Carter & Ogden (2023) — bodily states, and the thresholds that turn them into symptoms
The wiki’s first purely qualitative interview study, and its first source to describe symptom perception in participants’ own words rather than through a task, a questionnaire, or a generative model. Twelve adults (3 men, 9 women; aged 23–53; mixed — four with no disclosed condition, the rest with asthma, epilepsy, Crohn’s/colitis, anaemia, EDS, eczema, dyspraxia, anxiety, depression) were interviewed about how they feel, notice, make sense of, and talk about their bodily states, and how some of those states cross over into being “symptoms.” Thematic Analysis (Braun & Clarke) yielded four themes and a fifth transcending one.
Its value here is not new mechanism but a new register. Everything the wiki holds about symptom perception — Pennebaker schemata, precision-weighted priors, fear of bodily sensations — is stated by clinicians and theorists. This is the same phenomenon narrated from the inside, by people who have never heard of a prior, and it lands on the wiki as the phenomenological counterpart to machinery the wiki otherwise carries only in expert vocabulary.
The four themes, compressed
- The qualia of bodily states. Participants defined and located felt states with widely varying detail — tiredness as “heaviness”/“weakness,” pain as “achy”/“stabbing”/“a dull ache,” emotion in the “stomach” (both happiness and sadness), anger in the heartbeat and shoulders. A folk bodily map of emotion emerges unprompted, gut-centred, consistent with the emBODY topography (nummenmaa-2014-bodily-maps) but with no physiological warrant — exactly the caution that map literature carries.
- Attending to bodily states. A “baseline” body-attention varied between people (“I just spend a lot of time thinking about it” vs “not necessarily what my focus is on”), and specific properties raised the odds a state was noticed: novelty (“something that’s alien”), unusual location, duration beyond an expected window, and emotional response — negative appraisal amplifying attention in a cyclical, self-feeding way (“if you’re in a bad mood you tend to see things worse than they actually are”).
- Becoming symptoms. A state became a symptom when it deviated from a personal norm (norms that a diagnosis can quietly reset — “for me it’s kind of normal to have diarrhoea”) and when the person found meaning / attributed causality. Assigning a medical cause did two things at once: it recruited the state into an “ill-health” frame (making it a symptom) and it reduced the fear (“once they can understand it… it’s reassuring, because now they can put a name to this”).
- Reifying symptoms. The abstract felt state is consolidated into something communicable — through an internal voice (“if I keep telling myself my head hurts, it probably will start hurting”), through metaphor (“drumbeat in the head,” “run over by a bus,” “moving through treacle”), and through medical vs lay language chosen for its effect on the listener. Some states can only be signalled non-verbally (“I just stomp around and shout”); “hidden” conditions frustrate this because there is nothing for others to see.
The transcending theme: a series of thresholds
The paper’s one genuinely novel move is to make explicit what symptom-perception models leave implicit. Bodily states become symptoms by surpassing a threshold, and the four themes are each a way that threshold can be moved:
- Lowered by habitual body-focus, by negative affect (anxiety/depression “affects the whole body”), by expecting a symptom to recur, and by reading any new sensation in line with an existing diagnosis (a Crohn’s patient attributing “any niggle” to a flare “even if it didn’t necessarily turn into” one).
- Raised by deliberately ignoring or normalising states (“if I just don’t think about it… it doesn’t come up”), by respecting the body’s normal activity (“if I have a sensation it’s just doing something it needs to do”), and by simply not having a chronic condition that primes expectation.
Crucially, a state can pass one threshold (attention) without passing another (finding-meaning / reifying), so it is noticed but not symptomatised — “most of the time if I exercise it produces a sensation and I just… let it pass like you would with a thought.” This is a staged model, not a single cut-point. See symptom-threshold for the concept and its mapping onto the wiki’s predictive-coding material.
Why this matters to the wiki
It is schema-guided-symptom-perception and interoceptive-inference said in the first person. Strip the vocabulary and the alignment is close:
| Carter & Ogden (2023), participants’ terms | wiki’s expert terms |
|---|---|
| threshold lowered by expecting a symptom / reading it as one’s diagnosis | strong prior dominating perception |
| threshold lowered by negative affect and body-focus | precision up-weighted on interoceptive channels; schema activated by threat and attentional shift |
| a state “noticed but not made a symptom” | prediction error registered but not attributed / explained by a higher-level model |
| finding meaning / attributing medical cause | model selection — which hypothesis explains the sensation |
| threshold raised by normalising / ignoring | habituation; low prior probability of pathology |
The paper reaches this without a generative model, from interview transcripts — which makes it, like Van der Does et al. and Ehlers, a clinical/lay arrival at the predictive-coding picture from outside the predictive-coding camp. It belongs on the same shelf: the convergence is evidence the picture is not an artefact of one vocabulary.
It supplies the folk end of two other wiki threads. The gut-centred emotion localisation is emBODY topography reported spontaneously. And the participants’ varying insight into how their emotions colour their bodily states — some highly aware (“I’m fully aware of the extent to which depression can make you feel physically ill”), some not — is alexithymia observed as a continuum in ordinary talk, which the authors themselves flag (citing Bird/Bagby). Their own prior work adds a construct the wiki did not have a name for: “Interoceptive Crossover” (Carter & Ogden 2020) — individual differences in the extent to which people perceive emotions and physical symptoms as similar, a plausible mediator of both the alexithymia link and the gut-for-emotion mapping.
Brakes
Held loosely, and the authors are candid about most of it. N = 12, self-selected, mostly recruited from the authors’ own prior questionnaire study, so a sample of engaged people with strong health narratives. Interviews were remote (telephone/Skype-audio), which may have thinned rapport. The analysis is inductive Thematic Analysis run by two health psychologists whose symptom-perception priors are declared to have shaped the coding, and the interview schedule’s deliberately vague language (“states,” “feelings,” “sensations”) may have led participants to mirror it. Most importantly for this wiki: the thresholds are described, never measured — there is no accuracy task, no physiology, no operational definition of a threshold, so the model is a rich hypothesis, not a tested one. The paper closes on the same definitional problem the wiki keeps hitting (desmedt-2023-interoception-discrepancies): “sensation,” “symptom,” “somatic complaint” and “bodily state” are used inconsistently across disciplines, and a headache can be a bodily state below threshold and a symptom above it — the threshold idea is partly a way of dissolving that terminological knot.