Impact of a sensory activity schedule intervention on cognitive strategy use in autistic students: A school-based pilot study

Abstract

Introduction

Autistic students may experience difficulty performing classroom tasks due to atypical sensory processing and inefficient use of higher-order cognitive strategies. Limited research has investigated the influence of in-class sensory activities to enhance the thinking strategies required for task performance. This study evaluated a classroom-based sensory activity schedule and its impact on cognitive strategy use.

Methods

A quasi-experimental, non-equivalent groups design was used. Students (n = 30, mean age 7.4 years) with atypical sensory processing negatively impacting classroom performance, and their teachers (n = 23), from six autism-specific schools were grouped into intervention (Sensory Activity Schedule and usual teaching) and control (usual teaching only) groups. Students’ cognitive strategy use during the performance of classroom tasks was evaluated at baseline and post-intervention using Perceive, Recall, Plan, Perform Stage Two Cognitive Task Analysis.

Results

Statistical analysis (Mann–Whitney U test) indicated that students who received the Sensory Activity Schedule intervention improved significantly more than control group students in overall cognitive strategy use (Z = –2.32, p = 0.02), and with strategy items involving attention and sensory perception (perceive, Z = –2.26, p = 0.02), and planning and organisation (Plan, Z = –.254, p = 0.01).

Conclusion

The Sensory Activity Schedule may enhance autistic students’ capacity to apply cognitive strategies more effectively during performance of classroom tasks.

Keywords

Sensory processing autism school therapy cognition disability occupational therapy

Introduction

Recognition of the impact of atypical sensory processing on the school performance of autistic students has led to using school-based sensory interventions that are integrated into curriculum activities (Mills and Chapparo, 2017). Specifically, the inclusion of teacher-led, sensory-based activities designed by occupational therapists in the classroom curriculum has gained momentum in special school settings as one way to improve student performance (Mills and Chapparo, 2017; Mills et al., 2016). The assumption underpinning the use of such sensory activities is that readiness to learn can be enhanced by using sensory input to achieve regulatory, functional, and adaptive outcomes (Bodison and Parham, 2018). Despite conceptual and practice support for this approach, research describing student outcomes from classroom-based interventions is scarce (Bodison and Parham, 2018).

Complex sensory, behavioural and cognitive profiles have been reported in autistic students (Robertson and Baron-Cohen, 2017). Occupational therapy interventions may enhance the function of sensory regulation in controlling emotions and the ability to use internal cognitive strategies to think through problems of everyday school activities (Chapparo and Lowe, 2012). Data are emerging to support the use of sensory-based activities to enhance classroom performance in specific tasks. For example, Mills et al. (2016) utilised individually targeted classroom-based sensory activities including deep pressure and movement breaks before and during challenging tasks, with promising results. However, there is little evidence about how these activities may impact the ability of students to use their thinking skills effectively during task performance (Mills and Chapparo, 2017).

School-based learning requires a student to use a range of cognitive skills and apply specific cognitive or thinking strategies to task demands when required (Martinez, 2010). While the term ‘cognition’ broadly refers to mental processes relating to the acquisition, storage, manipulation, and retrieval of information, ‘cognitive strategy use’ refers specifically to how internal thinking techniques are used in a particular way to plan and perform tasks (Chapparo and Ranka, 2011). The process of strategic thinking at school assumes that students can know the goal to be achieved, engage in planning and initiation of behaviour, and monitor and evaluate their own performance in relation to the goal (Chapparo and Ranka, 2011). A student must process salient classroom information ‘in the moment’ in order to use cognition in a functional way (Chapparo, 2010).

A wide range of cognitive profiles are apparent in autistic students, with high levels of cognitive capacity in some (Wallace et al., 2009) and significant cognitive impairment in others (Matson and Goldin, 2013). Some students exhibit constrained information processing, impacting the flexibility required for using thinking strategies across different classroom tasks and performance contexts (Williams et al., 2015).

Definitions of sensory processing interventions for autistic students vary, as do reports of effectiveness and intervention context (Case-Smith et al., 2015). Sensory integration therapy (SIT), which involves intensive, manualised, clinic-based intervention, has generated promising findings (Schoen et al., 2019) but is not always feasible in a school setting (Ashburner et al., 2014). Sensory Activity Schedules were found to be effective for autistic students in a school setting (Mills et al., 2016), provided that students were assessed for atypical sensory processing, intervention was individually tailored to the child, educational needs were considered, and teachers were consulted and trained in its use. Sensory differences in autistic students are singular, requiring individualised, multi-element interventions (Mills et al., 2016). Evidence for the effectiveness of generic sensory-based interventions such as sensory diets or weighted vests is limited (Bodison and Parham, 2018; Lang et al., 2012), with various outcomes targeted, including a reduction in stereotypy and an increase in correct responding. Evidence for the impact of sensory-based intervention on strategic thinking during task performance is scarce. At the time of writing, only one small study was located (Mills and Chapparo, 2017), which addressed the impact of sensory interventions on cognitive strategy use in a school setting.

