Abstract
Competing stimulus assessments (CSAs) are used to empirically identify stimuli associated with low levels of problem behavior. For some individuals with automatically maintained behavior, it can be difficult to identify effective competing stimuli. Recent research shows that prompting engagement and response blocking can be employed during the CSA to obtain significant reductions in problem behavior. The purpose of the present study was to replicate and extend prior research on the use of these tactics not only with competing stimuli, but also competing tasks, which require the active completion of a discrete response or response sequence. In addition, the current study validated the results of these pretreatment assessments in an extended treatment analysis, and examined the isolated and combined effects of prompting and response blocking within a component analysis. Future research directions and implications for clinical practice are discussed.
Keywords: competing stimulus assessment, competing task assessment, engagement, response blocking, stereotypy
1 |. INTRODUCTION
Many treatments for automatically maintained behavior, including stereotypy and problem behavior, rely on providing noncontingent access to alternative sources of reinforcement (see Gover et al., 2019; Rooker et al., in press). Competing stimulus assessments (CSAs) have utility in identifying stimuli that can produce reductions in automatically maintained behavior, ostensibly via reinforcer competition or substitution (Haddock & Hagopian, 2020). CSAs involve systematically presenting test stimuli singly while recording engagement with the stimulus and the occurrence of the problematic behavior of concern (e.g., self-injury or stereotypy). Stimuli associated with marked reductions in the problematic behavior, relative to a no-stimulus control condition, are then used in the context of treatment. However, for some individuals, simply making competing stimuli freely available may not be sufficient to shift allocations in responding from the problematic behavior to engagement with stimuli. When individuals show almost exclusive responding toward the problematic behavior, and/or limited engagement with stimuli during the CSA, additional tactics have been used to promote engagement with stimuli and disrupt problematic behavior.
For example, Jennett et al. (2011) reported on one individual diagnosed with cerebral palsy and an intellectual disability with automatically maintained self-injurious behavior (SIB) where the initial CSA failed to identify any stimuli that markedly reduced SIB. The initial CSA conducted involved offering each stimulus freely. After not obtaining a significant reduction in SIB, the authors then modified the CSA to include additional tactics which were applied across two conditions. In one condition, the CSA was repeated (stimuli were made freely available), but if the participant did not contact the stimulus (or did not maintain contact with the stimulus) for 5 s, the therapist represented the stimulus (i.e., CSA with re-presentation). Re-presentation of the stimulus increased engagement with most stimuli and produced some reductions in SIB with some stimuli. In the second condition, stimuli were available, and re-presentation was used if contact ceased (as in the previous condition), but the therapist also blocked all SIB attempts (i.e., CSA with re-presentation and response blocking [RB]). The combination of re-presentation and RB greatly decreased SIB and increased engagement with most stimuli. An extended treatment analysis comparing the various conditions demonstrated that access to competing stimuli with re-presentation and RB was highly efficacious in reducing SIB. The study also demonstrated that modifications such as re-presentation and RB could be employed during the CSA when free access to stimuli does not produce reductions in problem behavior.
More recently, Hagopian et al. (2020) described outcomes of an augmented-CSA (A-CSA), employing similar tactics with six consecutively encountered cases with treatment-resistant subtypes of automatically maintained SIB or stereotypy. Individuals were diagnosed with intellectual and/or developmental disabilities, and their SIB was classified as Subtype 2 or 3, and stereotypy as Subtype 2 (as defined by Hagopian et al., 2015). When test stimuli were made freely available, only one effective competing stimulus (defined as a stimulus that reduced SIB or stereotypy by at least 80%) was identified in two of six cases. Extending the procedures described by Jennett et al. (2011), the CSA was repeated using additional tactics: engagement with stimuli was prompted in one condition, and was combined with RB and redirection in a second condition. Improvements in stimulus engagement and reductions in problematic behavior were observed across all individuals. Most importantly, when those tactics were withdrawn and the test stimuli were again freely available without the adjunctive procedures, the number of effective competing stimuli increased in all six cases (range of 1–7 stimuli across cases). The results suggested that tactics promoting engagement with stimuli and disruption of problem behavior may facilitate contact with reinforcement available via engagement. In another study, Leif et al. (2020) employed prompting and contingent reinforcement to increase engagement in the context of a CSA with four participants with automatically maintained problem behavior. These initial findings are encouraging; however, further replication is needed, along with extended evaluations of interventions applied over longer periods of time to determine if those effects can be maintained (see Leif et al., 2020).
