Case Study: Extension of a Systematic Data-Based Reinforcement System to Increase Independence in Public-School Inclusion Settings to a Student with Down Syndrome and Intellectual Disability.
Prompt dependency in inclusion-setting students with intellectual disabilities is a challenge. Using a scientist-practitioner model, an intervention to shape independent responding was implemented by paraprofessionals for a student with Down syndrome. It included systematic reinforcement-schedule thinning, reduced paraprofessional proximity, and incorporation of peer models into prompting strategies. Independence on nine pre-academic skills improved by 25-100% (mean =68.1%) as reinforcement criteria increased and paraprofessional proximity decreased. Although the intervention significantly departed from treatment as usual, social validity among stakeholders was high.
Keywords: Down syndrome, inclusion independence, learning-to-learn, paraprofessional
Children with Down syndrome, a neurodevelopmental disability occurring in approximately 14 of every 10,000 live US births (Presson et al., 2013), are frequently educated in inclusive settings (Lorenz, 1999). This has increased the need for resources that assist teachers in promoting student success. Social skills and independently following classroom routines are among the teacher-identified "learning-to-learn" skills that better predict inclusion success than specific academic skills (Kemp & Carter, 2006), but students with disabilities often have difficulty with them. For example, expressive language delays in Down syndrome affect the student's ability to participate in group work and initiate peer interactions, which are essential inclusion-setting skills (Engevik, Naess, & Bernstsen, 2016). In addition, skills taught to students in one-to-one settings often do not generalize to group-instruction settings (Charania et al., 2010).
There is an extensive literature on effective strategies to promote independence on learning-to-learn skills in children with autism spectrum disorder (ASD) in inclusive settings (e.g., Saini, Miller, & Fisher, 2016), but it has not been extended to Down syndrome. This represents an area of need, especially since parents of children with Down syndrome are more likely to endorse inclusive placements than parents of children with ASD (Kasari, Freeman, Bauminger, & Alkin, 1999).
In inclusion classrooms, students with severe disabilities are often assigned 1:1 paraprofessionals who repeat/modify teacher instructions and prompt student responses. Often, student attending is directed toward the paraprofessional rather than the teacher, paraprofessionals may remain near the student (Giangreco, Edelman, Luiselli, & MacFarland, 1997), and prompt dependency develops. Often, paraprofessionals are not trained on strategies to increase student independence (Giangreco et al., 1997). Whereas 76% and 83% of paraprofessionals reported receiving training on teaching academic skills and implementing (unspecified) behavior-management programs, respectively (U.S. Office of Special Education Programs, 2002), training on strategies to increase student independence infrequently occurs (Hundert, 2009).
The field of applied behavior analysis has championed the scientist-practitioner model (SPM) for over 50 years (Baer, Wolf, & Risley, 1968; Kelley et al., 2015), and the current intervention evolved using that model. The first author had been tasked with improving independence by students with ASD in inclusion settings. They displayed difficulty generalizing previously learned skills to the inclusion environment and were dependent on paraprofessional-provided prompts. The intervention was a systematic, data-based reinforcement system that allowed paraprofessionals to use daily data to determine reinforcement criteria for independence on pre-academic skills. This case study describes an extension of that intervention to a student with Down syndrome, representing an initial step toward meeting the need for behavioral interventions for this population. The goal was to gradually increase independence while simultaneously decreasing the student's proximity to the paraprofessional so that stimulus control over behavior transferred from the paraprofessional-provided prompt to the natural classroom cues.
The participant was 55 months old and had diagnoses of Down syndrome and intellectual disability. She attended two daily 2.75-hour sessions in an inclusion preschool four days per week. She received 40-60 min per day of discrete trial instruction (DTI) beginning in January of the school year.
