Restricted Interests and Autism: Further Assessment of Preferences for a Variety of Leisure Items

Abstract

Researchers have yet to identify the conditions under which people diagnosed with autism spectrum disorder demonstrate restricted interests; it is possible that the impression of restricted interests is strengthened when a limited variety of items are included in a client’s preference assessments. This study will extend past research on preferences of children with autism by (a) examining participants’ preferences for unreplenished (familiar) play or leisure items versus items that are replenished frequently, (b) assessing if participants who prefer replenished items select items with properties that are matched or unmatched to their most preferred unreplenished item, and (c) assessing if participants who show an exclusive preference for unreplenished items will select replenished items during response-restriction and enhanced-replenished pool manipulations. Participants were four adolescents with autism and a caregiver-reported history of restricted interests. One participant selected both unreplenished (familiar) items and replenished (novel) items without further manipulations. The remaining three participants only selected replenished-matched leisure items after additional manipulations. Results are discussed in terms of the ethical and practical importance of assessing a range of potential reinforcers, particularly with clients who demonstrate restricted interests.

Keywords

leisure preferences restricted interests adolescents autism

Restricted Interests and Autism: Further Assessment of Unfamiliar Leisure Items

Reinforcer identification plays a key role in designing effective, individualized interventions. For example, Milo, Mace, and Nevin (2010) found that varied reinforcers maintained higher response rates, participants were more resistant to distraction in the presence of varied reinforcers, and varied edible reinforcers were preferred over constant edible reinforcers. Many studies demonstrated that more potent competing reinforcers are associated with greater reductions in problem behavior; these studies addressed a range of behaviors such as face slapping, head banging, body hitting (Ringdahl, Vollmer, Marcus, & Roane, 1997), hand mouthing (Goh et al., 1995), and vocal stereotypy (Lanovaz & Argumedes, 2009). Thus, reinforcer variety has implications for treatment efficacy as well as human rights (e.g., right to a therapeutic environment; Van Houten et al., 1998).

Practitioners may encounter unique challenges when identifying a range of reinforcers for people with autism. Indirect assessment data (e.g., parent report) suggest the tendency of people with autism spectrum disorder (ASD) to demonstrate restricted interests (e.g., Hattier, Matson, May, & Whiting, 2012; Tilton & Ottinger, 1964). Direct assessments of the range of preferences demonstrated by people diagnosed with ASD are rare in published research.

One way to assess a more complete range of reinforcers for clients with autism is to systematically introduce novel items to assessment pools. Kenzer, Bishop, Wilke, and Tarbox (2013) compared preferences for caregiver-nominated familiar toys and experimenter-selected unfamiliar toys with three children with autism. Unfamiliar toys, which shared features or functions with the familiar toys, were highly selected during preference assessments; unfamiliar items also functioned as reinforcers during a progressive-ratio reinforcer assessment. Findings of Kenzer et al. and others (Ciccone, Graff, & Ahearn, 2015) suggest that novel stimuli matched to putative reinforcing aspects of preferred items are preferred by people with autism. The present study acknowledges that, whether or not item novelty is a predictor of reinforcing value for people with ASD, practitioners need methods to identify an expanded range of reinforcers for scenarios in which (a) a client’s most preferred item is unavailable due to health concerns or resource constraints, (b) a client’s most preferred item is not currently functioning as a reinforcer, or (c) a client has historically limited opportunities to develop or demonstrate a preference for items beyond the typical assessment pool.

The study by Kenzer et al. was limited in at least two ways with respect to the goal of identifying a variety of toy or leisure preferences for clients with ASD. First, the purpose of Kenzer et al. was to evaluate preferences for familiar stimuli and matched unfamiliar stimuli, but just three unfamiliar stimuli were included per participant for the duration of the study. Because a relatively small variety of items was presented, questions remain about the breadth of leisure preferences per participant. Second, Kenzer et al. mentioned that restricted interests were characteristic of individuals diagnosed with ASD, but it was unclear if participants were included due to caregiver concerns with restricted interests.

