Interoception—the brain’s ability to sense internal bodily signals like heartbeat and breathing—is often portrayed as a key “brain–body bridge” to mental health. In recent years, researchers have proposed that people with psychiatric vulnerability should show altered interoceptive performance, linking shifts in symptom experience to how accurately and confidently individuals detect internal sensations. But a new study in Nature Mental Health challenges that popular idea, suggesting that the story is more complicated than many models assume.
The research team tested this claim in a large cross-sectional community sample, using psychophysically optimized tasks designed to measure interoception with high rigor. Participants completed objective assessments in two domains: cardiac interoception (N = 456) and respiratory interoception (N = 245). Rather than relying on simple correlations, the study applied hierarchical Bayesian modeling to evaluate relationships between interoceptive metrics and a broad set of symptom dimensions.
Crucially, the analysis targeted multiple components of objective interoceptive performance, including sensitivity (how well participants detect signals), precision (how reliably they separate signal from noise), and metacognition (how accurately confidence tracks performance). These metrics were then tested across several model types—linear, categorical, and even network-based frameworks—to see whether any consistent signature of psychiatric symptoms emerged.
The results were striking: objective interoceptive performance measures were largely unrelated to mental health dimensions. In other words, variations in how well participants could detect internal bodily signals did not translate into meaningful differences in symptom profiles, regardless of the analytic strategy. The findings therefore run counter to the expectation that interoceptive detection accuracy should serve as a broad marker of psychopathology.
By contrast, self-reported interoceptive sensibility—how people believe they experience internal signals—showed moderate associations with symptoms. However, semantic similarity analyses indicated that these links may reflect higher-order interpretation and affective beliefs, not the perceptual processes measured by the objective tasks.
This distinction matters, the authors argue, because “believing you feel your body” may not mean “perceiving your body accurately.” The study implies that interoception research may need to separate interpretative styles and emotional appraisal from genuine sensory performance. Otherwise, symptom associations could be misattributed to sensory mechanisms.
Overall, the work suggests a shift in how interoception is operationalized in mental health science. If objective performance does not reliably track psychiatric vulnerability across symptom domains, researchers may need new frameworks—and better measurement approaches—to identify what aspects of interoception are truly clinically informative.
Instead of a universal interoceptive deficit, the results point toward a more selective, context-dependent relationship between internal signal processing and mental illness. For readers hoping for a clear “interoception → symptoms” pathway, the message is sobering—but scientifically valuable: the brain–body connection in mental health may involve beliefs and interpretations at least as much as sensory accuracy.
Subject of Research: Interoception and mental health symptoms (cardiac and respiratory interoceptive performance)
Article Title: Interoceptive performance is unrelated to mental health symptoms in a large multi-domain psychophysical investigation.
Article References: Banellis, L., Nikolova, N., Ehmsen, J.F. et al. Nat. Mental Health (2026). https://doi.org/10.1038/s44220-026-00688-4
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s44220-026-00688-4
Keywords: interoception; mental health; psychophysics; hierarchical Bayesian modeling; cardiac and respiratory sensing
