What question did this study set out to answer?

This research aims to explore the disconnect between how language models represent brands and how they make choices on behalf of users.

July 4, 2026Open Access

From Stated Perception to Revealed Choice: A Pre-Registered Instrument for the Choice-Perception Gap in AI Brand Perception (PRISM-C)

Key Points

This research aims to explore the disconnect between how language models represent brands and how they make choices on behalf of users.
Utilized PRISM-C, a pre-registered instrument, to assess eight dimensions of brand perception.
Conducted 1,144 counterbalanced trials across 40 brands and 40 scenarios involving four different model families.
Measured divergences between predicted picks and revealed choices using a systematic brand-level choice policy.
Mean pick unanimity among model families was .972, while divergence from predicted picks reached .633 on trials, significantly exceeding the measurement floor of .044.
Observed that stated dimensions influenced choices unequally, with some sign reversals in preferences.
Position effects emerged only when options were not easily distinguishable, emphasizing the systematic nature of choices.

Abstract

As language models move from describing brands to choosing them on users' behalf, an unexamined assumption carries the weight: that what a model says about brands predicts what it does. PRISM-C measures that assumption. The instrument elicits a model's stated eight-dimension brand readings, projects each buyer need into the same space, and compares the need-nearest predicted pick with the model's revealed pick in a simulated agentic choice task – with the divergence judged against the choice elicitation's own cross-family operator floor, so a gap is measured, never asserted. In the pre-registered campaign (40 brands, 40 scenarios, 1,144 counterbalanced trials across four model families), the families were near-unanimous with one another (mean pick unanimity .972; floor .044) yet diverged from the predicted pick on .633 of trials – fourteen times the floor. Choice weighted the stated dimensions unequally, with sign reversals; position effects appeared only when options were perceptually indistinguishable; and pre-registered contrasts ruled out presentation order and tie-breaking noise in favor of a systematic brand-level choice policy the stated profile only partially carries. Stated-perception measurement does not transfer to agentic choice for free; the wedge between them is now an instrument-measured quantity. Includes zharnikov-2026bb-from-stated-perception-to-revealed-choice.yaml (Paper Spec v0.1.0) – a machine-readable specification of the paper's claims, assumptions, and dependencies. The paper's full machine-first bundle (the SPINE claim/dependency graph and the ONTOLOGY term module) lives in the public repository; see https://github.com/spectralbranding/paper-spec for the standard. This PDF is generated programmatically from that machine-first source under a research-as-repository model.

Read Full Paperexternally

AI에게 질문

Bookmark

View Full Paper