The Sense You Were Never Taught to Have

You are asked how you feel and something goes looking. Not for a word, though a word will be needed shortly. For the thing the word is meant to be about. There is activity in the chest, or somewhere behind it. The jaw is doing something. Whether any of this constitutes an emotion, or which one, is not obvious from the inside, and the pause before you answer is longer than the question deserved.

Most accounts of this treat it as an emotional problem. There is a physiological one underneath it, and it has a name that almost nobody uses outside a laboratory.

Interoception is the sensing of the body’s internal state. Heartbeat, breath, gut, temperature, the signals arriving from inside rather than from the world. It has been called the eighth sense, mostly by people trying to make it sound approachable, and the phrase does it no favours because it implies something as clean and unified as vision. It is not. That is the first thing worth knowing, and the research has spent the last decade establishing it.

What interoception actually means

The standard definition is broader than the popular one. Interoception covers not just sensing internal signals but interpreting, integrating and regulating them. The sensing is the part that gets discussed. The interpreting is where most of the difficulty lives.

The concept also splits. In 2014, Sarah Garfinkel and Hugo Critchley proposed that what was being loosely called interoceptive awareness was three distinguishable things, and the distinction has held up well enough to organise the field since.

Interoceptive accuracy is objective performance: can you actually detect the signal, measured against a machine that knows the answer.

Interoceptive sensibility is what you believe about your own interoception, gathered by questionnaire. Your own report of how tuned-in you are.

Interoceptive awareness is the relationship between the two. Whether your confidence tracks your accuracy. Metacognition about the body.

In a normative sample all three came apart. Accuracy was only partly predicted by the other two, and they aligned only among the people who were already most accurate. Which produces the finding that matters most here. How good you think you are at reading your body tells you almost nothing about how good you are.

That is not a rhetorical flourish. It has been demonstrated repeatedly, including in the population you would least expect. Experienced meditators, tested on objective heartbeat detection, performed no better than controls despite considerably greater confidence in their interoceptive ability. Years of attention to internal states raised sensibility. Accuracy did not follow.

The measurement problem, which you should know about

There is a complication here, and skipping it would be dishonest, because it undercuts a great deal of what gets written confidently about interoception.

The dominant measure for forty years has been the heartbeat counting task: sit still, count your heartbeats without taking a pulse, compare to the ECG. It is simple and cheap and it has produced most of what people believe about interoceptive accuracy.

It also appears to measure something other than what it claims. When researchers adapted the instructions to stress reporting only heartbeats the participant could actually feel, rather than estimated ones, scores dropped by around half. Performance turns out to be substantially predicted by two things that have nothing to do with sensing your body: time estimation, and prior knowledge of what a normal resting heart rate is. Knowledge of resting heart rate mediates the relationship between intelligence and task performance, which is a polite way of saying that clever people who know roughly what a heart rate is score better without necessarily feeling anything at all.

There is an active dispute about how damning this is, with published exchanges running in both directions, and the field has not settled it. What is not disputed is that a large body of confident claims about interoception rests on an instrument whose validity is under serious question.

There is a further problem for the tidy version of the story. If interoception were a single ability, people good at one channel would be good at others. They mostly are not. Correlations between cardiac, respiratory, gastric and thermal accuracy are weak or absent. Training transfers poorly: cardiac biofeedback improves heartbeat perception without improving anything else. One recent paper takes the position far enough to title itself There Is No Such Thing as Interoception, arguing the term is a useful heading for organising research rather than a coherent single phenomenon.

So the honest position is this. Something real is being described. The instruments for measuring it are worse than the literature’s confidence suggests, and it is probably several things rather than one.

Why the standard framing misses it

Search for interoception and you will mostly find occupational therapy resources aimed at children, usually autistic children, and usually framed around identifying when you need the toilet or when you are hungry. That work is legitimate and it helps people. It is also a small corner of the subject, and it has come to stand in for the whole.