The purposes of the present study were to (a) replicate and extend aforementioned findings demonstrating that tactics aimed at promoting engagement and/or disrupting problematic behavior can be employed in the context of CSAs, (b) evaluate the generality of this approach to identify competing tasks, which involve the active completion of a discrete response or response sequence, and (c) conduct an extended treatment evaluation and component analysis to validate the findings of these pretreatment assessments over an extended period of time, and to determine the relative contribution of prompting and RB components.
2 |. GENERAL METHOD
2.1 |. Participant and setting
Jeb was a 21-year-old male diagnosed with autism spectrum disorder, admitted to a specialized inpatient unit for the assessment and treatment of problem behavior; however, motor stereotypy was also targeted for treatment. He had limited functional communication (e.g., he could communicate using gestures or one-word utterances) and could follow one-step directions. Results from a functional analysis (FA; Iwata et al., 1982/1994) of motor stereotypy indicated it was maintained by automatic reinforcement, and would be further classified as Subtype 2. Jeb engaged in stereotypy at high levels across the alone, attention, toy play, and demand sessions of the FA, during 98.98%, 91.47%, 98.03%, and 98.69% of sessions, respectively; there was a lack of differentiation amongst conditions (FA data are available from corresponding author). Sessions were conducted in padded session rooms (approximately 3 × 3 m) 5 days per week for approximately 30–60 min per day. Each room had a one-way observation window and contained two chairs, a desk, and relevant session materials.
2.2 |. Response measurement and interobserver agreement
A computerized data collection program (BDataPro; Bullock et al., 2017) was used by trained observers across all sessions. Continuous duration recording (CDR) methods were used to record the duration of stereotypy, which was defined as repetitive touching of the lips, nose, or eyes, or repetitive movements of one or both fingers/hands (3-s onset/offset). Engagement (with stimuli) was defined as any attempt to manipulate the stimulus (3-s onset/offset) that was not destructive or harmful to self or others. This did not include mere contact with the stimulus or use of the stimulus in a manner consistent with the targeted stereotypy. Data were also collected on RB of stereotypy (which occurred with immediate onset and 3-s offset); this was necessary to provide a measure of treatment efficacy as the duration of RB reflects how often stereotypy was occurring. Interobserver agreement (IOA) was calculated on a 10-s interval-by-interval basis using the partial interval agreement method. That is, IOA was calculated by dividing the smaller duration of stereotypy by the larger duration of stereotypy for each interval and multiplying this proportion by 100. IOA was collected during 63% of CSA sessions, 78% of CTA sessions, and 44% of treatment sessions, and partial agreement coefficients averaged 93%, 90%, and 96%, respectively.
3 |. PRETREATMENT ASSESSMENT METHODS
3.1 |. Competing stimulus assessment
Leisure stimuli selected to be tested during the CSA were selected based on the results of caregiver interviews, behavioral observations, and a multiple stimulus without replacement preference assessment (DeLeon & Iwata, 1996) suggesting they were preferred and could have potential to compete with stereotypy. Stimuli tested were: a stress ball, magazine, iPad (highest preferred item from preference assessment), hand clapper toy, suction toy, octopus toy, and plastic rings. An expandable ball toy and slinky were later included in the assessment after we were initially unable to identify effective competing stimuli. The CSA procedures were similar to those described by Hagopian et al. (2020) and included three conditions that sequentially became more treatment intensive; however, we did not repeat the initial free access in our assessment (see descriptions below). In the first condition, stimuli were made freely available (free access), then engagement with stimuli was prompted (prompted engagement [PE]), and finally, engagement was prompted with RB of stereotypy (see below for procedural details). For each condition, three series were conducted with each stimulus before moving to the next condition.
