sessions or continued conducting sessions until a steady state of vocal stereotypy was observed. If a steady state was not achieved within 10 sessions, then the next condition was introduced. This was also the criterion for introduction of all subsequent conditions.

MS + RIRD

The procedures included those described by Rapp (2007) for MS and Ahearn et al. (2007) for RIRD. The participant had continuous access to one of two toys that produced auditory stimulation. The experimenter rotated the two sets of toys when conducting more than one session per day. If the participant stopped engaging with the toy for longer than 2 s, the experimenter manipulated the toy to produce sound. The experimenter activated the toy approximately five times per session; however, in some sessions the toy was never activated.

Contingent on vocal stereotypy, the experimenter removed the toy, interrupted the participant's stereotypy, and redirected the participant by saying the boy's name in a neutral tone of voice before issuing a demand to engage in a vocal response (as described by Ahearn et al., 2007). The removal of the toys during RIRD functioned both to increase the likelihood of participants attending to demands during RIRD and to increase the MO for the emission of the request. If the toy produced sound at the time of RIRD implementation, the experimenter did not silence the toy. A secondary observer stopped the session clock during implementation of RIRD.

The vocal demands during RIRD were a series of three vocal imitations of words (e.g., “Troy say ‘hopscotch,' say ‘binder,' say ‘sofa'”) from a previously generated list of words. If the child engaged in stereotypy between appropriate responses, the experimenter continued to present instructions until he emitted three consecutive appropriate vocalizations without vocal stereotypy. If he emitted a mand during RIRD, the experimenter acknowledged the response and replied with, “in a minute, we can have —,” and then continued to deliver instructions. After the participant complied with three consecutive instructions without engaging in vocal stereotypy, the experimenter delivered verbal praise (e.g., “Nice job talking”) and started the session clock. In addition, the experimenter returned the leisure item to the participant contingent on a mand. We required the participant to mand to increase the likelihood that he contacted reinforcement for appropriate vocalizations (Ahearn et al., 2007). However, the experimenter activated a toy to produce auditory stimulation (MS) if the participant did not mand for an item or activate it within 2 s after completion of RIRD.

If the participant engaged in vocal stereotypy when the experimenter returned the item, the experimenter immediately removed the item and reimplemented RIRD. No consequences were provided for behaviors other than vocal stereotypy and appropriate vocalizations (e.g., throwing items, hitting the table). Sessions ended when 5 min elapsed (timed by the session clock).

RIRD alone

The procedures were similar to those described in the MS + RIRD condition, although the two moderately preferred leisure items included in the sessions did not produce any sound. The purpose of this condition was to evaluate the effects of RIRD alone on vocal stereotypy when the MO for stereotypy was assumed to be present. We assumed the MO was present, because the toys did not make sounds. Therefore, sounds produced by toys did not have the opportunity to function as an abolishing operation for the auditory products of stereotypy.

MS alone

The purpose of this condition was to evaluate levels of stereotypy in the presence of toys that produced MS (auditory stimulation). This condition was similar to the MS + RIRD condition; however, RIRD was not implemented. That is, the participants had access to one of two toys that produced sound throughout the session, and the experimenter activated a toy if 2 s elapsed without the production of MS. No programmed consequences were delivered contingent on stereotypy.

Experimental Design

We compared the effects of RIRD alone, MS alone, and MS + RIRD on vocal stereotypy and appropriate vocalizations in a multitreatment reversal design (Holcomb & Wolery, 1994). For both participants, we introduced MS + RIRD after the initial baseline. We counterbalanced the introduction of MS alone and RIRD alone across participants. The experimenter completed the evaluation with the reinstatement of the treatment associated with the greatest suppression of vocal stereotypy.

Social Validity

We obtained two seperate measures of social validity from both participants' parents at the conclusion of the study. One measure evaluated parents' opinion of the RIRD procedure, and the other measure evaluated validity of the MS procedure. The experimenter provided each measure after the parent viewed a brief video clip of an RIRD or MS treatment session. Scores for each question ranged from 1 (strongly disagree) to 5 (strongly agree). A score of 3 was neutral.