This study was guided by the following research question: What is the impact of Sensory Activity Schedule (SAS) intervention on cognitive strategy use for autistic students in a classroom setting as determined by the Perceive, Recall, Plan, and Perform (PRPP) Stage Two Cognitive Task Analysis (Chapparo and Ranka, 2011).

Methods

Ethics approval was obtained from The University of Sydney Human Research Ethics Committee (No: 2014/305) and Autism Spectrum Australia’s (Aspect) research approvals committee (No: 1430).

Study design

The study used a quasi-experimental, non-equivalent groups design with baseline and post-intervention tests (Reichardt, 2005). Teachers and their participating students were allocated to one of two groups (intervention and control) using an alternating selection pattern based on order of recruitment of teacher participants into the study. A quasi-experimental design was used instead of a randomised controlled study for the following reasons: ethical (inability to randomly assign interventions in a school situation); difficulty of randomising some students and teachers equally; and a small available sample size. After an initial process of simple 1:1 random allocation of teachers and students, three participating teachers had more than one student in their class participating who were assigned to competing research conditions. As it was impossible for these teachers to act in accordance with requirements of both intervention and control conditions simultaneously, and in consideration of the small sample size, the students participating with the three teachers were allocated the intervention/control condition assigned to their teacher. Non-equivalent groups design has been described as consistent with ethical design, and offering valid contribution in early feasibility studies such as this one, where the aim was to explore intervention efficacy for future research.

The intervention condition consisted of SAS intervention in addition to usual classroom teaching according to the Aspect Comprehensive Approach to Education (ACAE). The control condition was usual classroom teaching according to ACAE only. Data were collected for all students at two data collection points: a baseline measure at the start of the study, prior to any intervention and a post-intervention measure following 7 weeks of the intervention period.

Participants

Two groupings of participants were recruited: autistic students and their classroom teachers from a large, autism-specific, non-government organisation that operates a network of schools in and around Sydney, Australia.

Inclusion criteria included: primary school age (4 to 12 years); attending one of the participating autism-specific schools; in a class with a teacher who agreed to participate in the study; diagnosis of autism as confirmed by school records; and evidence of sensory processing difficulties that negatively affected school performance as determined by classroom teacher and school occupational therapist. Several participating students in the study also had a diagnosis of intellectual disability (ID) in addition to autism. Students were not excluded based on ID as this is a common co-occurring diagnosis.

Teachers were 23 registered primary school teachers in NSW who were teaching classes of four to seven autistic students in one of the participating autism schools. As noted above, not all students in a teacher’s class were recruited for the study: 17 teachers had one participating student, five teachers had two participating students, and one teacher had three participating students. Class sizes were small (4–7 students), with two staff members in each class (teacher and teacher’s aide), allowing for the implementation of an individualised SAS for participating students. Figure 1 illustrates the flow of participant recruitment and group allocation. Eight students were excluded from analysis as their data were misplaced by the schools and not returned for analysis.

Figure 1.

Participant flow chart for study.

Study procedure

Following ethical approval, schools were approached and school principals invited teachers within their school to participate. Teachers who volunteered and provided written informed consent were asked to select between one and three students in their class who they, in consultation with the school occupational therapist, identified as having atypical sensory processing that affected school participation. Teachers had some knowledge of sensory processing prior to the commencement of the study as, following school protocol, they were required to participate in relevant training provided by school occupational therapists and comment on sensory processing in all students’ individual education plans. Written informed consent was obtained from parents of child participants. While the children were unable to provide formal written or verbal consent, all children were given the opportunity to signal their intent to participate in sensory activities using gestural and/or visual modes of consent.

A sensory processing assessment of all student participants was conducted to confirm the presence of atypical sensory processing and describe the impact this may have had on students’ classroom occupational performance. This assessment comprised the Short Sensory Profile 2 (SSP2) and Sensory Profile School Companion (SPSC2) (Dunn, 2014). Both questionnaires use a five-point rating scale where parents and teachers identify how often their child/student responds to different sensory situations and sensory inputs. Occupational therapy evaluation using non-standardised methods including school observations and interviews with parents and teachers complemented the use of the SSP2 and SPSC2 in identifying the unique sensory processing needs and responses of participating students.

Teachers and their corresponding students were then allocated to either the intervention or control conditions (as explained above).

Baseline data were collected for all students prior to the 7-week intervention period, which occurred during either term 2, term 3, or term 4 of the Australian school year depending on when teachers were recruited. Post-intervention data were collected immediately following the 7-week intervention period. An Australian school term is usually 10 weeks and this timeframe for intervention fit the timeframe of conducting research within a school. Student data from the intervention group were only included in the analysis if students completed at least 80% of intervention sessions. Figure 1 shows students who were excluded from analysis.

Intervention conditions

Control condition: usual classroom teaching according to the ACAE

The ACAE is used to guide teaching practices within the autism schools where the research was conducted and was used as the usual classroom teaching control condition in this study. The ACAE is a multi-element, evidence-based framework comprising structured teaching, positive behaviour support, family involvement, individualised planning, environmental supports, transition and inclusion, mental health and wellbeing, and the presence of a multidisciplinary team in schools (Aspect, 2015).