Prior to the initial trial with each stimulus, Jeb was prompted to sample each stimulus for 5–10 s. A series was comprised of: (a) test trials, where each stimulus was presented in randomized order without replacement, and (b) a no-stimulus control trial. The no-stimulus control trial was always conducted first in each series, serving as a control to isolate the effect of each test stimulus. The control trial procedures mirrored the ignore FA condition, in which no test stimuli were present in the room and all topographies of problem behavior were ignored. Trial duration (120 s) was derived from the mean duration of stereotypy within the ignore condition of previously conducted FA. Because stereotypy occurred at a high rate, the trial duration of 120 s was deemed to be of sufficient duration to obtain a representative sample, and allow for rapid evaluation of the effects of test stimuli on stereotypy. Contingencies for stereotypy during test trials varied across conditions (described below). An effective competing stimulus was defined as a stimulus associated with ≥80% reduction in stereotypy relative to the average of the control trials (excluding the RB condition).
3.1.1 |. Free access
Each trial began with the therapist placing the test stimulus on the table in front of Jeb and saying, “You can play with this if you want.” The therapist did not prompt Jeb to engage with the test stimulus or provide any consequences for stereotypy. The therapist did not prompt engagement with the stimuli or block stereotypy. If problem behavior occurred, it was blocked and ignored.
3.1.2 |. Prompted engagement
Each trial began with the therapist placing the test stimulus in Jeb’s hands and saying, “You can play with this if you want.” Contingent upon 10s of no engagement with the test stimulus, the therapist placed it in Jeb’s hands (if necessary) and physically and verbally prompted him to engage with the stimulus for 5 s (e.g., “You can play with your toy like this”).
3.1.3 |. Prompted engagement with response blocking
This condition was identical to the PE condition, except a therapist also immediately blocked the occurrence of stereotypy and redirected Jeb’s hands to the test stimulus, and briefly (up to 5 s) PE with it (using physical guidance and verbal prompting as above). Of note, attempts to engage in stereotypy were immediately blocked but the operational definition for stereotypy included a 3-s onset criterion; because the occurrence of stereotypy was effectively blocked in many instances, data were collected on duration of blocking and reported for this phase. Accordingly, the blocking procedure lasted a minimum of 3 s to compensate for the 3-s onset criteria for stereotypy. When no effective competing stimuli were identified following this condition, additional stimuli (i.e., expandable toy ball and slinky) were evaluated in subsequent trials within this prompted engagement with response blocking (PE + RB) condition. The additional stimuli were not evaluated within the free access or PE conditions because their effects (in combination with prompting and RB) could be assessed within the subsequent extended treatment evaluation.
3.2 |. Competing task assessment
The competing task assessment (CTA) was procedurally identical to the CSA in this study, except tasks were evaluated in lieu of leisure-related stimuli. In contrast to stimuli assessed in the CSA, with which the participant can typically engage with in many different ways, tasks assessed in the CTA consisted of discrete tasks that required a specific response. Tasks were selected based on caregiver interviews (e.g., parents, education coordinator, and behavioral staff) and behavioral observations indicating that Jeb could complete them rapidly and accurately with minimal to no assistance from others. Tasks selected for assessment in the CTA were: stringing beads, sorting colors, screwing bolts, completing a puzzle, lacing, taking objects apart, and stacking items. Similar to the CSA, Jeb was exposed to Free Access, PE, and PE + RB conditions. All conditions were identical to those described in the CSA except in the conditions with PE, Jeb was initially prompted to engage with the tasks every 30 s in the absence of engagement (as opposed to 10 s of no engagement with stimuli in the CSA). This timeframe for prompting was selected based on clinical observations regarding Jeb’s latency to engage with and complete the potential competing tasks independently.
