nucleation

THEORY-662: The feeling of fluency is an inference about the precision of one's own processing, triggered by a change in precision against expectation; because its cause is internal and not given to introspection, a felt discrepancy is attributed to the most salient external source, as familiarity, liking or a sense of error

4 min read

Properties4
StatusProposed
Tagsemotion-and-affect, cognition, consciousness
SourceLIT-957 — Relative fluency (unfelt vs felt) in active inference
SupportsCLAIM-337 — Active inference already runs an internal loop: mental action selects among higher-level cognitive states and sets the precision of lower-level observations, structure learning reshapes the generative model without action, and a judgement is behaviour, so a judgement–observation loop may be action and perception one Markov blanket inward, CLAIM-439 — A regime of valence is felt, in the sense of attended and available to the system's other processes, when the system that is in it also models it: when a deeper process infers which ordering wins its negotiation, which way the system is moving along it and how strongly it prevails, and the inference is used elsewhere in the system; unfelt valence is the potential without that model, as unfelt fluency is a change in precision without the inference about it

Graph View

Source

Brouillet and Friston (2023), LIT-957, read in full in NOTE-785 from the authors’ manuscript; page numbers are the manuscript’s. The experiments it reviews were not read and are reported as the paper states them.

What was actually shown

The reviewed findings (pp. 8–13), secondhand.

  • Relative fluency, ease against expectation, is distinguished from absolute fluency (Whittlesea and Leboe 2003).
  • In the hension effect (Whittlesea and Williams 1998, Exp. 3), new orthographically regular non-words drew more false recognitions than new real words or irregular non-words, although real words were read fastest. It is read as surprise: easy to pronounce against hard to identify, attributed to the stimulus.
  • Motor fluency of an unrelated preceding gesture, or of a colour associated with a fluent gesture, raised “old” responses to new items (Brouillet et al. 2017, 2022). In typing, the sense of having erred came from a gap between expected and experienced fluency (Wilbert and Haider 2012).

The proposal (pp. 14–18). Unfelt fluency is the subpersonal recognition of a change in precision, which orients attention, likened to exogenous attention. Felt fluency is “a hierarchically deeper representation of fluency”, made opaque by mental action, that is, endogenous attention. Because felt fluency “arises from internal inferences about the precision of predictive processing, not the sensorium”, it “cannot be attributed to its subpersonal cause—and is therefore plausibly attributed to the sensorium” (p. 18).

What could have come out otherwise. In the reviewed experiments, judgements could have tracked absolute ease. They tracked discrepancy. The precision reading itself was not exposed to any test.

What this does not say

  • That fluency is precision. That identification is proposed by analogy; nothing is simulated or derived, and felt and unfelt fluency are not variables in any model (C2–C4, weak).
  • That fluency is affect. The paper treats fluency as a metacognitive feeling of ease. It touches valence only by citing Hesp et al. (2021) and the hedonic-fluency literature.
  • That “felt” means phenomenally felt. Felt fluency is the state endogenous attention makes reportable. The paper gives no argument that unfelt fluency lacks experience, and speaks of “a phenomenal sense of non-felt fluency” (p. 11). It cannot separate felt from unfelt in the sense QUESTION-059 asks about (NOTE-785).
  • That the neuromodulatory story is shown. The link to dopamine and noradrenaline is by citation.
  • What would refute it. The paper does not say.

Connections

  • Barrett’s categorisation of affect by context (THEORY-057) has the same structure: a feeling with an internal cause is attributed to the most salient source. The paper does not draw the parallel.
  • If both this and THEORY-045 hold, some felt states are about the brain’s own processing and not about the body. Felt fluency here involves no body in the loop, unlike Damasio’s hybrid feelings (THEORY-093). Neither is a contradiction, since fluency is not offered as a homeostatic feeling.
  • Overt somatic markers (THEORY-500) are a felt quality entering a judgement about its source, with a body state as the feeling.
  • Cross-modal fluency (LIT-190, NOTE-146) is the record’s other reading of fluency misattribution.

Note of 2026-10-29: what holds it back

The title states as the account of fluency what the paper proposes by analogy: no generative model with felt and unfelt fluency as variables has been built, and none has predicted a new reversal that an experiment then found, as the promote_when asks. A split into what Brouillet and Friston propose would free neither CLAIM-337 (also grounded in THEORY-564 and THEORY-577) nor CLAIM-439 (resting on CLAIM-440), all Proposed.

Note of 2026-10-30: what holds it back

Rechecked after CLAIM-439’s version 3 (2026-10-30), which repaired its edges to THEORY-873 and THEORY-882 and still rests on CLAIM-440, Proposed. The wait is unchanged: no generative model with felt and unfelt fluency as variables has been built, and none has predicted a new reversal that an experiment then found. A split into what Brouillet and Friston propose would still free neither CLAIM-337 nor CLAIM-439.

Backlinks