RESULTS

The results of the functional analyses for Ivan and Troy are depicted in Figure 1. For both participants, vocal stereotypy occurred in the absence of social consequences. Ivan's vocal stereotypy persisted under extended exposure to the no-consequence condition, and Troy's vocal stereotypy was highest in the no-consequence condition during the second half of the functional analysis. Results from both participants' functional analyses suggested that vocal stereotypy was maintained by automatic reinforcement. Figure 2 depicts the percentage of vocal stereotypy during the MS assessment for both Ivan and Troy. Results showed reduced stereotypy when toys with batteries were present.

Figure 1.

Figure 1

Open in a new tab

The percentage of vocal stereotypy during Ivan's and Troy's functional analysis.

Figure 2.

Figure 2

Open in a new tab

The percentage of vocal stereotypy during the matched stimuli assessment for Ivan and Troy for Toy Sets 1 (top) and 2 (bottom).

Figure 3 depicts the percentage of Ivan's vocal stereotypy during baseline and treatment conditions. Vocal stereotypy occurred at moderate levels during the initial baseline (M = 37%), and Ivan emitted appropriate vocalizations infrequently (M = 0.7). During the MS + RIRD condition, vocal stereotypy rapidly decreased to low levels (M = 2.1%, M = 5.8%) and appropriate vocalizations increased but were highly variable (M = 2.6, M = 7) during both exposures to this condition. Vocal stereotypy was maintained at lower levels during the MS-alone condition (M = 5.2%, M = 11.8%), although appropriate vocalizations decreased to baseline levels (M = 0.8, M = 0.5) during both phases of this condition. RIRD alone was associated with increased levels of appropriate vocalizations (M = 4.6, M = 6, M = 3.9) and low levels of vocal stereotypy (M = 5%, M = 3%, M = 4.3%) during the three exposures to this condition. Ivan's final treatment phase was RIRD alone because this treatment was associated with moderate rates of appropriate vocalizations and low rates of stereotypy, comparable to the MS + RIRD condition. Also, Ivan did not have regular access to toys producing sound in other environments (e.g., school), thereby decreasing the likelihood of parents or teachers implementing MS + RIRD.

Figure 3.

Figure 3

Open in a new tab

The percentage of vocal stereotypy and the frequency of appropriate vocalizations during treatment for Ivan. BL = baseline; MS = matched stimulation; RIRD = response interruption and redirection; AV = appropriate vocalization.

Figure 4 depicts the percentage of vocal stereotypy for Troy during baseline and treatment conditions. Vocal stereotypy occurred at moderate levels during the initial baseline (M = 16%), and he infrequently emitted appropriate vocalizations (M = 1.3). During the MS + RIRD condition, stereotypy was low (M = 1%, M = 0.7%, M = 1.5%) and appropriate vocalizations increased (M = 3.8, M = 3.3, M = 6.75) during the three phases of this condition. Stereotypy occurred at a slightly higher level during RIRD alone (M = 6.7%) compared to MS alone (M = 2.4%). A higher frequency of appropriate vocalizations was associated with RIRD alone (M = 16.6) compared to MS alone (M = 4.4). Troy's final treatment phase was MS + RIRD because this condition was associated with the lowest frequency of implementations of RIRD (see further description below) and a moderate frequency of appropriate vocalizations.

Figure 4.

Figure 4

Open in a new tab

The percentage of vocal stereotypy and frequency of appropriate vocalizations during treatment for Troy. BL = baseline; MS = matched stimulation; RIRD = response interruption and redirection; AV = appropriate vocalization.

Figure 5 (top) displays the average number of occasions in which the experimenter implemented RIRD per session for each participant. Ivan's frequency of RIRD implementations was relatively equal across conditions. The MS + RIRD condition averaged 4.5 implementations, and the RIRD-alone condition was slightly higher with an average of 5.8 implementations. We observed a larger discrepancy among treatments for Troy. The MS + RIRD condition was associated with a considerably lower frequency of implementations, averaging 2.3 per session. In comparison, the RIRD-alone condition averaged 15 implementations per session.