4 |. PRETREATMENT ASSESSMENT RESULTS
In the CSA, Jeb engaged in high levels of stereotypy across all control trials. Additionally, within the free access and PE conditions, Jeb continued to engage in high levels of stereotypy across all test stimuli (Figure 1, top two panels). Stereotypy remained high (M = 97.5%) in the control trials within the PE + RB condition (Figure 1, bottom panel). During subsequent PE + RB test trials, test stimuli associated with high levels of stimulus engagement and that required minimal blocking were identified: a slinky (blocking, M = 0.97%; engagement, M = 80.7%) and an expandable ball (blocking, M = 5.6%; engagement, M = 82.6%). In the CTA, Jeb also engaged in high levels of stereotypy across test and control trials during the free access and PE conditions (Figure 2, top two panels). During the PE + RB condition, several test tasks associated with high levels of engagement and minimal blocking (that also required infrequent prompting) were identified: stringing beads (blocking, M = 3.5%; engagement, M = 80.6%) and screwing bolts (blocking, M = 10.4%; engagement, M = 82.4%; Figure 2, bottom panel). Together, these competing stimuli (expandable ball, slinky) and tasks (screwing bolts, stringing beads) were included in the subsequent treatment evaluation.
FIGURE 1.
Percentage of session with stereotypy/blocking (open triangles) and engagement (black bars) across free access (FA), prompted engagement (PE), and prompted engagement with response blocking (PE + RB) competing stimulus assessment (CSA) conditions. *Competing stimulus used in subsequent treatment evaluation
FIGURE 2.
Percentage of session with stereotypy/blocking (open triangles) and engagement (black bars) across free access (FA), prompted engagement (PE), and prompted engagement with response blocking (PE + RB) competing task assessment (CTA) conditions. *Competing task used in subsequent treatment evaluation
5 |. EXTENDED TREATMENT EVALUATION AND COMPONENT ANALYSIS METHODS
5.1 |. Purpose and design
Stimuli and tasks identified in the CSA and CTA, respectively, that were associated with the lowest levels of stereotypy and highest levels of engagement were evaluated in an extended treatment analysis. This analysis was performed to (a) validate the predictive utility and findings of the CSA and CTA, (b) determine the contribution of the PE and PE + RB components in combination with CSA and CTA items, and (c) to determine if these procedures could produce meaningful behavior change over a longer period of time across a series of sessions. Using a withdrawal with embedded multielement design, we further examined the effectiveness of the top two competing stimuli (expandable ball, slinky) identified in the CSA and the top two competing tasks (screwing bolts, stringing beads) identified in the CTA as separate treatments. We opted to include two of each item (in lieu of a single item) to mitigate potential satiation with any individual stimulus or task. Using a multielement design, treatment sessions alternated between those where competing stimuli were available and those where competing tasks were available. The component analysis included conditions where (a) the stimuli or tasks were made freely available (without prompting engagement or blocking stereotypy), (b) where engagement with stimuli or tasks was prompted, and/or (c) stereotypy was blocked but no stimuli or tasks were present to determine the contribution of each component (i.e., stimuli and tasks, blocking, and blocking with stimuli and tasks). Sessions were initially 5 min in duration.
5.2 |. General procedures
During all the competing stimuli treatment conditions, Jeb had access to two competing stimuli; during all the competing task treatment conditions, he had access to two tasks. At the outset of each session, Jeb could select which competing stimulus or task to engage with first. At the session midpoint (i.e., after 150 s), the therapist replaced the previously selected competing stimulus or task with the other respective item for that condition. Thus, within each session, Jeb was exposed to each of the two competing stimuli or competing tasks for an equal duration. Beginning with Session 19, the prompting schedule within the competing task condition was modified to occur after 10 s of no engagement to be commensurate with the prompting schedule used in the competing stimuli condition. Beginning with Session 78, session duration for both competing stimuli and competing task conditions were extended to 10 min, to determine if treatment effects would be maintained over longer session durations.