The purpose of this study was to examine if participants with autism select stimuli from a Replenished Pool of potential reinforcers more or less often than items from an Unreplenished Pool. We chose the terms replenished and unreplenished to distinguish reinforcer pools in our study, rather than familiar and unfamiliar, because our purpose was to assess the practical utility of including a wide variety of similar and dissimilar items in preference assessments, not to demonstrate novelty as a reinforcing property of the preferred items. More specifically, we compared participants’ cumulative selections per pool to inform practitioner decisions about expanding beyond caregiver-nominated items for clients with a history of restricted interests. The current study extends prior research on conditions under which people with ASD select a variety of reinforcers by including participants with a history of restricted toy and leisure preferences, by replenishing stimuli frequently to sample preferences for a wider range of items, and by manipulating assessment procedures and pools until an expanded range of relative reinforcers is identified.

Method

Participants

Four males diagnosed with autism were recruited from a school for children with developmental disabilities. Requirements for participation included historically low to moderate levels of severe problem behavior (e.g., self-injurious behavior, aggression), a history of compliance with tabletop tasks lasting up to 10 min, and a history of scanning and selecting among multiple items as confirmed by a clinical case manager. All participants were recommended by caregivers for this study based on a history of restricted toy and leisure preferences and a history of problem behavior associated with denied access to preferred items. Standard practice in the participants’ school was to record the approximate frequency and duration of problem behaviors (e.g., property destruction, aggression) hourly from 8 a.m. to 8 p.m. We consulted each participant’s daily data sheet, which included a notes section for documenting antecedents and consequences of problem behavior, to corroborate caregiver reports.

Stu was a 16-year-old male with a wide variety of leisure skills. He engaged appropriately with most leisure items when prompted (e.g., books, Legos®, and puzzles). His primary mode of communication was vocal-verbal and he could speak in full sentences. He waited quietly and engaged with teacher-selected books or puzzles during transitions. Stu seldom requested leisure items other than use of the computer or other Internet-ready electronics. Items were rarely withheld from Stu, but more severe problem behaviors such as screaming, grabbing caregivers, and environmental destruction were documented under these conditions.

Alan was also 16 years old and demonstrated vocal-verbal skills but spoke in one- or two-word phrases and used an augmentative communication device. He demonstrated some leisure skills but rarely used these skills when asked to do so (e.g., during transitions). Electronics were rarely withheld from Stu, but under these conditions he cried, flipped over tables, hit his head, or bolted away from his teacher as confirmed by review of his daily data sheets.

Phil was a 17-year-old male who communicated in complete sentences and could carry on a conversation for three to four exchanges. He demonstrated a wide range of leisure skills when prompted. Phil requested both electronics and crafts in a variety of mediums (e.g., painting, drawing with gel pens). If Phil’s requested craft or electronic item was not available, he bargained for the requested item or, less often, screamed and bit his hand.

Doug was a 16-year-old male who communicated primarily through an augmentative device but he also used basic signs and gestures. Doug rarely interacted with leisure items other than electronics. Doug verbally perseverated about the unavailability of electronic items to the point of disrupting instructional programs for himself and his peers. Items were rarely withheld from Doug, but review of his daily data sheets suggested that more severe problem behaviors such hitting or pinching his teachers sometimes occurred under these conditions.

Setting and Materials

Sessions took place within a typical classroom at the participants’ school. Participants were seated at a desk in a quiet area of the classroom. A minimum of two sessions were conducted per visit up to 5 days per week. Each session included a one-trial choice per pool (three trials total), followed by a one-trial prompted exposure per item (three exposure trials per session), one concurrent-chain worksheet trial (Kenzer et al., 2013), and a reinforcer-access period.

Materials included academic task materials (i.e., worksheet) and three pools of leisure items (i.e., Unreplenished, Replenished-Matched, and Replenished-Unmatched). Procedures used to assemble each stimulus pool are described below.

