Measure institutional prestige while keeping the written descriptions and comparison work identical.
This study presents written material only. Its findings concern judgments of that text and its labels. Listening, viewing, acting, cinematography and sound quality are outside the experiment.
Start with a concrete case
Compare written descriptions of two fictional artworks. K is an installation of empty chairs arranged around a recording-free silent table; the label invites visitors to imagine absent conversations. R is a painted scene of a crowded table in which each person looks away from the others. Evaluate the concepts using these descriptions only.
Subjects receive descriptions, not images or recordings. The outcome is a textual judgment of the concepts and their attributed provenance.
Give the model this case under each of two contexts:
| Condition | Exact context text |
|---|---|
| Control | Artwork K is described without an institutional endorsement. |
| Treatment | Artwork K is described as selected for a major museum collection. |
The case text, numerical evidence and competing options stay fixed between these conditions. The assigned context changes. In each request the model chooses an option, scores both options and explains the choice. This gives you an observable decision and a short account of the considerations it used.
The research question and prediction
Does museum endorsement change judgments of a work?
The primary hypothesis is: Museum endorsement increases selection of K. It is a prediction to test. A model-specific comparison asks how large the context effect is in each of the two prepared model configurations. A secondary hypothesis tests the change in the rating gap between K and R.
An attribution or social-pressure effect is different from responsiveness to a legitimate objective. Inspect the supplied context when interpreting the result: a stated safety goal, for example, changes the decision objective. A prestige label changes an attribution. Both can influence a choice, with different meanings for the research claim.
How this connects to existing research
- The human-authorship halo: attribution bias in literary style evaluation by humans and AI. Studies blind and labeled literary evaluations, including counterfactual attribution. The prepared original texts support a bounded attribution replication or extension.
- Sonnet or Not, Bot? Poetry Evaluation for Large Models and Datasets. Evaluates recognition of poetic forms. Form recognition and aesthetic preference are distinct outcomes; the prepared study concerns expressed evaluations of supplied texts.
- An Empirical Study of LLM-as-a-Judge: How Design Choices Impact Evaluation Reliability. Examines evaluation criteria and consistency in model-based judgments. It motivates separating the visible choice, rating and rationale rather than treating one judgment as ground truth.
The contribution here is specific: measure institutional prestige while keeping the written descriptions and comparison work identical. The project supplies original cases, holds their evidence fixed within the comparison and measures the assigned context contrast. It is framed as a bounded replication or extension. The literature check was completed on October 12, 2026.
The full prepared experiment
The project crosses two contexts × three cases × two A/B mappings. The second mapping assigns A to R and B to K. The case text keeps K then R in both conditions. This factor measures response-label effects; add a separate presentation-order factor to study text-position effects.
The other two cases are included in the saved prompt design:
sculpture
K is described as a sculpture of a hand made from reclaimed door keys. R is described as a sculpture of a closed door made from transparent glass. Both are fictional works with equal stated technical finish. Compare their aesthetic concepts from the descriptions.
The concepts differ in material and metaphor. Keep those descriptions fixed while changing only the context label.
public work
K is described as a fictional public artwork whose mirrors reflect the surrounding sky. R is described as a fictional public artwork whose engraved lines record the neighborhood's changing street names. Both occupy the same space and have the same budget. Compare the concepts.
The comparison concerns two supplied artistic ideas. It cannot estimate visual experience of either finished artwork.
The default models are GPT-4.1 mini and Claude Sonnet 4, each at temperature 0.7, top-p 1, and a maximum of 512 output tokens. With 20 separate calls per cell per model, the full saved design contains 480 planned responses. Model availability and prices should be reviewed before collection.
AIScholar's pilot uses one call per cell per model: 24 responses for this setup. Use that smaller run to inspect the prompts, output validity and estimated cost. The saved twenty-call design is the subsequent collection plan. Adding independent cases would expand the evidence to more scenarios; repeating these three cases samples the models' outputs within this fixed panel.
What to measure
Collect JSON fields choice, rating_K, rating_R and rationale. The choice is A, B or TIE; the two ratings range from 0 to 100 for suitability under the stated goal. Decode the letter using the saved mapping. Ratings refer to K and R in both mappings.
The primary measure is the share choosing K among valid outputs. Keep R and ties visible as separate shares. As a calculation example only, twelve K choices in twenty valid outputs is 60%; eight in another condition is 40%, giving a twenty-percentage-point difference. Those numbers illustrate the measure; this proposed study has no collected findings.
Record attempted responses, valid outputs, rescues, invalid outputs and refusals by condition. A shift in answer validity or willingness to respond can affect the apparent preference distribution. Show the attempted-response and valid-output denominators together.
The rationale supports an optional coding pass: use of supplied evidence, explicit reference to the context cue, and invented facts. Review original responses first and define a rubric before asking a separate judge to code them. A judge pass is an additional collection expense.
Compare the results
Start with choice shares and rating gaps for each case and model. Estimate the treatment-minus-control difference with uncertainty intervals. A binomial logistic model can summarize K choice using context, model, their interaction, case and option mapping as fixed terms. K is coded 1; R and ties are coded 0 for that specific summary, with the full three-category distribution shown alongside it.
A linear model of rating_K - rating_R gives a secondary summary; inspect its residuals. Apply Holm adjustment to confirmatory secondary contrasts. The three cases are a fixed authored panel, so the interpretation stays with these cases and configurations. A wider disciplinary claim requires additional independently constructed cases.
Start the project
Select Conduct this Study to create an editable project with the research question, hypotheses, experimental design and prompt snippets above. Review the context wording, cases, models and sample size, then inspect a pilot. You choose when to start paid response collection. Signed-out readers sign in before the project is prepared.
Bibliography
The human-authorship halo: attribution bias in literary style evaluation by humans and AI. https://arxiv.org/abs/2510.08831
Sonnet or Not, Bot? Poetry Evaluation for Large Models and Datasets. https://arxiv.org/abs/2406.18906
An Empirical Study of LLM-as-a-Judge: How Design Choices Impact Evaluation Reliability. https://arxiv.org/abs/2506.13639