Figure 5.

Figure 5

Open in a new tab

The average frequency of response interruption and redirection (RIRD) implementations (top) and the average session length in seconds (bottom) per participant during RIRD alone and matched stimulation (MS) + RIRD conditions. The area above the horizontal line indicates the average time required to implement the RIRD procedure.

Figure 5 (bottom) displays the average session length for the MS + RIRD and RIRD-alone conditions for each participant. The horizontal line at the 300-s mark indicates the programmed amount of time for the session. The portion of the bar above the line indicates the average amount of extra session time in each condition. Ivan's treatment conditions had similar session lengths. Overall, both participants' session lengths did not display any decreasing trends throughout the experiment.

Table 1 displays the questions and average ratings across both parents on the social validity questionnaire. Overall, parents rated both the RIRD and MS procedures favorably. Parents were also asked to circle the name of the preferred procedure or none at all. If they preferred both procedures equally, they could circle both procedures. Both parents indicated a preference for MS + RIRD. One parent indicated that she would like to use the combined procedure at home and school.

Table 1.

Social Validity Questions and Average Parent Ratings for RIRD and Matched Stimulation

graphic file with name jaba-45-03-08-t01.jpg

Question
Average rating
RIRD
MS
 1. I liked the vocal RIRD + MS procedure. 4.5 4.5
 2. If trained, I could easily implement the vocal RIRD + MS procedure in my home. 4.5 4.5
 3. I am satisfied with my child's response to the vocal RIRD + MS procedure. 4 4
 4. Vocal RIRD + MS reduced my child's stereotypy during the session. 4 4.5
 5. I thought the vocal RIRD + MS procedure was intrusive. 1.5 1
 6. Vocal RIRD + MS is a socially appropriate intervention method for my child. 4 4.5
 7. I think vocal RIRD + MS improved my child's rate of appropriate vocalizations during the session. 4.5 4
 8. My child will benefit in the long run from the procedures implemented in this study. 4.5 4.5
 9. Vocal RIRD + MS produced negative emotional reactions in my child. 1 1
10. I feel vocal RIRD + MS produced little change in my child's stereotypy rates during session. 3 3
Open in a new tab

 Note. Items were rated on a Likert-type scale of 1 (strongly disagree) to 5 (strongly agree). Items 5, 9, and 10 were reverse coded.

DISCUSSION

Our results indicated that RIRD decreased stereotypy and increased appropriate vocalizations for both participants. These findings are consistent with previous studies on RIRD (e.g., Ahearn et al., 2007; Liu-Gitz & Banda, 2010; Miguel et al., 2009). Both participants also emitted a higher frequency of appropriate vocalizations in conditions that contained RIRD. In addition, we observed decreased levels of vocal stereotypy for both participants during conditions that contained MS. The observed reduction in vocal stereotypy in the presence of MS may have been a result of the toys' auditory stimulation functioning as an abolishing operation for the product of vocal stereotypy. These findings support results obtained by Rapp (2006) that stereotypy is sensitive to the manipulation of MOs.

RIRD alone and MS alone produced similar outcomes for both participants. For Troy, MS + RIRD produced lower levels of stereotypy with a shorter session time and considerably fewer implementations of RIRD than RIRD alone. For Ivan, the treatment conditions were associated with similar levels of stereotypy, session lengths, and number of RIRD implementations. The comparable suppressive effects of RIRD alone and MS + RIRD on Ivan's stereotypy may be attributed to the inclusion of RIRD. Moreover, he engaged in fewer appropriate vocalizations during MS alone than during conditions that contained RIRD. The increase in vocalizations during RIRD may be related to the toy-removal component embedded in our RIRD procedure. This component may have functioned as a conditioned transitive MO (Michael, 2003), establishing the value of the removed toy as a reinforcer and increasing the probability of appropriate requests. In addition, toy removal may have functioned as punishment (i.e., response cost) for inappropriate vocalizations (e.g., Falcomata, Roane, Hovanetz, & Kettering, 2004). In contrast, Troy's results suggested that MS + RIRD was associated with the most positive treatment outcome. These effects are consistent with O'Reilly et al. (2006, 2007, 2008), who found that the addition of noncontingent reinforcement increased the effectiveness of procedures designed to suppress problem behavior. Troy's results also are consistent with those of Fisher et al. (1998), who found that MS plus response blocking was more effective for reducing one participant's stereotypy than MS alone.