Response Definitions and Interobserver Agreement

Trained observers scored session-by-session data live or from video. Approach responses, defined as reaching toward an item, were scored for choice-per-pool trials and for each concurrent-chain trial. After a participant selected an item during the concurrent-chain trial, the name of the item’s original pool was also recorded. Observers scored whether or not problem behavior occurred at any point during concurrent-chain trials and whether task completion on the concurrent-chain trial was independent or prompted. Point-by-point interobserver agreement (IOA) was calculated for each dependent measure (i.e., name of item selected on each choice-per-pool trial; name of item selected on concurrent-chain trial; occurrence or non-occurrence of problem behavior; prompted or independent task completion) during at least 32% of sessions. We calculated IOA by dividing the number of agreements by the total number of agreements plus disagreements for each dependent measure and multiplying by 100. Mean agreement scores were as follows: 91.7% agreement for Phil (range, 67%-100%), 96.7% agreement for Stu (range, 87%-100%), 96.4% agreement for Alan (range, 86%-100%), and 92.8% agreement for Doug (range, 71%-100%). For duration of item engagement during the reinforcer-access period of each concurrent-chain trial, IOA was calculated by dividing the shortest duration recorded by the longest duration recorded per access period and multiplying by 100. Mean agreement for duration of item engagement for Phil, Stu, Alan, and Doug was 89.69% (range, 81%-99%), 100%, 86.92% (range, 83%-100%), and 93.91% (range, 83%-100%), respectively.

Procedural integrity was assessed on the following experimenter behaviors during at least 31% of sessions per participant: presenting and removing each item during prompted exposure trials, providing access to selected items for up to 5 min after each trial of each assessment type, delivering prompts during the worksheet portion of each concurrent-chain trial, and randomizing the horizontal placement of items following each trial of each assessment type. The experimenter had to implement a step correctly on all opportunities within a session to receive credit for that step. The number of correct steps was then divided by the total number of prescribed steps for each session. Assessment sessions for Phil, Stu, Alan, and Doug were implemented with 92.86%, 93.75%, 100%, and 100% procedural integrity, respectively.

General Procedures

Participants completed a series of assessment procedures followed by a concurrent-chain assessment to determine which pool of items (i.e., Unreplenished, Replenished-Matched, or Replenished-Unmatched) was most preferred, overall. First, stimuli were identified for the Unreplenished Pool through a combination of caregiver interviews and paired stimulus (PS) preference assessments. Second, participants were exposed to stimuli in each of the three assembled pools during scheduled sampling sessions. Third, assessment sessions were conducted including three choice-per-pool trials and one concurrent-chain (worksheet) trial.

Items in the Replenished Pools were regularly updated, as follows. An item was replaced if that item was chosen during the choice-per-pool trial on four occasions, regardless of whether or not the item was chosen during the concurrent-chain trial. An item was also replaced if it had not been chosen during choice-per-pool trials for 10 sessions. The Replenished-Matched Pool and the Replenished-Unmatched Pool were fully replenished if no items were chosen during the concurrent-chain trial for four consecutive sessions. To illustrate, Table 1 lists all items that were presented to Alan in each pool over the course of the study.

Table 1.

Items Assessed per Pool With Alan.

Unreplenished Pool	Restricted Pool	Replenished-Matched Pool	Enhanced Replenished-Matched	Replenished-Unmatched
Computer Ball iPod®	Ball	Simon® Cymbals Electronic Drumsticks Maracas Sound Effects Flarp!™ Whoopee Cushion BopIt!™ Laser Fingers Light Up Ball Magic 8 Ball™ Pig Light Shaker Instrument Triangle Wooden Instrument Bongos Pop Gun Tambourine ViewMaster®	Fruit Ninja™ Horri-Ballz™ Monster Lite Brix™ PD Patrol Car	Fishing Game Tinker Toy® Perplexus™ Lincoln Logs® Rubik’s Cube® Peg Game Slinky® Tic Tac Toe Paint Posters B. Bristle Block® Jacob’s Ladder Kinetic Wheel Moon Dough™ Puppets Top Bowling Magic Hoops Tricky Fingers™

Note. Items are arranged in a descending order from the most (top) to the least preferred items (bottom) per pool.

