Humans do not stop at five neat senses. Oxford psychologist Charles Spence and colleagues have long argued for a working range of roughly 22 to 33 sensory channels — a range, not a census — a framing already tied to a £1.9 million Arts and Humanities Research Council project that started in 2013. This is an explainer of that evidence, not a 2026 lab breakthrough.
What happened
Aristotle’s five — sight, hearing, touch, taste, smell — is a teaching device, not a parts list. Philosophers and scientists treated the senses in isolation for a long time because the vocabulary did. Modern sensory neuroscience does not. Dedicated receptor classes and pathways handle far more than those five labels, and everyday experience is almost always multisensory: what you see changes what you hear; what you smell changes what you think you are tasting.
In 2013, the Centre for the Study of the Senses at the University of London’s School of Advanced Study launched Rethinking the Senses: Uniting the Philosophy and Neuroscience of Perception, an AHRC Science in Culture large grant. The principal investigator was the late Professor Sir Colin Blakemore. Co-investigators included Charles Spence (University of Oxford), Ophelia Deroy (London), Fiona Macpherson (University of Glasgow), and Matt Nudds (Warwick). The University of Glasgow’s Centre for the Study of Perceptual Experience still hosts the project page. The AHRC case-study page states the scientific provocation in one sentence: instead of five, “we might have to count up to 33 different senses, each served by dedicated sets of receptors.”
Barry C. Smith, director of the Institute of Philosophy at the School of Advanced Study and a collaborator on that project, restated the point in The Conversation (article 270697). His long-term collaborator, Professor Charles Spence, head of the Crossmodal Research Laboratory at Oxford, told him that neuroscience colleagues believe there are “anywhere between 22 and 33 senses.” That is the sentence the 2026 aggregators copied. It is an expert range. It is not a 2026 count, and Spence has not published a table that enumerates 33 named senses as a completed inventory.
What the range is pointing at is not mysterious. Among the systems that sit outside the schoolroom five:
| System | What it detects | Why it is not “just touch” |
|---|---|---|
| Proprioception | Limb position without looking | Muscle and joint receptors, not skin |
| Vestibular / equilibrioception | Head motion and balance | Inner-ear canals, integrated with vision |
| Interoception | Internal state — heartbeat, hunger, breath | Visceral afferents; clinical literature of its own |
| Nociception | Pain | Dedicated nociceptors |
| Thermoception | Temperature | Separate receptor classes from pressure |
| Agency / ownership | The feeling that a movement (or a limb) is yours | Can fail in stroke even when sensation remains |
| Gustation vs flavour | Salt, sweet, sour, bitter, umami on the tongue | Fruit “tastes” are mostly retronasal smell plus touch |
Touch itself is already a bundle: pressure, pain, temperature, itch. Flavour is a combination of gustation, retronasal olfaction, and touch. Smith’s essay is careful on raspberry: there is no raspberry receptor on the tongue, and no arithmetic of sweet-plus-sour that builds it. Odour compounds travel from the mouth to the nose via the nasopharynx. Aircraft noise, in work from the same group, reduces perception of salt, sweet and sour but not umami — which is why tomato juice on a plane is a sensory result, not a meme.
The live article cited “research published in 2024” on interoception and mindfulness without naming a paper. That sentence is gone. The field has a real literature; it does not need a ghost citation. A standard roadmap is Khalsa, Adolphs, Cameron et al., “Interoception and Mental Health: a Roadmap,” Biological Psychiatry: Cognitive Neuroscience and Neuroimaging 3, 501–513 (2018). DOI 10.1016/j.bpsc.2017.12.004. Interoceptive disruption shows up in anxiety and eating-disorder research. That is a clinical research programme, not a 2026 reveal that humans “have” a 23rd sense.
Blakemore’s public version of the expanded framework was already in circulation by 2014. He died in 2022. Treating him as a source for a 2026 discovery is a dating error.
Why it matters
The useful claim is not a number. It is that perception is constructed from many specialised channels that the brain binds, and that education, medicine, and interface design still talk as if there were five pipes.
In the clinic, interoception is a handle on conditions where body awareness is part of the pathology. Proprioceptive training is already routine in rehabilitation after injury or stroke. Vestibular processing is a reason some children cannot sit still in a classroom without it being a morality story. None of that requires 33 to be the True Count. It requires not collapsing those systems into “touch.”
In flavour science and aviation, the same group’s work is already applied: umami surviving cabin noise is a testable, dated observation from the Rethinking the Senses programme, not a lifestyle tip invented for this URL.
Virtual and augmented reality still undershoot because they simulate sight and sound and then are surprised when people feel sick. Vestibular and proprioceptive mismatch is a named problem. That is a design constraint, not a product announcement.
What this page must not do is tell a search engine that a 2026 paper “revealed” 33 senses. ScienceDaily and similar outlets reprinted Smith’s Conversation essay in 2026. Reprinting is not a result. The Good Signal’s job, on a URL that already ranks for this language, is to put the 2013 date in the first screen and to refuse a census that Spence never took.
What to watch next
- Named papers, not ranges. If a 2026–27 study actually enumerates receptor classes or maps a new interoceptive pathway, cover that study on its own terms. Do not bolt it onto this explainer as proof that “the number is 33.”
- Crossmodal results with methods. Footstep-sound body-weight illusions, Tate Britain audio-guide memory effects, cabin-noise taste — those are experiments from the AHRC project. New work should be cited the same way: lab, year, measure.
Sources
- Barry C. Smith, “Humans could have as many as 33 senses,” The Conversation (270697) — https://theconversation.com/humans-could-have-as-many-as-33-senses-270697
- AHRC Science in Culture, “Rethinking the Senses: Uniting the Philosophy and Neuroscience of Perception” — https://www.sciculture.ac.uk/project/rethinking-the-senses/
- University of Glasgow, Centre for the Study of Perceptual Experience, project page (PI: Colin Blakemore; CI: Charles Spence) — https://www.gla.ac.uk/research/az/cspe/projects/rethinking-the-senses/
- Khalsa, S.S. et al. Interoception and Mental Health: a Roadmap. Biol. Psychiatry Cogn. Neurosci. Neuroimaging 3, 501–513 (2018). https://doi.org/10.1016/j.bpsc.2017.12.004
- Oxford Crossmodal Research Laboratory (Charles Spence) — https://www.psy.ox.ac.uk/research/crossmodal-research-laboratory