5.3 |. Baseline
No competing stimuli or tasks were available and stereotypy was ignored.
5.4 |. Competing stimuli or tasks with PE + RB
This condition was identical to the PE + RB condition of the CSA and CTA. That is, competing stimuli or tasks were present, engagement was prompted when criteria were met, and stereotypy was blocked. The purpose of this condition was to determine if the effects observed during the CSA and CTA could be replicated in the context of a treatment analysis conducted over a series of consecutive sessions, and to provide additional opportunities to allow Jeb to contact reinforcement available via engagement with stimuli and tasks.
5.5 |. Response blocking alone
This condition was identical to baseline (i.e., no competing stimuli or tasks were present) except the therapist blocked stereotypy. The purpose of this condition was to determine the effects of RB alone.
5.6 |. Competing stimuli or tasks with PE
This condition was identical to the PE condition of CSA and CTA (i.e., the therapist prompted Jeb to engage with the competing stimuli or tasks when the criteria were met). Although PE alone was not found to significantly reduce stereotypy in the CSA and CTA, we sought to evaluate the impact of this specific component to (a) determine if reductions in stereotypy previously observed with RB in place might be maintained in the absence of RB (as a function of learning during extended exposure to the prompting and blocking procedures), and if so, (b) to reduce response effort of the intervention.
5.7 |. Competing stimuli or tasks with RB
During this condition, the competing stimuli or tasks were present, while stereotypy was blocked; however, the therapist did not prompt Jeb to engage with the stimuli or tasks contingent upon the absence of engagement. The purpose of this condition was to (a) determine if reductions in stereotypy could be maintained in the absence of PE following exposure to the previous treatment conditions, and if so, (b) to reduce response effort of the intervention.
6 |. EXTENDED TREATMENT EVALUATION AND COMPONENT ANALYSIS RESULTS
During the initial baseline phase (BL) of the treatment evaluation, Jeb displayed elevated levels of stereotypy (M = 98.4%; Figure 3, top panel). During the following phase, the full treatment package was implemented: competing stimuli or tasks were combined with PE + RB. Stereotypy was nearly eliminated, and blocking was low across both treatment conditions (Figure 3, bottom panel). Jeb’s engagement was elevated across both competing stimuli (M = 86.1%) and competing task (M = 88.6%) treatment conditions. During the return to baseline phase, Jeb’s stereotypy increased and occurred throughout the session (M = 99.0%). Initial treatment effects were replicated when the full treatment package with PE + RB involving either competing stimuli or tasks were reapplied.
FIGURE 3.
Percentage of session with stereotypy (top panel), engagement (middle panel), and blocking (bottom panel) across the evaluation for competing stimuli (CS) and competing tasks (CT). Conditions included baseline (BL), response blocking alone (RBA), prompted engagement (PE), response blocking (RB), and combined prompted engagement with response blocking (PE + RB). Beginning with session 19, the prompting schedule within the competing task condition was modified from every 30 s to every 10 s (similar to the competing stimulus condition)
In the response blocking alone phase (where the effects of blocking alone were evaluated in the absence of competing stimuli or tasks), Jeb required frequent blocking (M = 50.0%; Figure 3, bottom panel), suggesting that the blocking component alone was not effective in isolation. Upon return to the full treatment package where competing tasks or stimuli were used in combination with PE + RB, engagement again increased across both conditions and blocking of stereotypy was minimal.