What that framing cannot account for is the adult who has no difficulty knowing when they are hungry and considerable difficulty knowing what they feel. The signals arrive. Something goes wrong between arrival and interpretation, and the difficulty is not sensory in any simple sense.

This is where the concept earns its place, and where the interesting association sits.

Interoception and alexithymia

Alexithymia is difficulty identifying and describing your own emotions. It is a trait rather than a diagnosis, and it is common enough to be unremarkable, appearing across autistic populations, in depression, in eating disorders, and in a substantial number of people with no clinical presentation at all.

The obvious hypothesis was that alexithymia is interoceptive failure: you cannot name the feeling because the signal never arrives. A meta-analysis of 66 samples, more than 7,000 people, tested it.

The overall correlation was small and negative, r = −.162. Real, and much weaker than the theory predicted.

The detail is more interesting than the headline. The relationship depended heavily on which dimension was measured and how. And it was moderated by population: in psychiatric and developmental samples the association was moderate, while in healthy typically developing samples it was not significant at all.

Read that carefully, because it does not say what most summaries claim. It does not establish that people who struggle to name their feelings are failing to detect their bodies. In the general population, the connection largely is not there. Whatever alexithymia is, it is not simply a broken sensor.

Which leaves the more difficult possibility. The signal arrives. The difficulty is what happens next, at the point where a physiological state has to become an emotional one, and that step is not sensation at all. It is interpretation, and interpretation is learned.

What this means for the pause before you answer

Nobody teaches interpretation. Emotional vocabulary is acquired the way accents are, from whoever is around. If the people around you did not name states, or named them wrongly, you arrive at adulthood with a functioning body and no working index for it.

That is not a deficit in the person. It is an absence of training in something nobody announces as trainable, in a culture with a rich vocabulary for what things mean and a thin one for what they feel like.

It also explains why the standard advice fails. Sit with the feeling assumes the feeling is already labelled and available for sitting with. Name it to tame it assumes a name exists to be reached for. Both instructions are written by people for whom the interpretive step is automatic, addressed to people for whom it is the entire problem.

The research suggests something more modest and less satisfying. Attention to the body will raise your confidence in your interoception, reliably and measurably. It may not improve your accuracy, and the two are separable. Whatever is worth doing here is unlikely to be about detecting the signal harder.

The pause before you answer is not evidence that nothing is happening in there. It is evidence that something is happening and has not been assigned a word, which is an ordinary consequence of an ordinary developmental gap and is not a fault in the equipment. Whether the word ever arrives is a different question, and the literature does not answer it.

FAQ

What is interoception?

The sensing of the body’s internal state: heartbeat, breath, gut, temperature, the signals arriving from inside rather than from the world outside. The fuller definition covers not only sensing those signals but interpreting, integrating and regulating them, and the interpreting is where most of the difficulty tends to sit. It is sometimes called the eighth sense, which oversells how unified it is.

What is the difference between interoception and proprioception?

Proprioception is the sense of where your body is in space: limb position, joint angle, the reason you can touch your nose with your eyes shut. Interoception is the sense of the body’s internal condition: heartbeat, hunger, breath, temperature. Both are internal in the sense of not coming from the outside world, but they answer different questions. Proprioception answers where. Interoception answers how things are going in there.

Can you improve your interoception?

Partly, and the research draws a distinction worth knowing. Attention to the body reliably raises interoceptive sensibility, meaning your confidence in and self-reported sense of your own interoception. It does not reliably raise interoceptive accuracy, the objective ability to detect signals. Experienced meditators tested on objective heartbeat detection performed no better than controls despite much higher confidence. Training also transfers poorly between channels: improving cardiac perception does not improve gastric or respiratory perception.

Is alexithymia caused by poor interoception?

The evidence says no, or at least not simply. A meta-analysis of 66 samples and over 7,000 people found only a small negative correlation between alexithymia and interoceptive awareness, and the association was moderate in psychiatric and developmental populations but not significant at all in healthy typically developing samples. Whatever difficulty naming emotions consists of, it does not appear to reduce to a failure of detecting bodily signals in the general population.