Research gap at a glance
- Question: Does the interaction between interoceptive accuracy and anticipatory bodily differentiation in predicting decision quality replicate in an independent, larger sample, and does it hold on the Iowa Gambling Task as well as on the intuitive reasoning task?
- Why it matters: This interaction is the only evidence in the wiki that interoception does anything to decision-making, and it underpins the gain-term reading of interoception used across the applied pages. It rests on two unreplicated interactions in one 2010 paper.
- Theme: Emotion and self
- Status: open
What the wiki already knows
- Interoceptive accuracy was unrelated to intuitive decision quality (r = .08), but it moderated the link between per-choice bodily differentiation and decision quality (Delta R squared = .08, n = 92), helping when the body favoured good choices and hurting when it favoured bad ones (Dunn et al. (2010)).
- No simple-slope tests were reported, so it is unknown whether poor perceivers are uncoupled from their bodies or inversely coupled; both published figures show the low-accuracy line sloping negative (Dunn et al. (2010), Does somatic feedback guide decisions?).
- An extreme-groups study on the Iowa Gambling Task found a main effect of heartbeat perception on deck choice but recorded no per-choice bodily term, so the moderation has never been measured on the paradigm that generated the literature (Werner et al. (2009)).
- Interoceptive accuracy in both studies was measured with heartbeat counting, whose stroke-volume and guessing confounds matter more when it is the moderator (Is the heartbeat-counting task a valid measure of cardiac perception?).
- Between 20% and 37% of healthy controls perform disadvantageously on the IGT, and 27% preferred unprofitable decks on the intuitive reasoning task (Iowa Gambling Task, Intuitive reasoning task (IRT)).
What would answer it
- A preregistered multi-site replication with several hundred participants, running both the Intuitive reasoning task (IRT) and the Iowa Gambling Task, measuring interoceptive accuracy with a discrimination task as well as counting, recording per-choice Skin conductance response (SCR) and heart rate to compute a bodily-differentiation index, and testing the interaction together with simple slopes at low and high accuracy.
Connections
- Raised on: Does somatic feedback guide decisions?, Dunn et al. (2010), Is more interoceptive awareness better?, Age-related interoceptive decline
- Related concepts: Somatic marker hypothesis, Interoceptive sensitivity, Body loop and as-if body loop
- Related debates: Does somatic feedback guide decisions?, Is more interoceptive awareness better?, Is the heartbeat-counting task a valid measure of cardiac perception?
Does somatic feedback guide decisions? calls a replication of the Dunn moderation the single most valuable absent study on that page. The claim it would test is precise: accurate perception of the body is not a good in itself but a gain term on whatever the body is signalling, so it improves decisions when bodily signals point the right way and worsens them when they point the wrong way. That reading now carries weight well beyond card games, including on Is more interoceptive awareness better?, where it is the basis for predicting that awareness training could help some people and harm others.
The original evidence is thin for the weight it bears. Dunn et al. (2010) reported two interactions at moderate sample sizes, from the laboratory that built both the hypothesis and the task, and the robustness checks were re-analyses of the same participants. The plotted lines also raise a theoretical question the paper did not test: if poor perceivers show a negative slope, their bodies would be relating to choice in the opposite direction, which no version of the theory predicts.
The paradigm question matters too. The somatic marker literature was built on the Iowa Gambling Task, but Dunn deliberately replaced it with a task that fixes its known flaws. Werner et al. (2009) used the IGT but without the per-choice bodily term, so the two lines of evidence have never met in one design. The same unrun study appears from a lifespan angle on Does interoception decline with age because the signal fades or because perception does?.