In the next phase, when PE continued to be implemented with the competing stimuli or tasks but the RB component was removed, treatment effects were diminished. Specifically, engagement decreased considerably across both competing stimuli and tasks conditions (M = 55.0% and M = 39.0%, respectively). Stereotypy was generally elevated and variable across both conditions (M = 29.1% and M = 33.7%, respectively). In the next condition when PE was withdrawn and stereotypy was blocked, effects were nearly comparable to the full treatment package. Across both the competing stimuli and tasks conditions, engagement remained relatively high (M = 79.4% and 66.0% respectively), though therapists were required to block stereotypy more often during the competing task condition relative to the competing stimuli condition (M = 16.2% and 6.6%, respectively). These effects were largely replicated in the next two phases. Although it appears RB was an essential component, the need to block stereotypy during this condition was infrequent. After session duration was extended from 5 to 10 min, engagement decreased across sessions (particularly in the competing stimuli condition), however, there was not a commensurate increase in the duration of blocking.
6.1 |. Summary of findings and supplemental quantitative analyses
Competing stimuli or tasks with prompting only, or RB alone (without competing stimuli) produced comparable reductions in stereotypy which were not clinically meaningful. However, competing stimuli or tasks with (a) both prompting and RB, or (b) with RB and no additional prompting, produced the strongest effects. Engagement (see Table 1) with competing tasks was relatively higher than with competing stimuli when session duration was extended.
TABLE 1.
Mean engagement for percentage of session (range) by condition and phase
| Competing stimulus assessment | Competing task assessment | |
|---|---|---|
| Prompted engagement with response blocking | 90.4 (75.5–98.2) | 80.0 (62.1–92.1) |
| Prompted engagement | 54.6 (21.5–95.0) | 40.9 (21.6–67.1) |
| Response blocking | 79.8 (41.8–98.0) | 70.6 (50.7–84.7) |
Overall, reductions in stereotypy (relative to baseline) with competing stimuli in the last three sessions of the PE + RB, PE, and RB phases were 100%, 59%, and 100%, respectively. Similarly, with competing tasks, reductions were 100%, 55%, and 100%. Of course, it is also important to examine the level of RB required in the PE + RB and RB conditions necessary to achieve these robust decreases in stereotypy. With competing stimuli in the PE + RB and RB conditions, therapists blocked stereotypy for 8.1% and 7.7% of the session on average, respectively. Similarly, with competing tasks, therapists blocked stereotypy for 10.3% and 8.6% of session on average, respectively. To supplement analysis of these outcomes, we conducted a series of nonparametric Kruskal–Wallis and Mann–Whitney U tests to evaluate differences in engagement levels across PE + RB, PE, and RB conditions. Differences in engagement were significant across conditions (X2 = 15.58, df = 2, p < 0.001): with engagement being higher in PE + RB conditions relative to PE (p < 0.001) and RB (p < 0.001) alone conditions. Additionally, engagement was higher in PE alone conditions relative to RB alone conditions (p < 0.001). Engagement with competing stimuli was significantly higher with competing tasks (z = 2.62, df = 1, p = 0.008) in the PE + RB condition. It is also noteworthy that engagement with the competing tasks conditions did not change considerably when the prompting schedule was increased to be equivalent to the competing stimuli conditions (i.e., 10 s), and engagement was still significantly higher in the competing stimuli conditions relative to sessions in the competing tasks conditions with 10-s prompting (z = 1.97, df = 1, p = 0.04). Moreover, there was no statistically significant difference in levels of stereotypy within the PE condition (p = 0.43) between competing stimuli or tasks conditions, nor were there significant differences in the percent of session with blocking across competing stimuli or tasks in the PE + RB (p = 1.0) or RB (p = 0.08) conditions.