Intervention condition: SAS intervention in addition to usual classroom teaching according to ACAE

Teachers allocated to the intervention condition implemented SAS intervention with their students in addition to usual classroom teaching according to the ACAE. SAS is a five-element framework originally developed by Mora and Chapparo (2011) that guides sensory-based intervention in a school setting for autistic students and has been previously trialled with promising results (Mills and Chapparo, 2017; Mills et al., 2016). The five elements are discussed below. First, participating students must demonstrate evidence of need. Atypical sensory processing that negatively impacted classroom participation and performance was confirmed by an occupational therapist upon student recruitment into the study. Second, sensory activities used are selected in line with students’ unique sensory needs, preferences, and the resources available. In this study, activities took approximately 10 minutes and were facilitated by classroom teachers who received support and training from an occupational therapist. Students were given visual schedules to facilitate their participation and independence, and activities took place prior to participation in required class activities. See Table 1 for an example of sensory issues and corresponding activities. Third, SAS is task specific and activities are used directly prior to set work tasks such as reading, fine motor and writing tasks, and puzzles (Mills et al., 2016). Sensory activities were implemented at particular times to assist a student’s readiness to engage in specific tasks, rather than to improve their general behaviour throughout the school day. Fourth, work tasks and sensory activities used are teacher directed, as part of the school curriculum and in consultation with school occupational therapists (Boshoff and Stewart, 2013). The fifth element of SAS addresses the sensory, physical, social, and cognitive contextual fit of the activities used in the classroom (Chapparo and Lowe, 2012), considering the needs of other students and staff, to avoid implementation of the SAS impacting others. In this study, it was acknowledged that different classrooms and schools had different contexts, activities, resources, and people that were considered during the design of the SAS intervention for each student.

Table 1.

Summary table of sensory issues and corresponding activities (adapted from Mills et al., 2016).

Presenting sensory issue	Example of SAS activities
• Regularly rocking and bouncing on chair during class time and constantly seeking movement throughout the day, which was interrupting participation in school activities	• Before desk work commences, climb and crash using A frame and crash mat, bouncing on therapy ball for 3 minutes prior to desk work, shoulder squeezes from teacher
• Some sensory seeking and sensitivity observed, overstimulated at times, avoidance behaviours, hand flapping, needed more help maintaining optimal arousal level for classroom activities	• Before morning table work commences, jumping on a mini trampoline, lying flat on a scooter board and self-propelling forwards with arms, tactile fidgets such as putty available throughout group times, structured gross motor programme with rest of class
• Moves and climbs a lot during the school day. Has more difficulties in noisy environments, is quite distractible, and seeks touch input on different surfaces, which interferes with her performance in classroom activities	• Offered choice from fidget box before desk work (sensory activities to be presented on 1-2-3 visual board), 5 minutes on a trampoline, activity songs on a therapy ball to provide proprioceptive input, shoulder squeezes from teacher, offered choice of 5-minute break in dark tent in classroom
• Becomes overwhelmed frequently within the classroom and is wary of new sensory stimuli. Seeks movement throughout the day in response to being overwhelmed; this affects participation in and completion of classroom group and individual activities	• Chewing and deep pressure around the jaw and cheek, lie on tummy and roll slowly over therapy ball, taking weight on hands, deep touch pressure on shoulders provided by trusted adult, access to headphones, keep sensory input predictable, try to avoid alerting activities such as spinning or bouncing on therapy ball

SAS: Sensory Activity Schedule.

Outcome measure: Perceive Recall Plan Perform Stage Two Cognitive Task Analysis

The PRPP Stage Two Cognitive Task Analysis (Chapparo and Ranka, 2011) is a standardised, criterion-referenced assessment that measures how well cognitive strategies are used in context to meet particular task demands. The structural and conceptual background to the PRPP Stage Two utilises information processing theory, which proposes four main cognitive processing domains: sensory processing and attention (perceive); memory (recall); planning (plan); and performance monitoring (perform). When children engage in classroom tasks, they use their cognition strategically, applying ways of thinking (cognitive strategies) from all four processing domains for each activity.

The PRPP Stage Two measure contains 35 observable cognitive strategy items, each representing a generic cognitive strategy within four processing quadrants and which is applied during task performance. Within the perceive quadrant, strategies for registering and discriminating the sensory information required for task performance are observed, as well as those resulting in the capacity to shift and maintain focus of attention to critical sensory information. The recall quadrant involves assessment of the effectiveness of thinking strategies for remembering facts, schemes, and procedures for task performance. The plan quadrant comprises assessment of thinking strategies supporting the capacity to know the goal of the task, to choose and sequence its parts, and to evaluate problems and performance outcomes. Within the perform quadrant, strategies for starting, stopping, timing, and controlling task performance are observed (Chapparo and Ranka, 2011).