Assembling the Stimulus Pools

Selection of Unreplenished Pool items

To determine the toys and media items to assess for the Unreplenished Pool, a PS preference assessment was conducted (Fisher et al., 1992). Caregiver-nominated leisure items were included in the assessment. The caregiver providing input had worked with the participant for at least 6 months prior to the experiment. The number of items varied from three to six per participant because caregivers were free to nominate as many or as few items as relevant for each participant. We conducted the PS preference assessments to confirm which item was most preferred and to ensure that each item in the familiar pool was associated with some level of engagement. The top-ranked three to five items identified from the PS assessment (data available on request) were designated as the Unreplenished Pool. The number of items presented on any given trial for any pool was matched to the number of familiar items.

Electronic media items (i.e., movie and computer) were most preferred for all participants. Stu’s most preferred items were a movie, iPod®, puzzle, coloring book, and book. Alan’s Unreplenished Pool included the computer, an iPod®, and a yoga ball. Unreplenished items for Doug were the computer, Legos®, PlayDoh®, and coloring. The unreplenished items for Phil were the computer, a movie, and coloring.

Selection of replenished pool items

The top-ranked item from the PS assessment was matched to other items on the basis of sensory similarities to make up the Replenished-Matched Pool. Items included in the Replenished-Matched Pool shared visual or auditory qualities with the top-preferred item for each participant (e.g., computer matched to Simon® game; iPod® matched to sound effects toy). Items included in the Replenished-Unmatched Pool (e.g., blocks, Moon Dough™) had fewer features in common with audio-visual items in the Unreplenished or the Replenished-Matched Pools. The items from both Replenished Pools were unfamiliar in that they were not previously included in the participant’s preference assessments and they were not found in the participant’s school classroom or residential facility.

Assessing Preferences Within and Across Pools

Sampling sessions

Sampling sessions occurred prior to the first assessment session and at least every five sessions, thereafter. Sampling sessions were also provided in the event that both Replenished Pools were fully replaced at the same time (i.e., participants had not selected items from either replenished pool for four consecutive concurrent-chain trials). The purpose of the sampling sessions was to ensure participants were oriented to each item prior to making a choice. During sampling, the experimenter presented each item from each of the three separate pools for 1 min per item. The experimenter turned on items that required activation. If the participant did not interact with a specific item during the first sampling trial of the first day, the experimenter spent up to 1 min modeling and physically guiding the participant through ways to use the item. The sampling trial was then repeated; any item that was not associated with interaction on the second sampling trial was replaced with an alternative item from the relevant pool.

Choice-per-pool and concurrent-chain assessments

The concurrent-chain assessment included the most preferred item from each stimulus pool. Each session began with a series of three choice-per-pool trials. The experimenter instructed the participant to choose an item from the first pool (i.e., Unreplenished, Replenished-Matched, or Replenished-Unmatched). The participant was allowed to interact with the chosen item from that pool for up to 30 s. After the choice-per-pool trial was completed for each pool, the chosen items were placed behind the corresponding colored worksheet (i.e., green for the Unreplenished Pool, orange for the Replenished-Matched Pool, and blue for the Replenished-Unmatched Pool).

Worksheet content was identical across pools to control for differing levels of task difficulty. Worksheets were comprised of previously mastered tasks that were developmentally appropriate for each participant. Each worksheet included four questions related to the same task such as drawing a shape, identifying the value of different coins, identifying a shape, and counting items. The placement of worksheets and stimuli was randomized following each trial.

