If you’re working with autistic students in schools, you’ll be familiar with repetitive motor behaviours, commonly referred to as “stims” or “stimming.” These behaviours, which are a subtype of restricted and repetitive behaviours, can include hand flapping, rocking, spinning objects, or twirling hair. Whilst these behaviours are recognised as a means of emotion regulation and self-regulation in autistic persons, this study aimed to investigate the function of stereotypies across different co-diagnoses. The researchers expected that the results would show a high prevalence of stereotypies in autistic persons and persons with social phobia, medium prevalence in persons with OCD, and low prevalence in persons with ADHD. They expected that stereotypies function as self-regulation strategy across diagnoses, though the strongest agreement was expected for autistic persons. They also hypothesised that co-occurrence would increase prevalence among persons with both ASD and social phobia or OCD, but not for persons with ASD and ADHD.

I know this article is a step away from OT journals, but I think the results provides some very helpful insights into understanding our autistic children’s behaviours.

Methods

Participants were recruited from the Aspie Quiz website, which attracts over 100,000 visitors per year.

The study occurred in three phases. Although they don’t describe it as a Delphi study, the approached used appears very similar. Across the stages they narrowed down the options and continued to ask participants their experiences to reach a collective agreement. In the first phase, 3,967 participants took the quiz, with 1,558 scoring high on neurodiversity being offered to complete the first questionnaire. Participants were provided with 45 stereotypies and 32 internal states, selecting up to three internal states for each stereotypy or indicating “I do not do this stim” or “non-emotional reason.” A total of 779 participants completed this phase.

In the second phase, the researchers used an iterative process to refine stereotypy-internal state combinations. For example, I flap my hands when I’m excited. Stereotypies were listed with four response options: “I do not do this,” “I do this during a specific state” with the states being chosen from phase one results, “I do it independent of mood,” and “I do it for other reasons.”

In the third and final phase, 22 stereotypies and internal state pairs were included, plus one that served as a control—whistling. This questionnaire was completed by 4,990 participants (2,890 men, 2,100 women, mean age 30.93 years) before taking the Aspie Quiz.

Participants could indicate whether they self-identified as an autistic person, a person with ADHD, a person with OCD, or a person having social phobia. They could also state they had a formal diagnosis for any of the four conditions. There was no proof of diagnosis or interviews to confirm it—the researchers relied on participants’ honesty.

Among the 4,990 participants, 229 reported being diagnosed with ASD (4.6%) and 1,915 chose self-diagnosis (38.4%). A total of 459 reported being diagnosed with ADHD (9.2%) and 803 chose self-diagnosis (16.1%). A total of 205 persons reported being diagnosed with OCD (4.1%) and 951 chose self-diagnosis (19.1%). A total of 282 reported being diagnosed with social phobia (5.7%) and 1,582 chose self-diagnosis (31.7%).

Data analysis involved three main approaches: prevalence analysis (contrasting “I do not do it” with doing the behaviour), validity analysis (contrasting “I do it independent of mood” with doing it for self-regulation), and agreement analysis (examining whether the behaviour was done for the specific proposed emotion).

The Results

Selection of Stereotypies and Their Functions

In phase one, the researchers found that many stereotypies were done for multiple reasons, and various stereotypies were done for the same reason. For example, ‘talking to oneself’ was most commonly done when thinking, whilst ‘flapping hands’ was typically done when excited. Overall, 20 different stims were done when thinking, and 22 stims were done when bored, illustrating there’s rarely a one-to-one mapping of stims to internal states.

Furthermore, some stims were done for seemingly contradictory reasons. Rocking back-and-forth was done when thinking, for comfort, when overwhelmed, when overstimulated, to calm oneself, but also when distressed, when upset, and when sad. Stereotypies were done for over-arousal states like anxiety and overstimulation, and under-arousal states like boredom, but not for being tired or sick.

Prevalence

The highest prevalence was among autistic persons, followed by persons with social phobia, OCD, and least among persons with ADHD. For all four conditions, there were no significant differences in prevalence between self-diagnosed and formally diagnosed participants.

Validity: Self-Regulation Function

Doing the stereotypy for self-regulation was statistically significant for the ASD group but not for any of the other three conditions.

Co-occurrence Effects

There is also a lot of data looking at number of stereotypies used by the different diagnostic groups. Co-occurring diagnosis and autism had the higher aver number for example autism and ADHD had an average of 12.12 stereotypies. If you are interested this, it’s in the article along with inferential analysis on group differences. The function of the stereotypies was unaffected by co-occurrence.