The task and the criteria for successful task performance are set by known variables in the person’s ecology. In this study, the task and how well it should be performed was determined by the teacher, who was cognisant of the particular capacity of the student and the curriculum objectives of the class. The purpose of the assessment is not to make judgements about level of cognition per se, or compare performance of a common task among students, but to determine how well students are able to apply a generic set of cognitive strategies effectively during the performance of expected school tasks.

During observation of the student’s task performance, each cognitive strategy is assessed and rated according to whether it is used effectively for the task performance required (Chapparo and Ranka, 2011). A score of three represents error-free performance with no assistance or prompting required. A score of two is given when some prompting may be required, or slow processing occurs, but error-free performance is observed. A score of one denotes when excessive prompting is required and/or when deficit in this item has an impact on overall successful task performance.

Raw scores from each of the 35 processing strategies are summed and then converted into a percentage score out of the maximum possible score (105) to give a Total PRPP Stage Two raw score. This total score represents overall effectiveness of cognitive strategy use during the task. PRPP quadrant scores are generated for each of the four PRPP quadrants (perceive, recall, plan, and perform) by summing the cognitive strategy items in each specific quadrant and converting them to a percentage score.

Data collection procedure

Data were collected as part of a larger study evaluating the impact of SAS on the classroom participation and performance of autistic students. Participating students in control and intervention groups were filmed by school staff completing functional tasks in the classroom at baseline and then again post-intervention, 7 weeks later. Each participating student was filmed completing between one and four tasks, and videos ranged from 2 to 10 minutes depending on task duration. Tasks included classroom work such as reading, writing, fine motor, and matching tasks, and group and playground participation. Tasks were set by the classroom teacher and varied depending on the student’s age and level of ability.

Following video data collection, videos were placed in random order and scored by researchers who were occupational therapists experienced in using and scoring PRPP Stage Two Cognitive Task Analysis. To control for bias, raters were blinded to participant group allocation and timing of data collection (baseline or post-intervention). Data were subjected to test–retest and interrater reliability testing with high agreement between raters/ratings. Participating teachers were blinded to PRPP administration and scoring procedures.

Data analysis

The focus of analysis for this pilot study was to compare any changes in performance between students who received SAS intervention in addition to usual classroom teaching and students who received usual classroom teaching only. Data were analysed using IBM SPSS Version 23. Non-parametric statistical procedures were used as data collected were not normally distributed and therefore did not meet the assumptions required for parametric tests. A Mann–Whitney U test was conducted on baseline scores for PRPP Stage Two total and quadrant scores to identify if there were statistically significant differences in performance at baseline between the two groups, since differences at baseline can impact on the interpretation of overall results.

Since a parametric two-way, repeated measures ANOVA (two groups, two time points) could not be used, change scores were analysed to investigate any group by time interaction effect. Change scores were calculated for each student by subtracting their baseline score from their post-intervention score to represent improvement (positive change score) or decline (negative change score) in performance following the intervention period. A Mann–Whitney U test was then used to determine any differences between the two groups. Effect size correlation coefficients were calculated for significant Mann–Whitney U test results using the recommended effect size conversion calculation where the Mann–Whitney U test statistic (Z) is divided by the square root of the number of observations (Pallant, 2007). For the non-parametric Mann–Whitney U test used, power was calculated to be 80% with a sample size of 30 (group ratio of 1.31), effect size greater than 1, and a two-sided type-I error rate (alpha) of 0.05. In this study, for clinically significant results to be achieved, a larger effect of the intervention was required; therefore, the larger effect size to achieve 80% power was determined to be suitable.

Results

Characteristics of the 30 student participants (27 male, 3 female; mean age 7.4 years) are shown in Table 2. No significant differences were observed in SPSC-2 or SSP-2 scores between the two groups prior to the study. Scores showed that participants were, on average, rated as ‘more than others’ or ‘much more than others’ for each of the four areas of sensory processing as described in the SSP2 and the SPSC2 (seeker, avoider, sensor, bystander), indicating potential sensory difficulties that could impact classroom performance. The presence of intellectual disability could only be confirmed from the school records of 18 out of the 30 students. Eight of the intervention group and 10 of the control group students had a diagnosed ID in the mild or moderate range. For the remaining 12 children, ID data were not available. These participants may or may not have had ID.

Table 2.

Participant characteristics.

	SAS group	Usual teaching group	ANOVA for between group differences
	Mean (SD)	Mean (SD)	F	p
N	13	17	NA	NA
Gender	12 M, 1F	15 M, 2 F	NA	NA
Age	7.08 (1.91)	7.6 (2.21)	0.511	0.48
SSP-2 scores
Seeker	22.09 (5.28)	22.25 (5.41)	0.01	0.94
Avoider	28.64 (5.46)	28.50 (5.51)	<0.01	0.95
Sensor	35.64 (6.07)	36.06 (5.66)	0.04	0.85
Bystander	18.55 (7.62)	20.13 (5.02)	0.42	0.52
SPSC-2 scores
Seeker	25.69 (6.84)	22.81 (5.01)	1.71	0.20
Avoider	33.62 (10.19)	31.82 (8.97)	0.26	0.62
Sensor	30.38 (7.03)	28.19 (9.07)	0.51	0.48
Bystander	40.92 (11.41)	38.12 (10.56)	0.47	0.50

SAS: Sensory Activity Schedule; M: male; F: female; SSP-2: Short Sensory Profile 2 (completed by parents) (Dunn, 2014); SPSC-2: Sensory Profile School Companion 2 (completed by teachers) (Dunn, 2014)

As recorded in Table 3, there were no statistically significant differences between the two groups at baseline on all PRPP Stage Two scores, with the exception of the perceive quadrant. Students in the control group (who received usual classroom teaching only) started with a statistically significant higher score for perceive (median 83.33%) than students in the intervention group (median 75%).