After the choice-per-pool trials, a prompted exposure trial was conducted. Three worksheets were placed on the table in front of the most preferred item from each pool. The experimenter guided the participant to touch each worksheet, beginning with the first worksheet on the left. As the participant touched a worksheet, the experimenter said, “If you choose this worksheet, you can have this item” and then provided access to the corresponding item for 30 s. Participants were not required to complete the worksheet on prompted exposure trials.

During the concurrent-chain trial, the participant was asked to choose the worksheet he would like to complete to access the corresponding item from the choice-per-pool trials. Each participant was given 5 s to begin responding independently to the selected worksheet. The purpose of the worksheet trial was to verify that selected items would, at a minimum, maintain performance of previously mastered academic tasks. If the participant did not respond or responded incorrectly on the worksheet, he was then prompted using the least intrusive prompting necessary to complete the task (e.g., gesture → manual prompt → manual guidance). Regardless of whether task completion was independent or prompted, the experimenter delivered praise and the tangible item associated with the chosen worksheet for 5 min. The reinforcer-access period was terminated early if the participant indicated that he was done interacting with the selected item (e.g., saying “all done,” pushing the toy away) or if the participant had not interacted with the selected item for 20 s. Finally, sessions were terminated immediately if a participant asked to stop participating (e.g., participant says “All done,” “No more”).

Response-restriction procedure

Three participants (Phil, Alan, and Doug) responded almost exclusively to electronic items in the Unreplenished Pool. In an attempt to identify a wider range of effective reinforcers, electronic items were removed from the Unreplenished Pool (Hanley, Iwata, Lindberg, & Conners, 2003) for a subsequent assessment. In this way, new reinforcers could be identified without the distraction of highly preferred items. A response-restriction procedure was instituted for an item and any closely related items when it was selected during four out of five consecutive trials (i.e., only non-electronic items remained available in the Unreplenished Pool).

Response-restriction plus Enhanced Replenished-Matched Pool

If items in the Unreplenished Pool were, again, selected on four consecutive concurrent-chain trials, a final modification was introduced. The Replenished-Matched Pool was enhanced in that items were added to more closely approximate the putatively reinforcing sensory stimulation garnered by the most preferred unreplenished item. As all participants preferred unreplenished items of an auditory-visual nature, Enhanced Replenished-Matched Pool items yielded at least one intensified audio or visual capability. Although multi-sensory items were supplied in some of the original Replenished-Matched Pools, all items within the Enhanced Replenished-Matched Pool contained multi-sensory capabilities.

To illustrate, Alan’s original Replenished-Matched Pool included items matched to the auditory or visual dimensions of his most preferred unreplenished item (YouTube™ on a computer): instruments, laser fingers (colored lights that can be worn on fingers), and a Flarp!™ Whoopee Cushion. When the pool was enhanced, one of the items introduced was Fruit Ninja™ on an iPhone®. Although still handling an electronic item, Alan only had access to a specific game rather than free range to explore the Internet as he did with the computer. We also introduced a Horri-Ballz™ Monster that yelled silly phrases and vibrated. Finally, we included a Lite Brix™ PD Patrol Car that lit up with flashing multicolored lights, could be driven around, and could be disassembled and recreated in different ways. The purpose of presenting the Enhanced Replenished-Matched Pool was to see if matched items with intensified features would influence response allocation by our three participants with persistent restricted preferences.

Results

Results of the concurrent-chain assessment are shown in Figures 1 to 3. Stu’s selections (Figure 1) varied between the Unreplenished and Replenished-Unmatched Pools. Stu rarely selected items in the Replenished-Matched Pool. Alan, Phil, and Doug allocated responding almost exclusively to items in the Unreplenished Pool. As a result, all electronic items were removed from the Unreplenished Pool during subsequent response-restriction trials of the concurrent-chain assessment. Problem behavior was rarely observed during the concurrent-chain trials (e.g., one instance of property destruction by Phil; occasional out-of-seat behavior exhibited by all participants).

Figure 1.

Stu’s concurrent-chain selections are displayed.

