Do we know how well we judge pain as well as we judge sight?
People's confidence tracked the accuracy of their pain judgements just as well as for vision and warmth, but a person's skill at judging their own pain decisions was unrelated to their skill for vision.
Source
Metacognition across sensory modalities: Vision, warmth, and nociceptive pain
Study at a glance
- Design
- Human experiment — Within-subject two-interval forced-choice intensity discrimination in three modalities (visual contrast, innocuous warmth, noxious heat), each followed by a 1-4 confidence rating, with staircases holding accuracy near 70.7%; two sessions on separate days.
- N
- N=36 · 36 healthy adults in the final sample, set by a sequential Bayes-factor stopping rule; incomplete or at-chance datasets were excluded and replaced.
- Population
- Healthy adult volunteers at University College London (18 female, aged 19-38) with normal vision and skin sensation
- Outcome
- Perceptual sensitivity (d'), metacognitive sensitivity (meta-d'), metacognitive efficiency (meta-d'/d') and mean confidence in each modality, and their correlations across modalities
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
Thirty-six adults compared pairs of stimuli and said which was stronger: which of two striped patches had higher contrast, which of two warm touches on the hand was warmer, and which of two hot touches was more painful. After each choice they rated their confidence from 1 to 4. An adaptive staircase kept each task hard enough that people were right about 70.7% of the time, and signal detection theory was used to separate how well people perceived (d') from how well their confidence distinguished right from wrong answers (meta-d', and the ratio meta-d'/d', called metacognitive efficiency).
What they found
Contrary to the authors' prediction, metacognitive efficiency did not differ between pain, warmth and vision; on average people were close to optimal, with efficiency near 1 in all three tasks. Average confidence also did not differ between modalities, and people who were generally confident in one task were confident in the others. However, individual differences in metacognitive efficiency correlated between vision and warmth (r = 0.42) but not between vision and pain (r = -0.04), suggesting that confidence about pain draws on something extra, perhaps its emotional or arousing quality.
The limits
What it doesn't show
With only 36 participants, estimates of correlations between tasks were imprecise; the warmth-pain correlation was inconclusive, and the authors say the pain-specific finding needs replication. Safety limits meant the warm and painful temperature ranges sat close together, so some 'warm' stimuli felt slightly painful and some 'painful' ones did not, and the warmth task ended up slightly harder than the others. The study measured experimentally induced heat pain in healthy young adults, so it says nothing directly about chronic pain, and the explanation that affect or arousal drives the pain-specific variance was not tested.
Key terms
- Metacognition
- The ability to monitor and evaluate one's own mental processes, here measured by how well confidence ratings separate correct from incorrect decisions.
- Metacognitive efficiency (meta-d'/d')
- How much of the information available for the perceptual decision is reflected in confidence, expressed as a ratio so it is not confounded by how good someone's perception is; 1 means confidence uses all of it.
- Metacognitive bias
- A general tendency to be over- or under-confident regardless of whether one is right; measured here as mean confidence.
- Two-interval forced choice (2IFC)
- A task where two stimuli are presented one after the other and the participant must choose which one is stronger.
- Staircase procedure
- A method that makes the task easier after errors and harder after correct answers so that accuracy stays at a fixed level for everyone.
- Nociceptive pain
- Pain produced by activating pain receptors (nociceptors) with a potentially harmful stimulus, such as noxious heat.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What was the key dependent measure used to compare metacognition across modalities?
Common questions
Why not just compare how confident people were in each task?
Average confidence (bias) is different from whether confidence tracks accuracy (sensitivity). Someone can be very confident on every trial yet have no idea which answers were wrong, so the study measured both separately.
Why divide meta-d' by d'?
People who perceive better have more information to be confident about, so raw metacognitive sensitivity partly reflects perception. The ratio removes that, which mattered here because the warmth task turned out slightly harder than the others.
Does this mean pain metacognition is a separate brain system?
Not necessarily. The authors offer two explanations: a separate monitoring process for pain, or one shared system that is pushed around by pain's emotional and arousal responses. The data cannot distinguish them.
More on Metacognition