7 |. GENERAL DISCUSSION
Extant research suggests certain subtypes of automatically maintained behavior may be more resistant to first-line interventions (i.e., alternative reinforcement alone via NCR with competing stimuli) for automatically maintained SIB (Hagopian et al., 2018) and other behaviors (cf. Haddock, 2017). The current study replicated and extended the findings described by Jennett et al. (2011) and Hagopian et al. (2020), and more generally, the findings reported by Leif et al. (2020)—which demonstrated the utility of employing additional tactics to promote engagement and disrupt problematic behavior within the context of CSAs. In addition to replicating these findings, the current study builds upon the literature by demonstrating that this methodology can be extended to identify competing tasks. Whereas competing stimuli assessed in the CSA could more broadly be characterized as leisure items, competing tasks assessed in the CTA consisted of discrete tasks that required a discrete response or response sequence. Despite differences, both competing stimuli and competing tasks are intended to provide alternative sources of reinforcement to compete with reinforcement produced by automatically maintained behavior. The current study also contributes to the literature by partially validating the results of the CSA and CTA, demonstrating that effective competing stimuli and tasks identified in those assessments could be efficacious when employed in an extended treatment evaluation.1 However, the component analysis revealed that competing stimuli or tasks were most effective in suppressing stereotypy when used in combination with RB. Additionally, it was shown that competing stimuli and tasks were generally equivalent in terms of their effects on stereotypy, although engagement was somewhat higher with competing stimuli.
There are several limitations in this study that warrant further discussion and also have relevance for future research and applied practice. One limitation of the current study is that the extended treatment analysis was limited to only 10 min sessions. While the extended analysis and component analysis validated the CSA and CTA results and informed us about the contribution of the various components, it will be important for future studies to observe and record the effects over much longer periods of time. Thus, the impact of these procedures over a more extended period of time is warranted given concerns regarding changes in motivating operations and satiation (e.g., Lindberg et al., 2003).
One premise for using tactics to promote engagement with stimuli and tasks (and disrupt problem behavior) during these assessments is to shift allocation of responding away from problematic behavior and thereby facilitate contact with reinforcement produced by engagement. To the extent engagement with stimuli and tasks are strengthened, the continued need for those tactics may diminish over time (see Hagopian et al., 2020). Thus, additional research is needed to help determine if increasing duration of exposure to those procedures (by increasing trial duration, or the number of series of the CSA and CTA) could improve outcomes. This seems plausible, considering that individuals with intellectual and developmental disabilities (IDD) have learning challenges that often necessitate extensive learning trials.
Although we evaluated the effects of competing stimuli and tasks across separate conditions, future studies may also consider making these concurrently available during treatment to combine their effects, allow individuals to independently allocate their responding across those alternative options, and reduce potential satiation. To this end, increasing the number of available stimuli and/or tasks (beyond two items) may also mitigate the risk of satiation across items and tasks during extended applications. However, it may be optimal to initially evaluate only a few stimuli or tasks in an extended treatment evaluation that are likely to be most effective (based on results of the initial CSA and/or CTA). Once the effectiveness of these items/tasks and procedures are validated in the extended analysis, additional stimuli or tasks that demonstrated an effect (but were not as pronounced as the prior items selected for further evaluation) in the CSA or CTA could be incorporated into the extended analysis for evaluation. Our results indicate that the same empirically validated tactics for increasing engagement and decreasing problem behavior in the context of CSAs (e.g., Hagopian et al., 2020) may have generality to CTAs as well. Continued clinical research is needed to more systematically evaluate the necessity and predictive utility of modifications made in CSA/CTA procedures in the context of treatment for automatically maintained behavior among individuals with IDD. Such work will serve to enhance the clinical utility of these procedures, as well as identify procedural modifications to CTAs and CSAs that increase their conditional effectiveness in extended application or when used in conjunction with additional behavioral interventions.
ACKNOWLEDGMENTS
Manuscript preparation was supported by Grant 2R01HD076653 from the Eunice K. Shriver National Institute of Child Health and Human Development (NICHD), and P50 HD103538 from the Intellectual and Developmental Disabilities Research Centers (IDDRC). Its contents are solely the responsibility of the author and do not necessarily represent the official views of NICHD or IDDRC.