This study adds to the literature evaluating MS in conjunction with response blocking. Specifically, the reduction in vocal stereotypy during MS + RIRD and MS conditions was notable, because stereotypy could still occur during the manipulation of matched items. Participants in Fisher et al. (1998) had access to matched toys, but manipulating them was incompatible with their topographies of stereotypy. One limitation of RIRD is the effort that may be necessary to implement it consistently in applied settings. Troy's results indicate that the addition of MS to RIRD may lower the number of implementations needed to suppress stereotypy, which may increase the likelihood that caregivers will implement RIRD consistently. Overall, the addition of MS may address a possible limitation of RIRD and make it more user friendly. However, MS requires intervention agents to identify stimuli that compete with stereotypy, and providing continuous access to toys may not be possible during all classroom activities. Thus, additional research is needed to evaluate the feasibility of implementing MS + RIRD in more natural settings.

Several additional limitations warrant consideration. First, because RIRD could be conceptualized as a treatment package in this study, it is difficult to discern which component of RIRD (e.g., vocal demands, response cost) was responsible for the change in behavior. Results of more recent RIRD studies (e.g., Ahrens et al., 2011; Duffy-Casella et al., 2011) indicate that RIRD functions as positive punishment. Toy removal contingent on stereotypy may function as negative punishment. In addition, Ivan's increased frequency of appropriate vocalizations under conditions with RIRD seemed to be correlated with the number of times the response-cost procedure was implemented. Specifically, the more times response cost was implemented, the more times he manded to get the item back. Response cost was not included in the MS-alone condition. The omission of response cost during this condition may have provided fewer opportunities for Ivan to emit mands, which may explain the observed decrease in appropriate vocalizations. Future studies may evaluate the use of response cost in all conditions.

Second, we did not collect data on appropriate object manipulation. Anecdotally, participants manipulated the toys themselves during most of the session. Although experimenter activation of toys ensured consistency across sessions and conditions, adults may not always be available in natural settings to activate toys if the child does not do so independently. Finally, it may be helpful to examine more naturalistic demands during RIRD instead of using vocal imitation of arbitrary words. Examples might include taking the toy the child currently is playing with (e.g., a tiger toy) and prompting the appropriate tact (e.g., “say tiger,” or “say yellow tiger”) More relevant demands may increase the likelihood that appropriate vocalizations will come under the control of the appropriate stimuli in the environment.

Additional research on MS and RIRD is necessary to examine the long-term effects of these interventions. It would be helpful to gain information on whether post-RIRD sessions contained higher levels of stereotypy compared to presession levels. This information may be used to identify the utility of these procedures in natural settings.

Footnotes

This study is based on a thesis submitted by the first author under the supervision of the second author to the Psychology Department at California State University, Sacramento in partial fulfillment of the requirements for an MA degree in Psychology: Applied Behavioral Analysis. Jessica Love is now at the University of Nevada, Las Vegas. We thank Timothy Fechter for his assistance with data collection and Tiffany Kodak for her editorial suggestions.

