The gap
Despite extensive research characterising sensory processing patterns in autism and growing evidence that autism is heterogeneous at both the genetic and phenotypic level, nobody has systematically investigated whether a person’s sensory processing profile predicts how they respond to different interventions. In particular, the phenotypic subgroups identified by cluster-analytic and latent class approaches — including the classes described by Litman & Sauerwald (2025), already in this wiki — have not been linked to sensory processing profiles or to differential intervention outcomes.
Why it matters
If autism is heterogeneous — and the evidence overwhelmingly says it is — then one-size-fits-all interventions are unlikely to work equally well for everyone. Sensory processing patterns are a promising candidate for personalising intervention, because they are relatively stable, measurable, and directly relevant to daily functioning. A person who is predominantly hyporesponsive may need a very different environmental and therapeutic approach than a person who is predominantly hyperresponsive, and both may need something different from a person with mixed patterns.
The Litman & Sauerwald (2025) phenotypic classes — which include a “Sensory/repetitive” class characterised by high sensory sensitivity and repetitive behaviours — suggest that sensory processing may define meaningful subgroups within autism. But without research linking these subgroups to intervention outcomes, the clinical utility of the classification remains unrealised.
What exists so far
Sensory subtyping research using instruments like the Short Sensory Profile, SEQ, and SP-2 has consistently identified 2–4 sensory subtypes within autism (e.g., Ausderau et al., 2014; Lane et al., 2014; Ben-Sasson et al., 2008). These subtypes — typically described as hypo-dominant, hyper-dominant, mixed, and sensory-seeking — are replicable across studies, though the exact classification depends partly on the instrument used.
However, virtually all of this subtyping research stops at description. The question “Do people in different sensory subtypes respond differently to specific interventions?” has not been addressed. A small number of studies have examined whether baseline sensory features predict response to specific sensory integration therapy protocols (e.g., Schaaf et al., 2014), but these are limited in scope and have not been linked to broader phenotypic classification schemes.
What would fill it
A research programme that combines phenotypic classification (using genetic, behavioural, and sensory data) with prospective intervention studies. Specifically: administer a comprehensive sensory assessment (ideally covering all senses including interoception) at baseline; assign individuals to phenotypic subgroups using validated classification methods; deliver a defined intervention; and measure differential response by subgroup. This would require large sample sizes, multi-site collaboration, and the willingness to treat sensory processing as a moderator of intervention effectiveness rather than just a description of the population.
For the broader Dutch autism practice community, this gap represents an opportunity: the combination of detailed sensory assessment (via the prikkelprofiel), access to well-characterised autistic populations (including people with intellectual disability), and a strong clinical tradition in sensory processing work could make the Netherlands an ideal setting for this research.