The gap
Virtually all empirical research on interoception in autism has been conducted with autistic adults without intellectual disability. The interoceptive experiences, abilities, and needs of autistic people with intellectual disability — who represent a substantial proportion of the autistic population — are almost entirely unstudied.
Why it matters
Interoceptive differences have direct practical consequences: not recognising hunger, not identifying the need to toilet, not detecting pain, not connecting bodily sensations to emotions. These consequences are arguably more impactful for people with intellectual disability, who may have fewer alternative strategies for identifying and communicating internal states, and who may depend on carers to recognise what they cannot report.
Behaviours that are commonly attributed to “challenging behaviour” in people with intellectual disability — sudden agitation, self-injury, food refusal, apparent non-compliance — may in some cases have interoceptive roots. A person who cannot detect that they are in pain, or who cannot distinguish hunger from anxiety, may express that unrecognised internal state through behaviour that is then managed as a behavioural problem rather than addressed as a sensory one.
Without research in this population, we are operating on assumption rather than evidence.
What exists so far
The existing interoception literature in autism (DuBois et al., 2016; Garfinkel et al., 2016; Williams et al., 2023) explicitly acknowledges that findings cannot be generalised to autistic people with intellectual disability. The measurement tools used — heartbeat counting tasks, self-report questionnaires like the MAIA — require cognitive and linguistic capacities that exclude most people with significant intellectual disability. Mahler’s (2017) clinical interoception curriculum includes adaptations for people with developmental disabilities, but empirical evaluation of these adaptations is minimal.
Observational and informant-report approaches to assessing interoception in this population have not been systematically developed or validated.
What would fill it
Research is needed that develops and validates methods for assessing interoception in people who cannot complete self-report questionnaires or standard behavioural tasks. This might include carer-report instruments, observational protocols, or physiological measures (e.g., skin conductance, heart rate variability) as proxy indicators of interoceptive processing. Studies should be conducted with autistic people with a range of intellectual abilities, not only those at the upper end of the cognitive spectrum. Participatory research approaches involving carers, support workers, and (where possible) autistic people with intellectual disability themselves should inform the design of such studies.