Action Editor, Tiffany Kodak

REFERENCES

  1. Ahearn W. H, Clark K. M, DeBar R, Florentino C. On the role of preference in response competition. Journal of Applied Behavior Analysis. 2005;38:247–250. doi: 10.1901/jaba.2005.36-04. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ahearn W. H, Clark K. M, MacDonald R. P. F, Chung B. I. Assessing and treating vocal stereotypy in children with autism. Journal of Applied Behavior Analysis. 2007;40:263–275. doi: 10.1901/jaba.2007.30-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ahrens E. N, Lerman D. C, Kodak T, Worsdell A. S, Keegan C. Further evaluation of response interruption and redirection as treatment for stereotypic behavior. Journal of Applied Behavior Analysis. 2011;44:95–108. doi: 10.1901/jaba.2011.44-95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (4th ed., text rev.) Arlington, VA: Author; 2005. [Google Scholar]
  5. Athens E. S, Vollmer T. R, Sloman K. N, St. Peter Pipkin C. An analysis of vocal stereotypy and therapist fading. Journal of Applied Behavior Analysis. 2008;41:291–297. doi: 10.1901/jaba.2008.41-291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. DeLeon I. G, Iwata B. A. Evaluation of a multiple-stimulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis. 1996;29:519–533. doi: 10.1901/jaba.1996.29-519. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Duffy-Cassella M, Sidener T. M, Sidener D. W, Progar P. R. Response interruption and redirection for vocal stereotypy in children with autism: A systematic replication. Journal of Applied Behavior Analysis. 2011;44:169–173. doi: 10.1901/jaba.2011.44-169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Durand V. M, Carr E. G. Social influences on “self-stimulatory” behavior: Analysis and treatment applications. Journal of Applied Behavior Analysis. 1987;20:119–223. doi: 10.1901/jaba.1987.20-119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Falcomata T. S, Roane H. S, Hovanetz A. N, Kettering T. L. An evaluation of response cost in the treatment of inappropriate vocalizations maintained by automatic reinforcement. Journal of Applied Behavior Analysis. 2004;37:83–87. doi: 10.1901/jaba.2004.37-83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fisher W. W, Lindauer S. E, Alterson C. J, Thompson R. H. Assessment and treatment of destructive behavior maintained by stereotypic object manipulation. Journal of Applied Behavior Analysis. 1998;31:513–527. doi: 10.1901/jaba.1998.31-513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fisher W. W, Piazza C. C, Bowman L. G, Amari A. Integrating caregiver report with a systematic choice assessment to enhance reinforcer identification. American Journal on Mental Retardation. 1996;10:15–25. [PubMed] [Google Scholar]
  12. Holcomb A, Wolery M. Comparative single subject research: Description of designs and discussion of problems. Topics in Early Childhood Special Education. 1994;14:119–127. [Google Scholar]
  13. Iwata B. A, Dorsey M. F, Slifer K. J, Bauman K. E, Richman G. S. Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis. 1994;27:197–209. doi: 10.1901/jaba.1994.27-197. (Reprinted from Analysis and Intervention in Developmental Disabilities,2, 3–20, 1982) [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Koegel R. L, Covert A. The relationship of self-stimulation to learning in autistic children. Journal of Applied Behavior Analysis. 1972;5:381–387. doi: 10.1901/jaba.1972.5-381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lanovaz M. J, Fletcher S. E, Rapp J. T. Using relative measures to identify stimuli that alter immediate and subsequent levels of vocal stereotypy: A further analysis of functionally matched stimulation. Behavior Modification. 2009;33:682–704. doi: 10.1177/0145445509344972. [DOI] [PubMed] [Google Scholar]
  16. Lerman D. C, Iwata B. A. A methodology for distinguishing between extinction and punishment effects associated with response blocking. Journal of Applied Behavior Analysis. 1996;29:231–234. doi: 10.1901/jaba.1996.29-231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lewis M, Bodfish J. Repetitive behavior disorders in autism. Mental Retardation and Developmental Disabilities Research Reviews. 1998;4:80–89. [Google Scholar]
  18. Liu-Gitz L, Banda D. R. A replication of the RIRD strategy to decrease vocal stereotypy in a child with autism. Behavioral Interventions. 2010;25:77–87. [Google Scholar]