Figure 2.

Phil’s (top panel) and Doug’s (bottom panel) concurrent-chain selections are displayed.

Figure 3.

Alan’s concurrent-chain selections are displayed.

As shown in Figure 2, Phil and Doug continued to choose items in the Unreplenished Pool even when the top-preferred item was restricted. As a result, we introduced a final Enhanced Replenished-Matched Pool plus response-restriction phase (i.e., sensory features intensified) during which Phil and Doug began selecting items from the Enhanced Replenished-Matched Pool.

During response-restriction, Alan began selecting items from the Replenished Pools as well as less preferred items in the Unreplenished Pool (Figure 3). Electronic items were re-introduced to the Unreplenished Pool and Alan’s responding reverted to exclusive preference for electronics in the Unreplenished Pool. In the next phase, we combined an Enhanced Replenished-Matched Pool with response-restriction and Alan’s preferences shifted toward items in the Enhanced Replenished-Matched Pool.

Figure 4 depicts results for percentage independence on worksheet trials and duration of engagement (described below) for items that were selected most frequently during concurrent-chain trials. These data suggest that a variety of reinforcers were identified within and between pools. Regarding percentage independence, Stu and Allen were most likely to complete their worksheet without assistance when a replenished-matched item was selected in the concurrent-chain trial. Phil performed independently regardless of item category. Finally, Doug was most independent when unreplenished items were available for worksheet completion. He required prompts to complete the worksheet all five times that a replenished-matched item was available, casting some doubt on the potency of items identified from that pool.

Figure 4.

Independent performance on worksheet trials associated with items from each pool (top panel) and mean duration of engagement for top-preferred items on concurrent-chain trials (bottom panel).

Regarding item engagement, three of four participants engaged with one or more items from a replenished pool during concurrent-chain trials (range, 1 min 24 s to 5 min per item). The remaining participant, Doug, spent just 6 to 12 s engaged with the items he selected from the Enhanced Replenished-Matched and Replenished-Unmatched Pools. Phil, by contrast, played with his most preferred item from the Enhanced Replenished-Matched Pool (Fruit Ninja™) for a mean of 5 min across two concurrent-chain trials and he played with the second most preferred item (Plants v. Zombies™) from the same pool for a mean of 2 min 54 s across two concurrent-chain trials. Stu also spent the full 5 min engaging with Moon Dough™ from the Replenished-Unmatched Pool. Finally, Alan’s mean engagement data suggested a hierarchy of reinforcers between pools (top-preferred unreplenished item = 5 min; top-preferred enhanced replenished-matched item = 3 min 54 s; top-preferred replenished-matched item = 3 min 18 s; top-preferred replenished-unmatched item = 1 min 24 s).

Discussion

Results of this study extend previous research (Ciccone et al., 2015; Hanley et al., 2003; Kenzer et al., 2013) by demonstrating participants often selected unreplenished items but they also selected replenished items if (a) historically preferred items were unavailable and (b) the replenished items were matched to multiple sensory properties of the unreplenished items. These results suggest the importance of including a variety of items in preference assessments for people who demonstrate restricted interests. This study also extends the work of Kenzer et al. (2013) by assessing preferences for unfamiliar items that were replenished frequently (i.e., after a maximum of four selections). Participants in the current study were assesed on at total of 20 to 39 replenished items.

All four participants came to us labeled by caregivers as students with restricted interests for electronic items of one form or another. Participants were given carte blanche in selecting games and videos available on electronic devices in the Unreplenished Pool. Results might differ had we restricted the use of devices to a single preferred game or video (e.g., participants might have selected replenished items prior to response-restriction) or had we assessed the preferences of younger children with ASD as in Kenzer et al.’s (2013) work.

