Abstract
The results of this study suggest that differences in hyporeactivity for children with autism may play a bigger role in classroom behavior and learning than previous literature has suggested.
Sensory reactivity differences are a core diagnostic criterion for autism spectrum conditions (American Psychiatric Association, 2013). These differences are found across all sensory domains, including the tactile, visual, and auditory domains (Dunn, 1997). Differences fall into three subtypes: (1) hyperreactivity (a strong response to stimuli, such as finding noise painful); (2) hyporeactivity (an underresponsiveness, such as not noticing the cold); and (3) sensory seeking (fascination with or need for a certain input; Ben-Sasson et al., 2009). All subtypes may be found within the same individual (MacLennan et al., 2022). Although autistic individuals describe finding enjoyment or comfort in some of their sensory reactivity differences (MacLennan et al., 2020), these differences may be challenging in a range of areas (Dellapiazza et al., 2018).
Autistic students are at higher risk for underachieving academically (Mallory & Keehn, 2021), and research looking into the role of sensory reactivity differences in this underachievement is increasing. A correlation between sensory reactivity differences and poorer academic performance has been demonstrated among autistic students in mainstream classrooms (Ashburner et al., 2008). Dunn (1997) theoretically demonstrated why sensory differences affect behavior and learning, and Jones et al. (2020) described how teachers and parents see sensory differences causing distress and distraction and reducing classroom participation. Sensory differences have also been shown to reduce attention (Mallory & Keehn, 2021).
Investigating the role each sensory subtype plays in learning is vital because different support methods are needed to aid with each. Using parent reports, Liss et al. (2006) found a relationship between hyperreactivity and overfocus of attention and between hyporeactivity and lower adaptive functioning and communication skills. However, this link was with general adaptive functioning skills rather than classroom behavior specifically. Ashburner et al. (2008) found that autistic students with increased hyporeactivity and sensory-seeking differences (measured with the Short Sensory Profile [SSP]) were at increased risk of inattention to tasks in the classroom, a behavior that impedes learning. Using parent and teacher reports, Green et al. (2016) found sensory reactivity differences were linked to increased emotional, but not behavioral, challenges.
The limited previous literature in the field relies on parent and teacher reports. Although these are important sources of information, they are vulnerable to recollection bias and parent–teacher discrepancies (Jordan et al., 2019). More research using direct observational methods is needed. Furthermore, previous work has focused on students in mainstream schools and has not incorporated autistic students in special education settings.
This study investigated whether there is a link between sensory reactivity differences and classroom behaviors of autistic children in special education settings, using objective direct observational measures.
Method
Participants and Procedures
Fifty-three students (9 female, 44 male) ages 5–18 yr (M = 10.53, SD = 3.98) with a clinical diagnosis of autism participated. Participants were recruited from two special education schools in which all students have an Education Health and Care Plan and require a high level of support, reflected in high staff:student ratios ranging from 1:1 to 1:3. Informed consent was gained from parents via electronic forms. Participants were unable to provide written or verbal consent, so they were continuously monitored for signs of distress. If they appeared distressed or if school staff stated that they were displaying anxiety, the assessment stopped. Research was conducted in participants’ usual classroom, with behavior observations completed during a routine seated table work activity.
Measures
Sensory Assessment for Neurodevelopmental Differences
The Sensory Assessment for Neurodevelopmental Differences (SAND) consists of direct observation of an individual’s response to sensory stimuli and an accompanying parent interview (Siper et al., 2017). In line with study aims to use a direct observation method, only the observational aspect of the SAND was used. During the observation, individuals’ observable reactions to stimuli are scored; a verbal response is not required, which makes the tool suitable for individuals with minimal spoken language. Structured observation using standardized manipulatives is completed and lasts approximately 15 min. The SAND was designed specifically to capture sensory reactivity differences among autistic people; sensory hyperreactivity (adverse reaction to noisy toys or flashing lights), hyporeactivity (not noticing being touched by a cold pack or sudden unexpected noise), and seeking behavior (looking very closely at a spinning wheel or bringing a toy close to their ear) are examined across visual, tactile, and auditory domains. If a difference is observed, a score of 1 is given. If no difference is observed, a score of 0 is given. A severity rating is given for the hyperreactivity, hyporeactivity, and seeking categories in each domain (1 = mild differences; 2 = moderate–severe differences, such as when a reaction is shown multiple times). The number of differences observed and severity ratings are combined to give an overall score (out of 15) for each domain, with total scores ranging from 0 to 45. Higher scores represent higher presence of sensory reactivity differences. The SAND has high internal consistency (Cronbach’s α = 0.90) and strong interrater and test–retest reliability (>0.8 and 0.8, respectively; p < .001; Siper et al., 2017). Researchers were trained on the SAND by one of the codevelopers of the tool.