Table 3.

Students’ cognitive performance as measured by PRPP Stage Two, with results of the Mann–Whitney U test showing differences between groups for baseline scores and change scores.

PRPP Stage Two section	Group	n	Median (%)	Mean rank (SoR)	Z	p-value	Effect size(r_pbs)
Baseline scores
Total PRPP Stage Two	SAS	13	70.48	12.54 (163)	–1.61	0.11	N/A
Total PRPP Stage Two	UCT	17	76.19	17.76 (302)	–1.61	0.11	N/A
Perceive quadrant	SAS	13	75.00	11.27 (146.5)	–2.31	0.02*	–0.42
Perceive quadrant	UCT	17	83.33	18.74 (318.5)	–2.31	0.02*	–0.42
Recall quadrant	SAS	13	75.93	13.08 (170)	–1.32	0.19	N/A
Recall quadrant	UCT	17	77.78	17.35 (295)	–1.32	0.19	N/A
Plan quadrant	SAS	13	55.56	12.58 (163.5)	–1.60	0.11	N/A
Plan quadrant	UCT	17	62.96	17.74 (301.5)	–1.60	0.11	N/A
Perform quadrant	SAS	13	79.17	12.69 (165)	–1.53	0.13	N/A
Perform quadrant	UCT	17	91.67	17.65 (300)	–1.53	0.13	N/A
Change scores
Total PRPP Stage Two	SAS	13	12.11	19.77 (257)	–2.32	0.02*	–0.42
Total PRPP Stage Two	UCT	17	3.75	12.24 (208)	–2.32	0.02*	–0.42
Perceive quadrant	SAS	13	11.57	19.65 (255.5)	–2.26	0.02*	–0.41
Perceive quadrant	UCT	17	5.32	12.32 (209.5)	–2.26	0.02*	–0.41
Recall quadrant	SAS	13	12.35	18.58 (241.5)	–1.68	0.09	N/A
Recall quadrant	UCT	17	0.00	13.15 (223.5)	–1.68	0.09	N/A
Plan quadrant	SAS	13	18.52	20.15 (262)	–2.54	0.01*	–0.46
Plan quadrant	UCT	17	3.70	11.94 (203)	–2.54	0.01*	–0.46
Perform quadrant	SAS	13	12.50	18.04 (234.5)	–1.38	0.17	N/A
Perform quadrant	UCT	17	2.08	13.56 (239.5)	–1.38	0.17	N/A

PRPP: Perceive, Recall, Plan, Perform System of Task Analysis; SAS: Sensory Activity Schedule intervention in addition to usual classroom teaching; UCT: Usual Classroom Teaching; SoR: sum of ranks; r_pbs: effect size – non-parametric point biserial correlation

* = Statistically significant (p<.05)

Table 3 also shows the results of changes to students’ cognitive strategy use performance as measured by PRPP Stage Two Change Scores. Median scores are shown for PRPP Stage Two total, and the four PRPP quadrant sections (perceive, recall, plan, and perform) as this is the accepted convention in non-parametric testing (Gravetter and Wallnau, 2013). Results from Mann–Whitney U test analysis comparing changes in performance between the two groups are shown (mean ranks, z-scores, p-values), as well as effect sizes for statistically significant results.

PRPP Stage Two total

Analysis of students’ overall cognitive performance as measured by PRPP Stage Two total scores revealed that while students in both groups improved in their classroom cognitive strategy use, students in the SAS group improved more (median change score 12.11%) than students who received usual classroom teaching only (median change score 3.75%). This result was statistically significant (p = .02), with a medium to large effect size (r_pbs = –0.42) as shown in Table 3.

PRPP Stage Two quadrant sections

Comparison of change scores between the two groups in each of the PRPP quadrants yielded a statistically significant result in the perceive quadrant (p = .02) with a medium to large effect size (r_pbs = –0.41), and for the plan quadrant (p = 0.01) with a large effect size (r_pbs = –0.46). Results were not statistically significant for changes noted in the recall and perform quadrants.

Students who were in the SAS group improved significantly more from baseline to post-intervention (larger median change scores) in both the perceive and plan quadrants in comparison with students in the usual classroom teaching group, where performance remained relatively stable (see Table 3).