Findings of the current study should be interpreted in light of several limitations. First, the response-restriction procedure may or may not have contributed to effects of the Enhanced Replenished-Matched Pool in the final phase of assessment for Alan, Phil, and Doug. In addition, Phil’s and Doug’s initial response-restriction phase may have been too brief for them to contact reinforcing properties of replenished items. Future researchers could include an extended response-restriction phase to investigate the effects of repeated sampling on participants’ preference for new items when familiar items are reinstated. If repeated sampling is insufficient to establish new preferences, researchers could test strategies such as embedding known reinforcers and teaching relevant verbal or motor skills to promote ease of participation with socially important activities and toys (e.g., Hanley, Tiger, Ingvarsson, & Cammilleri, 2009; Potter, Hanley, Augustine, Clay, & Phelps, 2013).

Second, Phil’s and Alan’s top-preferred item in the Enhanced Replenished-Matched Pool was closely related to their most preferred item in the Unreplenished Pool (i.e., repeatedly watching segments of a small number of YouTube videos on the computer). It is unclear if such a minor expansion of documented leisure preferences is sufficient to improve quality of life for a person with disabilities.

A third limitation is that participants might have chosen some replenished items due to demand features of the procedures (i.e., the experimenter saying “Choose”) rather than the reinforcing effects of those items. Doug’s limited item engagement and his low levels of independence on academic tasks during the concurrent-chain trials lend support to this interpretation. Future studies should evaluate the effects of assessment-informed reinforcers on more socially valid target skills (e.g., unmastered skills, vocational tasks).

A fourth limitation was that we did not systematically choose items for the Replenished-Matched Pools based on the critical reinforcing dimensions of high-preference unreplenished items; this may explain why replenished-matched items were never preferred without a response-restriction procedure in place. Future researchers could compare duration of engagement across modified highly preferred items in an attempt to isolate the most reinforcing feature (e.g., put tape over character illustrations or light and video functions; disable auditory functions); results could then be used to identify unfamiliar items that share that critical feature. Finally, from a practical standpoint, assessing a wide range of potential reinforcers may require the purchase of additional leisure items in settings that cannot readily share or rotate stimuli between groups of students. Many of the items in our study were borrowed from a stock cabinet at the students’ school; however, we spent approximately US$100 to purchase additional items for the replenished pools.

A concerning issue for the rights and effective treatment of people with disabilities today may be caregivers’ low likelihood of assessing a range of potential reinforcers, particularly when clients do not request or show immediate interest in new items (Stocco, Thompson, & Rodriguez, 2011). By way of illustration, Stu was referred to the current investigation because he exclusively requested the computer. Given the opportunity, however, he selected and interacted with a variety of items (e.g., Moon Dough™, magic loops toy) instead of the computer. Results of this study showed that participants with ASD demonstrate additional preferences given a varied stimulus pool and response restriction. Further research can clarify conditions under which varied reinforcers enhance the efficacy (e.g., Keyl-Austin, Samaha, Bloom, & Boyle, 2012) and social validity of ABA services. At a minimum, caregivers who seek out a wide range of potential reinforcers will avoid a scenario in which assessment practices contribute to the impression or the actuality of restricted interests among people with ASD.

Footnotes

Acknowledgements

We thank Dr. Eileen Roscoe, Dr. Rick Graff, and Mariana Guimaraes for their comments on earlier drafts of this manuscript.

Authors’ Note

This study was carried out by the first author, supervised by the second author, in partial fulfillment of the requirements for the master of science degree at Western New England University, Springfield, Massachusetts. Monica A. Spear is now at Advances Learning Center, in Watertown, Massachusetts. Emily White is now at the Behavior Analysis Certification Board in Littleton, Colorado.

Declaration of Conflicting Interests

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

Funding

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

Author Biographies

Monica A. Spear is a consultant at Advances Learning Center in Watertown, MA. Her professional interests include methods for identifying reinforcers and providing training to caregivers of clients with ASD.

Amanda Karsten is an associate professor of Psychology at Western New England University. Her clinical and scholarly interests include treatment of social-communication deficits of people with ASD across the lifespan, practitioner training, and translation of behavioral interventions from research to practice.