Proactive Versus Reactive Self-Regulation

There was a significant overall difference between proactive and reactive processes, with stereotypies being more often done for the proposed proactive internal state than for the reactive internal state. There were higher prevalence and validity for proactive pairs than for reactive pairs, most pronounced among persons with a formal diagnosis of ASD. Thus, stereotypies are not non-functional behaviours but are self-regulatory behaviours used to reduce negative emotions and achieve positive emotions.

Practical Takeaways

For school-based occupational therapists, this research offers several important insights that should inform how we understand and respond to stimming behaviours in our students. Firstly, the findings consolidate the view that stereotypies are a means of emotion regulation and self-regulation, particularly for autistic persons. This isn’t new information, but having it confirmed with nearly 5,000 participants adds weight to what many of us observe in practice.

The finding that stereotypies serve both reactive and proactive self-regulatory functions is particularly relevant. Students aren’t just using these behaviours to cope with negative states—they’re also using them proactively to achieve positive states like pleasure and comfort. This challenges any remaining deficit-based views that see stimming as problematic behaviour that needs to be eliminated.

The research reveals that there’s no universal mapping of specific stereotypies to specific internal states. The same behaviour might be used for different reasons by different students, or even by the same student in different contexts. This means we cannot assume we know why a student is engaging in a particular behaviour without understanding their individual experience.

The co-occurrence findings are important for understanding the complexity of students’ presentations. When autism co-occurs with OCD or social phobia, we can expect to see approximately 1.5 more stereotypies on average. However, the function remains the same—they’re still primarily for self-regulation. This suggests we shouldn’t change our fundamental approach based on comorbidities, but we should be aware of potentially increased frequency.

The finding that stereotypies are more common for proactive reasons in autistic persons—doing them for pleasure or comfort—reminds us that stimming isn’t just about managing distress. It’s also about achieving positive states, which is a strength-based perspective that should inform how we support students.

This online survey study of 32 occupational therapy educators examined perspectives on ableism in occupational therapy, finding that 72% perceived the profession as inherently ableist. Themes emerged including cultural sensitivity, challenges engaging in disability studies, and power dynamics in OT education, with most respondents agreeing the profession focuses on independence, affirming past critiques.

This study using longitudinal data from four independent birth cohorts demonstrated that earlier- and later-diagnosed autism have different developmental trajectories and genetic profiles. Two modestly correlated autism polygenic factors were identified: one associated with earlier diagnosis and lower early social/communication abilities, and another associated with later diagnosis and increased socio-emotional difficulties with stronger genetic correlations to ADHD and mental health conditions.

This study of 155 undergraduate students used latent profile analysis to examine the relationship between misophonia and autistic traits. Findings suggest misophonia and autistic traits may co-occur and strong associations with emotional functioning, social functioning, attention-deficit symptoms, and repetitive behaviours also shown.

This study reviewed sensory processing in autism. The authors highlight that sensory processing differences, reported in up to 97% of autistic individuals, supports recognition in the updated diagnostic manuals as a defining feature of autism rather than just an associated characteristic. They note that sensory features don’t exist in isolation—they interact with co-occurring conditions like anxiety, ADHD, and sleep and feeding difficulties, which shapes developmental trajectories and impacts daily participation and quality of life for both individuals and families.

There is a discussion on how sensory differences present in each sensory system including touch, vision, auditory, smell and taste and interoception and how sensory differences interact across the lifespan. The review presents a conceptual model showing three core sensory profiles in autism—hyper-reactivity, hypo-reactivity, and sensory seeking—as overlapping domains rather than discrete categories. Each profile is associated with different co-occurring conditions and behavioural manifestations: for example, hyper-reactivity is linked with anxiety, food selectivity, and sleep disturbances, while hypo-reactivity is associated with reduced pain sensitivity and joint attention deficits.

The model also maps intervention strategies to specific sensory profiles, showing that Sensory Integration Therapy may benefit sensory seeking and hypo-reactivity profiles, whilst sensory-based interventions and the Sequential Oral Sensory Approach are more commonly used for hyper-reactivity. The article is open access and very comprehensive, if you’re working with autistic students, it is a good one to have a look at. To conclude, the authors suggest sensory processing provides a framework through which autism’s heterogeneity can be understood from both mechanistic and clinical perspectives, offering concrete opportunities to improve diagnosis and individualised intervention.