Materials and Reinforcers
Materials included the data book, daily "cheat sheets," clipboard, tokens, and back-up reinforcers. The data book contained daily data sheets with current and subsequent reinforcement criteria and acquisition charts for each targeted skill. Acquisition charts included introduction/mastery dates for each criterion and operational definitions. The cheat sheet was a chart attached to the front of the clipboard containing only current reinforcement criteria for each skill and a section for data entry. This allowed quick behavior tracking and served as an easy reference for the paraprofessional. Cheat-sheet data were transferred daily to the data book. Reinforcement criteria changes were noted on the cheat sheet.
Initially, reinforcers were toys, but a token economy was established after a few weeks. Tokens were introduced using backward chaining by initially providing all but the last token (out of five) when the reinforcement criterion was met, then all but the last two, etc. The token board was attached to the back of the clipboard. When a token was earned, it was immediately removed from a Velero strip, producing a distinctive sound, and placed on the token board. The student was not shown the token/told why it was delivered to allow continued attending to lessons. No procedure to ensure attending to the token was implemented; however, the sound of the token being removed seemed to serve as a conditioned reinforcer. Tokens were exchanged for back-up reinforcers (small toys, computer time, etc.) throughout the day. The same tokens were used for all target skills. Informal preference assessments (presenting options and asking the student to choose) were conducted each time the token board was filled.
The behavioral target was independence on nine learning-to-learn skills (see Table 1). The first author conducted baseline probes on 10 potential targets. The baseline percentage of independent intervals/opportunities to display the skills was zero for six of the skills probed. Baseline percentages for independently attending, following group directions, and imitating peers were 7%, 14%, and 50%, respectively. The baseline percentage for referencing peers during play was 87.5% and was not targeted for intervention.
Initially, the reinforcement criterion for all targeted skills was one 3-s interval (or one instance) of the skill with the paraprofessional nearby. The intervention shaped independent responding by incrementally increasing reinforcement criteria for each skill, with a least-to-most prompt hierarchy to transfer stimulus control from paraprofessional-provided prompts to peer models/other classroom cues. No concurrent strategies (e.g., visual schedules) were used. The first author pre-determined incremental increases to the reinforcement criteria, which gradually increased in terms of frequency of responses/ number of intervals with correct responses. A criterion for increasing paraprofessional proximity was included when appropriate. For example, the initial reinforcement criterion for attending was one 3-s interval with the paraprofessional nearby for three consecutive days. The number of required (not necessarily consecutive) intervals increased to 2, 3, 5, 7,10, and 12 as each respective criterion was met. Then, the criterion was reset to one, with the paraprofessional now at least 10 feet away; it subsequently increased again as previously described.
The paraprofessional implemented the intervention by observing the student while standing behind her and holding the clipboard with the cheat sheet and token board. Reinforcement occurred immediately upon the display of any target behavior at the current reinforcement criterion, every time the criterion was met. For criteria involving intervals, the paraprofessional counted time covertly so her attending remained directed at the student. The paraprofessional stood at varying distances from the student throughout the day, depending on the current reinforcement criterion for the target skill that was likely to occur during each activity.
Back-up reinforcer delivery when the token board was full was immediate, regardless of whether a classroom activity was completed. The student was shown the full token board, and she was allowed access to back-up reinforcers in a separate area. If an opportunity to engage in a target skill occurred but the student did not display it, she was prompted. The paraprofessional was instructed to use the lowest prompt level necessary and to vary prompts. Prompting peer referencing was stressed over strict adherence to the hierarchy, which included: (1) a delay of 10-15 s, (2) a distance (walking closer to the Student--increased distance was resumed upon independent responding) or audible prompt (tapping the clipboard with a pen), (3) a silent point/gesture at peers, (4) asking, "What are your friends doing?" with a point or gesture, (5) asking the teacher to repeat her direction with a model, and (6) more intrusive prompts as necessary. Only praise was available for prompted responses.