Discussion

This study used a quasi-experimental group comparison design to determine the impact of SAS intervention on cognitive strategy use during the classroom activities of autistic students who attended an autism-specific school. Results indicated that students who received SAS in addition to usual teaching according to the ACAE demonstrated significantly more improvement than students who received usual teaching only, in overall cognitive strategy use during school activities, as indicated by PRPP Stage Two total score. In particular, cognitive strategy items targeting attention, sensory perception, and planning (perceive and plan quadrant scores) showed significantly more improvement in the SAS group. A number of key findings are discussed.

The first finding was that use of the SAS had an influence on the general ability to use overall cognitive strategies during student performance of classroom tasks as measured by PRPP Stage Two Cognitive Task Analysis total score. Cognitive strategies are internal thinking techniques that involve making a goal, planning, monitoring, and carrying out task behaviour (Chapparo, 2010). This finding supported an earlier study by Mills and Chapparo (2017), who utilised SAS intervention and a single system research design with seven students and noted significant improvements in cognitive strategy use for some autistic students. In a study by Edgington et al. (2016), similarly promising results were reported following an 8-week cognitive behaviour therapy programme designed to target sensory issues experienced by autistic adolescents. Although that intervention included relaxation and a therapy ball to address body coping strategies, no further detail was provided about the nature of sensory activities used and their impact on specific school performance. Edgington et al.’s study was different to the present study in that it focused on facilitating a conscious awareness of sensory processing difficulties, and excluded students with co-occurring ID. No other published research had been found that indicated similar findings for autistic children. This is the first study to measure cognitive strategy use in relation to an intervention designed to target sensory processing using a group-based design.

Second, it was found that specific sensory input, comprising SAS activities providing chiefly deep touch and proprioception, appeared to have a regulatory influence on attention (noticing, focusing, and maintaining to suit the task) as measured by the perceive quadrant items and behavioural output (calibrating and refining responses to suit the task). Neural circuitry for proprioception and deep touch input is thought to contribute to level of arousal through regulation of parasympathetic nervous system (PNS) (calm alert state) and sympathetic nervous system (SNS) (flight, fight state) autonomic activity (Schaaf et al., 2015). Research has demonstrated that some autistic children may have difficulties disengaging their SNS, leading to lower PNS activation, more SNS activation, and, consequently, less time in a calm alert state throughout the day (Schaaf et al., 2015). It is possible that engagement in SAS before performing challenging tasks contributed to improved regulation of arousal as part of a pre-cognitive ‘getting ready’ stage in performance. While research has demonstrated such calming benefits from proprioceptive inputs (Reynolds et al., 2015; Schaaf, et al., 2015), no research has demonstrated the same impact on specific classroom learning activity.

Third, particular improvement in the use of cognitive strategies in the perceive quadrant was observed, indicating that SAS may have enabled more efficient cognitive strategy use related to the attending and sensory processing required for task performance. A recent systematic review of research that used randomised controlled trial methods noted similar improvements in attending and sensory processing as a result of clinic-based sensory integration therapy (Schoen et al., 2019). While these and the present studies differed in intervention type (SIT vs SAS) and context (school-based vs clinic-based), the notion that sensory input may be associated with improved attention and processing of sensory inputs for use is supported.

The fourth finding from the study was a significant improvement in the plan quadrant, which is assumed to reflect executive functioning and includes cognitive flexibility, response inhibition, and higher-order thinking (Ranka, 2011). An unexpected finding from the study was that the greatest improvement in cognitive strategy use was observed in the plan quadrant, which is assumed to reflect executive functioning and is a noted area of difficulty for autistic students (Van den Bergh et al., 2014). This indicates that SAS may have had a positive impact on students’ capacity to organise their thinking in a way that is required for contextual task performance. Aside from Mills and Chapparo (2017), no published literature was found that addressed the impact of sensory interventions on planning and executive function in autistic students.

There are several possible interpretations for the positive results observed in the plan quadrant. First, it is possible that SAS implementation enabled the self-regulation required for the internal thinking strategies to organise and carry out task performance. It is also possible that SAS facilitated a stable platform of central nervous system regulation for the specific amount of time needed to do the task. Second, the state regulation obtained from SAS activity immediately prior to task performance may have facilitated students to access and use higher-order executive functions already present ‘in the moment’ they were needed. Such executive functions are thought to originate from pre-frontal cortex functions (Dunn and Kronenberger, 2013), which assist in attending to sensory information, shifting attention, regulating emotions, and generating mental images in a way that is specifically required for a task. When sensory information is perceived and processed, representations are made from internal and external objects within the environment, which are necessary for working memory, encoding and retrieval, attention, and voluntary motor responses (Dunn and Kronenberger, 2013). Higher-order cognitive processes such as thinking and planning emerge from registering, organising, and prioritising basic sensory and motor inputs relative to particular needs for performance ‘in the moment’ (Dunn and Kronenberger, 2013). In this study, sensory activities were used immediately prior to specific class tasks, rather than for ‘general’ use. SAS may have provided the ‘just right’ amount of sensory and motor input at the exact moment in performance time for optimum use of higher-order executive functions measured by items in the plan quadrant. This hypothesis is supported by research that has investigated potential links between executive functions and sensory processing and suggests that inhibitory control and attention at an executive level play an important role in the regulation of sensory processing as a latent factor (Romero-Ayuso et al., 2018).