Funding information
Intellectual and Developmental Disabilities Research Center, Grant/Award Number: P50 HD103538; Eunice Kennedy Shriver National Institute of Child Health and Human Development, Grant/Award Number: 2R01HD076653
Footnotes
A full validation would require also testing stimuli or tasks not expected to be effective in the extended analysis based on results of the CSA and CTA, respectively (Tiger & Effertz, 2020).
CONFLICT OF INTEREST
On behalf of all authors, the corresponding author states that there is no conflict of interest.
ETHICAL APPROVAL
All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- Bullock CE, Fisher WW, & Hagopian LP (2017). Description and validation of a computerized behavioral data program: “BDataPro”. The Behavior Analyst, 40(1), 275–285. 10.1007/s40614-016-0079-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- DeLeon IG, & Iwata BA (1996). Evaluation of a multiple-stimulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29(4), 519–533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gover HC, Fahmie TA, & McKeown CA (2019). A review of environmental enrichment as treatment for problem behavior maintained by automatic reinforcement. Journal of Applied Behavior Analysis, 52(1), 299–314. [DOI] [PubMed] [Google Scholar]
- Haddock JN (2017). Predicting treatment effects for problem behavior maintained by automatic reinforcement (Doctoral dissertation). University of Florida. https://ufdcimages.uflib.ufl.edu/UF/E0/05/14/83/00001/HADDOCK_J.pdf [Google Scholar]
- Haddock JN, & Hagopian LP (2020). Competing stimulus assessments: A systematic review. Journal of Applied Behavior Analysis, 53(4), 1982–2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hagopian LP, Frank-Crawford MA, Javed N, Fisher AB, Dillon CM, Zarcone JR, & Rooker GW (2020). Initial outcomes of an augmented competing stimulus assessment. Journal of Applied Behavior Analysis, 53(4), 2172–2185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hagopian LP, Rooker GW, & Zarcone JR (2015). Delineating subtypes of self-injurious behavior maintained by automatic reinforcement. Journal of Applied Behavior Analysis, 48(3), 523–543. 10.1002/jaba.236 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hagopian LP, Rooker GW, & Yenokyan G (2018). Identifying predictive behavioral markers: A demonstration using automatically reinforced self-injurious behavior. Journal of Applied Behavior Analysis, 51(3), 443–465. 10.1002/jaba.477 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Iwata BA, Dorsey MF, Slifer KJ, Bauman KE, & Richman GS (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27(2), 197–209. Reprinted from Analysis and Intervention in Developmental Disabilities, 2, 3–20. 10.1901/jaba.1994.27-197 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jennett H, Jann K, & Hagopian LP (2011). Evaluation of response blocking and re-presentation in a competing stimulus assessment. Journal of Applied Behavior Analysis, 44(4), 925–929. 10.1901/jaba.2011.44-925 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Leif ES, Roscoe EM, Ahearn WH, Rogalski JP, & Morrison H (2020). Increasing item engagement and decreasing automatically reinforced problem behavior within a modified competing stimulus assessment. Journal of Applied Behavior Analysis, 53(3), 1638–1659. 10.1002/jaba.695 [DOI] [PubMed] [Google Scholar]
- Lindberg JS, Iwata BA, Roscoe EM, Worsdell AS, & Hanley GP (2003). Treatment efficacy of noncontingent reinforcement during brief and extended application. Journal of Applied Behavior Analysis, 36(1), 1–19. 10.1901/jaba.2003.36-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rooker GW, Hagopian LP, Haddock JN, Arevalo AR, Bonner AC, & Dillon CP in press. Analysis of unexpected disruptive effects of contingent food reinforcement on automatically maintained self-injury. Journal of Applied Behavior Analysis. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tiger JH, & Effertz HM (2020). On the validity of data produced by isolated and synthesized contingencies during the functional analysis of problem behavior. Journal of Applied Behavior Analysis. Advance online publication. 10.1002/jaba.792 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.