Daily data. Only one data point needed to be recorded per skill each day: whether the reinforcement criterion was met at least once during the day ("+") or not ("-"). However, reinforcement continued at the specified criterion throughout the day. When the student achieved three consecutive plusses at a particular criterion for a skill, the next criterion was implemented. If the criterion for a skill was not met, reinforcement continued at that criterion level.
Probe data. Probe data were collected by the first author for target skills that had opportunities to occur during classroom visits and included the number of successful responses out of the total opportunities observed and the number of successful intervals out of the total I intervals with opportunities observed for each target skill. Interobserver agreement with a second Board Certified Behavior Analyst[R] (BCBA[R]) was collected on 32% of probe observations and ranged from 92-100% (mean = 99.4%).
Staff training. The paraprofessional was trained in situ via modeling/coaching by the first author approximately 30-60 min per week. Measures of treatment integrity were not permitted, so training to mastery could not be determined. However, daily data were reviewed during every observation and ongoing coaching concentrated on treatment integrity.
Social validity. Three classroom teachers familiar with the intervention completed an anonymous questionnaire indicating on a scale of 1 (strongly disagree) to 5 (strongly agree) whether the intervention successfully improved target-skill independence and whether it could be adequately implemented by trained paraprofessionals. In addition, 10 paraprofessionals and teachers familiar with the intervention were asked to rate how important the identified target skills were for all students on a scale of 0 (not at all important) to 5 (very important).
Figure 1 shows probe data for each target skill during baseline and averaged over the last three probe observations. Independence at all skills increased by a mean of 68.1% (range: 25-100%). Probe data were consistent across multiple observations of the same skill, with three exceptions: transitioning, establishing joint attention, and responding to incongruities. Daily data (not shown) indicated less variability. Fading paraprofessional proximity was successful, as evidenced by continued improvement in probe data for five skills as paraprofessional support decreased while the required frequency of each skill increased to 4-25 times. Token reinforcement for remaining seated was discontinued because all criteria were met and behavior was maintained by natural contingencies. Less progress was seen on responding to peer speech (criterion: nearby, two times), establishing joint attention (criterion: two times), and responding to incongruities (criterion: five times).
Figure 2 depicts paraprofessional-delivered prompts per minute during baseline and three subsequent observations, selected based on video availability. Baseline prompt rate was 2.5 times per minute, and this decreased to less than once per minute.
Social validity data indicated a mean score of 4.6 on whether the system improved skill independence, 4.7 on successful implementation by a paraprofessional, and 4.2 on the importance of independence on the skills targeted.
This case study demonstrates successful extension of an intervention initially developed for students with ASD to a student with Down syndrome. Although our ability to make strong conclusions is reduced due to limitations inherent in case-study designs, these results suggest successful transfer of stimulus control over a majority of the targeted skills from paraprofessional-provided prompts to the naturally occurring cues in the classroom.
This intervention represents use of the SPM: published research on reinforcement, task analyses, stimulus control and previous clinical practice influenced the different components of the intervention and fading practices used in order to enable the paraprofessional to implement them independently with precision. With respect to clinical practice, when the intervention was previously implemented with students with ASD, it was observed that the sound of the first author's heels clicking on the floor seemed to function as a discriminative stimulus for students who were engaging in off-task behaviors to quickly resume on-task behavior. Because the shoe's sound seemed to signal the availability of reinforcement, intentional audible prompts (e.g., tapping the clipboard) were incorporated into the current study.