Limitations

There are several limitations of this study. This study was the first of its kind to evaluate the impact of SAS intervention in relation to cognitive strategy use in a school setting. As this was a pilot study, there are a number of considerations for the design of future studies. This study utilised two multi-element intervention frameworks (SAS and ACAE) that are individualised to student needs rather than manualised. Difficulties have been highlighted with designing research projects around multi-element interventions with individualised components (Donald, 2018) and manualised intervention protocols may not always fit the individualised needs of autistic children (Mesibov and Shea, 2011). SAS is a teacher-directed intervention and there may have been a teacher effect in the implementation as teachers can have an impact on the delivery of classroom interventions (Kelly and Barnes-Holmes, 2013). Full diagnostic details for intellectual disability in autistic students was not available and this presents a potential confounding variable that could not be controlled for in measuring cognitive strategy use. Future studies of cognitive strategy use would need to take intellectual disability diagnostic information into account upon analysis.

Caution should be taken when interpreting statistical results as this study had a small sample, elevating the possibility of type II error. Given the small sample size and the individuality of autistic students, it is possible that a different result may have been observed with a different group of students. Group allocation in this study occurred by teacher rather than by student and this was necessary to prevent students in the same class being allocated to different intervention conditions. This may have created a nesting effect, which could be addressed using alternative statistical analysis, revealing different results. It is also possible that the significant results were affected by regression to the mean, which could not be corrected during non-parametric statistical analysis. Replicating these findings in a further study utilising a more rigorous methodology will clarify this issue.

Conclusion

The current study was focused on autistic students who attended autism-specific schools and had sensory difficulties that were negatively impacting their classroom performance. This research study used PRPP Stage Two Cognitive Task Analysis to demonstrate that when suitable students are given the opportunity to access sensory opportunities and adjustments within their own context, it can have a positive impact on classroom cognitive strategy use during participation in classroom occupations. These findings add to the emerging evidence base for the use of sensory strategies to support autistic students.

Key points for occupational therapy

Autistic students may have improved overall cognitive strategy use when given access to appropriate sensory opportunities.

Targeted opportunities to meet sensory needs can result in improved cognitive planning strategies in the classroom.

PRPP is a suitable outcome measure for autistic students in schools.

Footnotes

Acknowledgements

The authors wish to thank the participating teachers, and the children and their families.

Research ethics

This study gained ethical approval in 2014 from The University of Sydney Human Research Ethic Committee (No: 2014/305) and Autism Spectrum Australia’s (Aspect’s) Research Approvals Committee (No: 1430).

Consent

Written informed consent was obtained from participating teachers and from parents of participating students.

Declaration of conflicting interests

The first author was employed by Autism Spectrum Australia (Aspect) at the time the research took place. This research was conducted as part of the first author’s PhD studies. Neither Aspect nor the authors will benefit financially from favourable results. Authors have no other conflicts to declare.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: A small amount of funding was received for this study from Autism Spectrum Australia (Aspect).

Contributorship

This research was conducted as part of CM’s PhD. CM was involved in the conceptualisation and design of the study, recruitment, data collection, data analysis, and final write up.

CC was CM’s primary PhD supervisor. CC was involved in the conceptualisation and design of the study, data analysis, and final write up.

JH was CM’s associate PhD supervisor. JH was involved in research design, data analysis, and final write up.

ORCID iDs

Caroline J Mills

Joanne Hinitt

References

Ashburner

Rodger

Ziviani

, et al. (2014) Optimizing participation of children with autism spectrum disorder experiencing sensory challenges: A clinical reasoning framework. Canadian Journal of Occupational Therapy 81(1): 29–38.

Aspect (2015) ACAE Manual Introduction. Available at: https://www.autismspectrum.org.au/about-aspect/who-we-are/aspect-comprehensive-approach (accessed 1st March 2018)

Bodison

Parham

(2018) Specific sensory techniques and sensory environmental modifications for children and youth with sensory integration difficulties: A systematic review. American Journal of Occupational Therapy 72(1): 7201190040.

Boshoff

Stewart

(2013) Key principles for confronting the challenges of collaboration in educational settings. Australian Occupational Therapy Journal 60(2): 144–147.

Case-Smith

Weaver

Fristad

(2015) A systematic review of sensory processing interventions for children with autism spectrum disorders. Autism 19(2): 133–148.

Chapparo

(2010) Perceive, recall, plan and perform: Occupational centred task-analysis and intervention system. In: Rodger

(ed.) Occupation Centred Practice with Children: A Practical Guide for Occupational Therapists. Hoboken, NJ: Wiley-Blackwell, pp.183–202.

Chapparo

Lowe

(2012) School: Participating in more than just the classroom. In: Lane

Bundy

(eds) Kids Can Be Kids: A Childhood Occupations Approach. Philadelphia, PA: FA Davis, pp.83–101.