Behavior Assessment for Children–Second Edition Student Observation System
The Behavior Assessment for Children–Second Edition Student Observation System (BASC–SOS; Reynolds & Kamphaus, 2004) assesses 13 adaptive behaviors that facilitate learning (e.g., following instructions, completing an activity, interacting with staff) and 58 problem behaviors that impede learning (e.g., aggression, self-injury, inattention). For this study, the BASC–SOS language was changed to be less stigmatizing; behaviors were referred to as behaviors that impede or facilitate learning. It is important to recognize that autistic students may learn in different ways than neurotypical students; however, in the context of a routine table-based learning activity (during which the BASC–SOS was completed), the behaviors assessed would either impede or facilitate task engagement. The BASC–SOS procedure entails watching the participant for 3 s and then recording the behaviors witnessed for 27 s, repeated over a 15-min period. The total number of each behavior type observed was used as the participant’s score. The BASC–SOS shows high internal consistency (0.8 with children, 0.9 with adolescents) and test–retest reliability (Reynolds & Kamphaus, 2004). Furthermore, it was designed specifically for use with children who are experiencing difficulties significant enough to impede academic progress and has previously been used in research with autistic students (Hodges et al., 2022).
Multiple researchers collected data across schools, and all BASC and SAND procedures were followed.
Results
The data were analyzed with IBM SPSS Statistics (Version 24). Alpha was set at .05. The Kolmogorov– Smirnov test (Berger & Zhou, 2014) was used to test normality. The behaviors that facilitate learning variable was normally distributed, D(53) = .09, p = .20. The behaviors that impede learning variable, D(53) = .13, p < .05, and the total number of sensory reactivity differences, D(53) = .13, p < .05; hyperreactivity differences, D(53) = .23, p < .001; hyporeactivity differences, D(53) = .268, p < .001; and sensory- seeking differences, D(53) = .15, p < .05, were not normally distributed; therefore, nonparametric tests were used.
Pearson correlations (Freedman et al., 2007) showed that age was correlated with behaviors that facilitate learning, r(51) = −.47, p < .001. Nonparametric Spearman’s rank correlation coefficient (Zar, 2005) showed that age was also correlated with sensory-seeking differences, r(51) = −.37, p < .05, but not with behaviors that impede learning, r(51) = .15, p = .28, or total number of sensory reactivity differences, r(51) = 1.17, p = .247; hyperreactivity differences, r(51) = 0.17, p = .247; or hyporeactivity differences, r(51) = .03, p = .844. Therefore, age was controlled for in analyses involving the behaviors that facilitate learning and the sensory-seeking differences variables. Descriptive statistics are provided in Table 1, correlation results are given in Table 2, and scatterplot matrices of significant results are provided in the Supplemental Materials (available online with this article at https://research.aota.org/ajot).
Descriptive Statistics for SAND Domains and BASC–SOS Classroom Behaviors
Note. BASC–SOS = Behavior Assessment for Children–Second Edition Student Observation System; SAND = Sensory Assessment for Neurodevelopmental Differences.
Correlation Matrix
*p = .05 (two-tailed).
Discussion
Sensory reactivity differences and autistic students’ classroom behavior were related. The more sensory reactivity differences students displayed, the more behaviors that impede learning and fewer behaviors that facilitate learning were observed. Our findings are consistent with those of previous research demonstrating that sensory differences have a negative effect on adaptive behavior and attention and participation (Dellapiazza et al., 2018; Jones et al., 2020; Mallory & Keehn, 2021). This finding adds to the field because we addressed a limitation of previous literature by using direct observational methods rather than teacher and parent reports. We also included autistic students with high support needs in special education settings, who are underrepresented in research.
