Citation
Williams, Z.J., Suzman, E., Bordman, S.L., Markfeld, J.E., Kaiser, S.M., Dunham, K.A. et al. (2023). Characterizing interoceptive differences in autism: a systematic review and meta-analysis of case-control studies. Journal of Autism and Developmental Disorders, 53, 947–962. doi: 10.1007/s10803-022-05656-2
Key findings
- Across studies, autistic individuals showed small-to-moderate reductions in interoceptive accuracy compared to non-autistic controls, though this effect was inconsistent and subject to significant heterogeneity.
- Findings on interoceptive sensibility (self-report) were highly variable and did not converge on a clear direction. Some studies reported higher, some lower, and some equivalent sensibility in autistic participants.
- The meta-analysis identified important methodological sources of heterogeneity, including the use of different self-report questionnaires that may measure subtly different constructs (e.g., the BPQ vs. the MAIA).
- The authors recommended that future research use more specific measures of interoceptive attention (e.g., the Interoceptive Attention Scale) and control for confounds including alexithymia, anxiety, and depression.
- The overall picture is one of genuine but modest and inconsistent interoceptive differences in autism, unlikely to be captured by any single measure.
Method in brief
This was a systematic review and meta-analysis of case-control studies comparing interoceptive variables between autistic and non-autistic participants. The authors searched multiple databases and applied stringent inclusion criteria. Meta-analytic pooling was performed separately for interoceptive accuracy (using heartbeat tracking and discrimination tasks), interoceptive sensibility (self-report measures), and interoceptive awareness (confidence-accuracy correspondence). Moderator analyses explored the impact of alexithymia, sex, IQ, and measurement approach.
Relevance
This is currently the most rigorous quantitative synthesis of the interoception-in-autism literature. Its value lies primarily in mapping the state of the field honestly: the evidence for interoceptive differences in autism exists but is far less clear-cut than some earlier narrative reviews suggested. The paper’s methodological recommendations — particularly the call to disentangle interoception from alexithymia and to use more precise measurement tools — provide a roadmap for future research.
For the wiki’s purposes, this paper is a corrective to oversimplified narratives. It would be inaccurate to state that “autistic people have impaired interoception” as a blanket claim. The more accurate statement is that interoceptive processing may differ in autism, that the pattern of difference is complex and may involve dissociations between accuracy and self-report, and that alexithymia is a critical confound that must be accounted for.
Limitations
- The included studies were predominantly of autistic adults without intellectual disability, limiting generalisability.
- The heartbeat tracking task, used in most accuracy studies, has known methodological issues that may introduce noise.
- Even the meta-analytic estimates are based on relatively few studies with modest sample sizes.
- The field is evolving rapidly, and the meta-analysis’s search window (ending mid-2022) means that some more recent studies are not included.