Chapparo

Ranka

(2011) The PRPP System of Task Analysis: User’s Training Manual – Research Edition. Sydney, NSW: Occupational Performance Network: School of Occupation & Leisure Sciences, The University of Sydney.

Donald

(2018) A brief summary of pilot and feasibility studies: Exploring terminology, aims, and methods. European Journal of Integrative Medicine 24: 65–70.

10.

Dunn

Kronenberger

(2013) Attention deficit. In: Dulac

Lassonde

Sarnat

(eds) Pediatric Neurology: Handbook of Clinical Neurology (3rd Series). Amsterdam, The Netherlands: Elsevier, pp.257–261.

11.

Dunn

(2014) Sensory Profile 2. Bloomington, MN: Pearson Psychcorp.

12.

Edgington

Hill

Pellicano

(2016) The design and implementation of a CBT-based intervention for sensory processing difficulties in adolescents on the autism spectrum. Research in Developmental Disabilities 59: 221–233.

13.

Gravetter

Wallnau

(2013) Statistics for the Behavioral Sciences, 9th ed. Belmont, CA: Wadsworth, Cengage Learning.

14.

Kelly

Barnes-Holmes

(2013) Implicit attitudes towards children with autism versus normally developing children as predictors of professional burnout and psychopathology. Research in Developmental Disabilities 34(1): 17–28.

15.

Lang

O’Reilly

Healy

, et al. (2012) Sensory integration therapy for autism spectrum disorders: A systematic review. Research in Autism Spectrum Disorders 6(3): 1004–1018.

16.

Martinez

(2010) Learning and Cognition: The Design of the Mind. Upper Saddle River, NJ: Merrill.

17.

Matson

Goldin

(2013) Comorbidity and autism: Trends, topics and future directions. Research in Autism Spectrum Disorders 7(10): 1228–1233.

18.

Mesibov

Shea

(2011) Evidence-based practices and autism. Autism 15(1): 114–133.

19.

Mills

Chapparo

(2017) Use of perceive, recall, plan, perform (PRPP) stage two cognitive task analysis for students with autism and intellectual disability: The impact of a sensory activity schedule. Journal of Occupational Therapy, Schools & Early Intervention 10(3): 232–253.

20.

Mills

Chapparo

Hinitt

(2016) The impact of an in-class sensory activity schedule on task performance of children with autism and intellectual disability: A pilot study. British Journal of Occupational Therapy 79(9): 530–539.

21.

Mora

Chapparo

(2011) Using a sensory diet to mediate behaviours of concern and to increase children’s participation in daily activities. In: Dossetor

White

Whatson

(eds) Mental Health of Children and Adolescents with Intellectual and 283 Developmental Disabilities: A Framework for Professional Practice. Melbourne, Vic: IP Communications Pty Ltd, pp.135–151.

22.

Pallant

(2007) SPSS Survival Manual: A Step by Step Guide to Data Analysis Using SPSS for Windows, 3rd ed. Berkshire: Open University Press.

23.

Ranka

(2011) The plan quadrant. In: Chapparo

Ranka

(eds) User’s Training Manual – Research Edition. Sydney, NSW: Occupational Performance Network, School of Occupation & Leisure Sciences, The University of Sydney.

24.

Reichardt CS (2005) Nonequivalent group design. In Everitt, BS and Howell DC (eds), Encyclopedia of Statistics in Behavioral Science (Volume 3). Chichester, UK: John Wiley and Sons Ltd. pp.1410–1411.

25.

Reynolds

Lane

Mullen

(2015) Effects of deep pressure stimulation on physiological arousal. American Journal of Occupational Therapy 69(3): 6903350010p1-5.

26.

Robertson

Baron-Cohen

(2017) Sensory perception in autism. Nature Reviews in Neuroscience 18(11): 671–684.

27.

Romero-Ayuso

Jorquera-Cabrera

Segura-Fragoso

, et al. (2018) Assessment of sensory processing and executive functions in childhood: Development, reliability, and validity of the EPYFEI. Frontiers in Pediatrics 6: 71, DOI: https://doi.org/10.3389/fped.2018.00071.

28.

Schaaf

Benevides

Leiby

, et al. (2015) Autonomic dysregulation during sensory stimulation in children with autism spectrum disorder. Journal of Autism and Developmental Disorder 45(2): 461–472.

29.

Schoen

Lane

Mailloux

, et al. (2019) A systematic review of Ayres sensory integration intervention for children with autism. Autism Research 12(1): 6–19.

30.

Van den Bergh

SFWM

Scheeren

Begeer

, et al. (2014) Age-related differences of executive functioning problems in everyday life of children and adolescents in the autism spectrum. Journal of Autism and Developmental Disorders 44(8): 1959–1971.

31.

Wallace

Anderson

Happé

(2009) Brief report: Information processing speed is intact in autism but not correlated with measured intelligence. Journal of Autism and Developmental Disorders 39(5): 809–814.

32.

Williams

Minshew

Goldstein

(2015) Further understanding of complex information processing in verbal adolescents and adults with autism spectrum disorders. Autism 19(7): 859–867.