Emily A. White is the director of credentialing at the Behavior Analyst Certification Board. She has expertise designing and overseeing home-based early intensive behavioral intervention programs as well as consultation to public and private schools. Her professional interests include behavior analyst credentialing, early intensive behavioral intervention, and practitioner training.

References

Ciccone

F. J.

Graff

R. B.

Ahearn

W. H.

(2015). Increasing the efficiency of paired-stimulus preference assessments by identifying categories of preference. Journal of Applied Behavior Analysis, 48, 221-226.

Fisher

Piazza

C. C.

Bowman

L. G.

Hagopian

L. P.

Owens

J. C.

Slevin

(1992). A comparison of two approaches for identifying reinforcers for persons with severe and profound disabilities. Journal of Applied Behavior Analysis, 25, 491-498.

Goh

Iwata

B. A.

Shore

B. A.

DeLeon

I. G.

Lerman

D. C.

Ulrich

S. M.

Smith

R. G.

(1995). An analysis of the reinforcing properties of hand mouthing. Journal of Applied Behavior Analysis, 28, 269-283.

Hanley

G. P.

Iwata

B. A.

Lindberg

J. S.

Conners

(2003). Response-restriction analysis: I. assessment of activity preferences. Journal of Applied Behavior Analysis, 36, 47-58.

Hanley

G. P.

Tiger

J. H.

Ingvarsson

E. T.

Cammilleri

A. P.

(2009). Influencing preschoolers’ free-play activity preferences: An evaluation of satiation and embedded reinforcement. Journal of Applied Behavior Analysis, 42, 33-41.

Hattier

M. A.

Matson

J. L.

May

A. C.

Whiting

S. E.

(2012). Repetitive/restricted behaviours and interests in children with cerebral palsy and autism spectrum disorder. Developmental Neurorehabilitation, 15, 178-184.

Kenzer

A. L.

Bishop

M. R.

Wilke

A. E.

Tarbox

J. R.

(2013). Including unfamiliar stimuli in preference assessments for young children with autism. Journal of Applied Behavior Analysis, 46, 689-694.

Keyl-Austin

A. A.

Samaha

A. L.

Bloom

S. E.

Boyle

M. A.

(2012). Effects of preference and reinforcer variation on within-session patterns of responding. Journal of Applied Behavior Analysis, 45, 673-641.

Lanovaz

M. J.

Argumedes

(2009). Using the three-component multiple-schedule to examine the effects of treatments on stereotypy. Journal on Developmental Disabilities, 15, 64-68.

10.

Milo

Mace

F. C.

Nevin

J. A.

(2010). The effects of constant versus varied reinforcers on preference and resistance to change. Journal of the Experimental Analysis of Behavior, 3, 385-394.

11.

Potter

J. N.

Hanley

G. P.

Augustine

Clay

C. J.

Phelps

M. C.

(2013). Treating stereotypy in adolescents diagnosed with autism by refining the tactic of “using stereotypy as reinforcement.” Journal of Applied Behavior Analysis, 46, 407-423.

12.

Ringdahl

J. E.

Vollmer

T. R.

Marcus

B. A.

Roane

H. S.

(1997). An analogue evaluation of environmental enrichment: The role of stimulus preference. Journal of Applied Behavior Analysis, 30, 203-216.

13.

Stocco

C. S.

Thompson

R. H.

Rodriguez

N. M.

(2011). Restricted interests and teacher presentation of items. Journal of Applied Behavior Analysis, 44, 499-512.

14.

Tilton

J. R.

Ottinger

D. R.

(1964). Comparison of the toy play behavior of autistic, retarded, and normal children. Psychological Reports, 15, 967-975.

15.

Van Houten

Axelrod

Bailey

J. S.

Favell

J. E.

Foxx

R. M.

Iwata

B. A.

Lovass

O. I

. (1998). The right to effective behavioral treatment. The Behavior Analyst, 11, 111-114.