  19. Lovaas O. I, Newsom C, Hickman C. Self-stimulatory behavior and perceptual reinforcement. Journal of Applied Behavior Analysis. 1987;20:45–68. doi: 10.1901/jaba.1987.20-45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Matson J. L, Kiely S. L, Bamburg J. W. The effect of stereotypies on adaptive skills as assessed with the DASH-II and Vineland adaptive adaptive behavior scales. Research in Developmental Disabilities. 1997;18:471–476. doi: 10.1016/s0891-4222(97)00023-1. [DOI] [PubMed] [Google Scholar]
  21. Michael J. L. Concepts and principles of behavior analysis (rev. ed.) Kalamazoo, MI: Association for Behavior Analysis International; 2003. [Google Scholar]
  22. Miguel C. M, Clark K, Tereshko L, Ahearn W. H. The effects of response interruption and redirection and selective serotonin reuptake inhibitors (sertraline) on vocal stereotypy. Journal of Applied Behavior Analysis. 2009;42:883–888. doi: 10.1901/jaba.2009.42-883. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mudford O. C, Martin N. T, Hui J. K. Y, Taylor S. A. Assessing observer accuracy in continuous recording of rate and duration: Three algorithms compared. Journal of Applied Behavior Analysis. 2009;42:527–539. doi: 10.1901/jaba.2009.42-527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. O'Reilly M, Edrisinha C, Sigafoos J, Lancioni G, Cannella H, Machalicek W, et al. Manipulating the evocative and abative effects of an establishing operation: Influences on challenging behavior during classroom instruction. Behavioral Interventions. 2007;22:137–145. [Google Scholar]
  25. O'Reilly M. F, Sigafoos J, Edrisinha C, Lancioni G, Cannela H, Choi H. Y, et al. A preliminary examination of the evocative effects of the establishing operation. Journal of Applied Behavior Analysis. 2006;39:239–242. doi: 10.1901/jaba.2006.160-04. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. O'Reilly M, Sigafoos J, Lancioni G, Rispoli M, Lang R, Chan J, et al. Manipulating the behavior-altering effect of the motivating operation: Examination of the influence on challenging behavior during leisure activities. Research in Developmental Disabilities. 2008;29:333–340. doi: 10.1016/j.ridd.2007.06.004. [DOI] [PubMed] [Google Scholar]
  27. Piazza C. C, Adelinis J. D, Hanley G. P, Goh H. L, Delia M. D. An evaluation of the effects of matched stimuli on behaviors maintained by automatic reinforcement. Journal of Applied Behavior Analysis. 2000;33:13–27. doi: 10.1901/jaba.2000.33-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Piazza C. C, Fisher W. W, Hanley G. P, LeBlanc L. A, Worsdell A. S, Lindauer S. E, et al. Treatment of pica through multiple analyses of its reinforcing functions. Journal of Applied Behavior Analysis. 1998;31:165–189. doi: 10.1901/jaba.1998.31-165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rapp J. T. Toward an empirical method for identifying matched stimulation for automatically reinforced behavior: A preliminary investigation. Journal of Applied Behavior Analysis. 2006;39:137–140. doi: 10.1901/jaba.2006.37-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Rapp J. T. Further evaluation of methods to identify matched stimulation. Journal of Applied Behavior Analysis. 2007;40:73–88. doi: 10.1901/jaba.2007.142-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Repp A. C, Barton L. E. Naturalistic observations of institutionalized retarded persons: A comparison of licensure decisions and behavioral observations. Journal of Applied Behavior Analysis. 1980;13:333–341. doi: 10.1901/jaba.1980.13-333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Roscoe E. M, Carreau A, MacDonald J, Pence S. T. Further evaluation of leisure items in the attention condition of functional analyses. Journal of Applied Behavior Analysis. 2008;41:351–364. doi: 10.1901/jaba.2008.41-351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Smith R. G, Russo L, Le D. D. Distinguishing between extinction and punishment effects of response blocking: A replication. Journal of Applied Behavior Analysis. 1999;32:367–370. doi: 10.1901/jaba.1996.29-231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Taylor B. A, Hoch H, Weissman M. The analysis and treatment of vocal stereotypy in a child with autism. Behavioral Interventions. 2005;20:239–253. [Google Scholar]
  35. Vollmer T. R. The concept of automatic reinforcement: Implications for behavioral research in developmental disabilities. Research in Developmental Disabilities. 1994;15:187–207. doi: 10.1016/0891-4222(94)90011-6. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Applied Behavior Analysis are provided here courtesy of Society for the Experimental Analysis of Behavior

RESOURCES