Hyporeactivity was the only sensory subtype independently linked to classroom behavior. This is consistent with Liss et al.’s (2006) work based on parent report, which found that increased hyporeactivity was linked to general lower adaptive functioning. The relationship we found between hyporeactivity and reduced behaviors that facilitate learning is supported by previous research demonstrating that hyporeactivity has a negative impact on key learning skills, such as joint attention (Baranek et al., 2013), motor skills (Jasmin et al., 2009), and communication (Watson et al., 2011). Ashburner et al. (2008) found that increased hyporeactivity and seeks sensation scores on the SSP were related to inattention and reduced academic achievement. Our use of the SAND allowed us to separately assess hyporeactivity and sensory-seeking differences. Our results suggest that hyporeactivity differences could have been driving this link and that sensory seeking might not have been significantly linked in Ashburner et al.’s (2008) work if it had been a separate variable. Hyporeactivity may be less noticeable; thus, the role of hyporeactivity differences may have been underreported in the previous literature given its reliance on parent and teacher reports.
Hyperreactivity and sensory-seeking differences can be disabling for autistic students (Howe & Stagg, 2016); therefore, it is intriguing that we found no significant link to classroom behaviors. Nevertheless, we found moderate effect sizes for sensory-seeking differences and classroom behavior, which is supportive of the previous literature (Jones et al., 2020). Our finding may reflect the fact that teachers in special education settings make various adaptions to classrooms to accommodate sensory needs, mainly for hyperreactivity and sensory seeking (Pillar & Pfeiffer, 2016), including lowering lighting or providing rocking chairs. Without direct assessment, hyporeactivity differences may be harder for teachers to notice, so they may be unaware of a need for accommodations. Increased hyperreactive and sensory-seeking differences might be better supported because these differences are more overtly noticeable.
If hyporeactivity is playing a larger role in classroom engagement, this has implications for teaching and classroom design. Current design guidance already considers sensory reactivity differences, with a focus on how to address sensory hyperreactivity (Tola et al., 2021). Hyporeactivity needs to be supported by enhancing task stimuli so the task can reach the higher sensory registration levels of hyporeactive students (Dunn, 1997). These findings also have implications for the design of classrooms that allow for increased saliency of learning cues and stimuli.
Limitations and Future Research
These data were collected at a single time point and therefore may not be representative of participants’ overall classroom behavior, which may have affected our findings. We were unable to collect more information about participant characteristics, which is significant given that there is literature linking sensory differences to communication ability (Dellapiazza et al., 2018) and lower cognitive ability (Zachor & Ben-Itzchak, 2014). The SAND is a novel tool for sensory assessment but is not yet widely used in research. The BASC–SOS categorizes behaviors on the basis of neurotypical students’ learning styles; therefore, it is possible that behaviors may be inappropriately categorized for autistic students’ learning. Future research should measure variables multiple times and collect detailed participant demographic information.
Implications for Occupational Therapy Practice
This study has the following implications for occupational therapy practice: ▪ Autistic students showing increased behaviors that impede their learning or reduced behaviors that facilitate their learning might benefit from a sensory assessment. ▪ More research into the role of hyporeactivity differences in classroom behaviors is warranted. Students who are particularly hyporeactive may be at increased risk for not engaging and need targeted support.
Conclusion
This study found a link between increased sensory reactivity differences and more behaviors that impede learning and fewer behaviors that facilitate learning when assessed using direct observations in a special education setting. Hyporeactivity was correlated with both fewer behaviors that facilitate learning and increased behaviors that impede learning. This has implications for how support is tailored to students with increased sensory differences, especially those with hyporeactivity.
Supplemental Material
Supplementary material for Relationship Between Directly Observed Sensory Reactivity Differences and Classroom Behaviors of Autistic Children
Supplementary material, sj-pdf-1-aot-10.5014_ajot.2024.050345.pdf for Relationship Between Directly Observed Sensory Reactivity Differences and Classroom Behaviors of Autistic Children by Hannah Marcham and Teresa Tavassoli in The American Journal of Occupational Therapy
Footnotes
Acknowledgments
We thank the autistic students and their schools for participating. We also thank past students who supported this study, specifically Brett Davies, Charlotte Daniels, and Frankie Englezou. Last, we thank the Economic and Social Research Council’s South East Network for Social Sciences doctoral program for funding this research.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
