AI Science Tracker
EN

GPT-4o’s gender associations for inattentive ADHD changed with cue wording

Aarhus University · 24 Oct 2026 · computer science

Discovery

GPT-4o’s gender associations for inattentive ADHD changed with cue wording

Source check: Source checked. Not independent scientific confirmation.

Finding: In a controlled association task, female-associated responses for inattentive ADHD prompts fell from 80.4% with explicit gender categories to 38.5% with implicitly gendered names. The study measures GPT-4o’s choices under these prompt conditions. It did not test patient diagnosis or estimate ADHD prevalence.

Technology: GPT-4o; Context Association Test

Family GPT-4: profile · models page

Publication: Preprint · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
GPT-4o assigned gendered associations to research prompts using differently expressed cues under a Context Association Test.
Test and result
The study compared explicit and name-based cue lists on researcher-written contexts, including a control category and an auxiliary clinical-sample proxy.
Limitation / still unknown
Effects may depend on the model, names and prompts. The clinical sample is selective and cannot establish true population prevalence.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/abs/2610.33697v1 · https://arxiv.org/html/2610.33697v1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: linked separately.

Review date: 21 Nov 2026. Review scope: v1 cue-list methods, reference sample notes and Study 1 results.

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse computer science discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.