One aspect particularly relevant for practice is that implementation of the intervention may have altered contingencies for the paraprofessional. Prompting provides opportunities for student interaction, which may be reinforcing to paraprofessionals. From the paraprofessional's perspective, prompting a student may be reinforced by correct responses. From the child's perspective, prompting results in attention, which may function as a reinforcer for waiting for the delivery of a prompt, thereby creating/maintaining prompt dependence. If prompting is reinforcing for the paraprofessional, then inappropriate/ incorrect behaviors may become discriminative stimuli over the paraprofessional's behavior. Thus, both the student and adult may be differentially reinforced in situations in which prompt dependence is demonstrated. In contrast, this intervention provides reinforcement for both the student and paraprofessional when the student independently responds. As paraprofessionals collect data on and provide reinforcement for independence, they are shaped to recognize it. Because independence is paired with student interaction during rein-forcer delivery, independence may become a conditioned reinforcer for the paraprofessional, thereby reducing the likelihood of over-prompting. The depiction of the acquisition data on the daily data sheet illustrates the effect the paraprofessional's actions have on the student's skill acquisition, which also may serve as a reinforcer. Research on paraprofessional behavior is needed to verify these hypotheses.
Social validity data were high even though the intervention was a significant departure from treatment as usual. This indicates potential for using this intervention to support a greater number of students with Down syndrome in inclusive settings, which is an area of need. In addition, it suggests that educators and paraprofessionals may be interested in strategies specifically targeting independence on learning-to- learn skills, particularly given the flexibility of the current intervention. For instance, new skills can be added by simply adding operational definitions and corresponding data sheets with criterion levels, which reduces the need for paraprofessionals to make "judgement calls" on what to do next. In addition, training on the intervention easily generalizes when the paraprofessional is assigned additional students, which may have contributed to the high social validity ratings. This is an area that should be explored in future studies.
Research directly comparing this system with other inclusion models that includes more extensive baseline data collection and application of experimental designs to rule out potential confounds (e.g., maturation, the midyear DTI addition) and allows measures of treatment integrity is warranted. Traditional support was provided to the student for approximately 13 months, but baseline levels for the targeted skills were generally low (0-50%). The dramatic (25-100%) improvement for many targeted skills within only a few months of the current intervention (and prior to introduction of DTI) illustrates the need for systematic approaches to providing support in inclusion settings. Finally, and in-keeping with the SPM, more information regarding the level typically developing peers display the targeted skills is needed to determine goals for each skill level.
Future work can involve the development of an intervention manual with specific prescriptions for training competencies, assessment practices, derivation of initial reinforcement criteria, and treatment integrity measures. The current use of preset reinforcement criteria evolved based on the first author's clinical experience; thus, additional research on more efficient reinforcement schedules for shaping independence is needed. This intervention was based on shaping technology that was originally designed using a SPM to increase deficits in independence for students with autism. The current data illustrate the potential efficacy of this intervention for supporting students with Down syndrome in inclusion settings as well, using resources that are commonly available in public school settings.
Capitol Region Education Council, Endicott College
Caroline C. Redpath
Capitol Region Education Council, River Street Outreach Program,
Karen M. Lionello-DeNolf
Address correspondence to: Audrey Gifford, Endicott College. E-mail: audrey.gifford@OAIautism.com.
Author Note: Audrey Gifford, Endicott College, Optimized Autism Interventions; Caroline C. Redpath, CREC, River Street Outreach Program, Communication Sciences and Disorders, Elms College; Karen M. Lionello-DeNolf, Department of Psychology, Assumption College. Further information about the intervention described in this paper, including example data sheets and other materials, is available upon request.
Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 1, 91-97. doi:0.1901/jaba.1968.1-91
Charania, S., LeBlanc, L. A., Sabanathan, N., Ktaech, I. A., Carr, J. E., & Gunby, K. (2010). Teaching effective hand raising to children with autism during group instruction. Journal of Applied Behavior Analysis, 43, 493-497. doi:10.1901/jaba.2010.43-493
Engevik, L. I., Naess, K-A, B., & Bernsten, L. (2016). Quality of inclusion and related predictors: Teachers' reports of educational provisions offered to students with Down syndrome. Scandinavian Journal of Educational Research, 1-18. doi: 10.1080/0031 3831.2016.1212252
Giangreco, M., Edelman, S., Luiselli, T., & MacFarland, S. (1997). Helping or hovering? Effects of paraprofessional proximity on students with disabilities. Exceptional Children, 64,7-18. doi:10.1177 /001440299706400101
Hundert, J. (2009). Inclusion of students with autism using ABA-based supports in general education. Austin, TX: Pro-Ed Publishing.
Kasari, C, Freeman, S. F. N., Bauminger, N., & Alkin, M. C. (1999). Parental perspectives on inclusion: Effects of autism and Down syndrome. Journal of Autism and Developmental Disorders, 29, 297-305. doi:10.1023/A:1022159302571
Kelley, D. P., Wilder, D. A., Carr, J. E., Rey, C., Green, N., & Lipschultz, J. (2015). Research productivity among practitioners in behavior analysis: Recommendations from the prolific. Behavior Analysis in Practice, 8, 201-206. doi:10.1007/s40617-015-0064-l
Kemp, C., & Carter, M. (2006). Active and passive task related behavior, direction following and the inclusion of students with disabilities. Education and Training in Developmental Disabilities, 41,14-27.
Lorenz, S. (1999). Making inclusion work for children with Down syndrome. Down Syndrome News and Update, 1,175-180. doi:10.3104/practice.149
Presson, A. P., Partyka, G., Jensen, K. M., Devine, O. J., Rasmussen, S. A., McCabe, L. L., & McCabe, E. R. B. (2013). Current estimate of Down syndrome population prevalence in the United States. The Journal of Pediatrics, 163, 1163-1168. doi:10.1016/j.jpeds. 2013.06.013
Saini, V., Miller, S. A., & Fisher, W. W. (2016). Multiple schedules in practical application: Research trends and implications for future investigation. Journal of Applied Behavior Analysis, 49, 421-444. Doi:10.1002/jaba.300
U.S. Office of Special Education Programs (2002). Study of Personnel Needs in Special Education. SPENSE Fact Sheet. Retrieved from www.spense.org
Caption: Figure 1. Probe data for each target skill during baseline and averaged over the last three probe observations
Caption: Figure 2. Paraprofessional-delivered prompts per minute during baseline and three subsequent observations, selected based on video availability
Table 1 Operational Definitions of Targeted Skills Target Skill Operational Definition Remaining seated Being seated in any group lesson with quiet during group hands and mouth for the entire interval. lessons Attending was not required. Other behavior (attempts to leave, use of a special seat, etc.) was scored as "-." Attending to group During lessons, keeping face and eyes oriented lessons toward the speaker addressing the group, or at an item the speaker indicates, with an appropriate (quiet) mouth and body (seated, with hands and feet kept to herself). Instruction directly to the student was not included. Imitation of peers' Repeating peers' motor movements motor movements (e.g., dancing) that were not part of direction following (approximations were acceptable). On-task, fine motor Moving writing implements on paper while holding the paper down for the entire interval, or manipulating a toy, including time a hand moves with a toy or to select another toy/puzzle piece. Responding to peer Reacting to peer speech directed at the speech student by doing what the peer asked or answering with appropriate speech. Following group Carrying out directions from any adult directions addressing the class. Directions to specific students were not included. Transitions Appropriately moving from place to place with the group using only the same cues as typically developing classmates. Responding to Visibly reacting to any unusual event incongruities (i.e., startling, looking toward a sound, etc.). Responding to Visibly reacting to an unusual event, looking incongruities and at peers, and/or speaking to peers about it. establishing joint Opportunities were any incongruity noted by attention with a typical peers within sight or hearing range peer (e.g., a fire truck driving by).
|Printer friendly Cite/link Email Feedback|
|Author:||Gifford, Audrey; Redpath, Caroline C.; Lionello-DeNolf, Karen M.|
|Publication:||Education & Treatment of Children|
|Article Type:||Case study|
|Date:||Aug 1, 2018|
|Previous Article:||Increasing the Wearing of Multiple Prescription Prosthetic Devices.|
|Next Article:||A Comparison of Task Analysis Training Procedures.|