R9990/2026-03-31/C001/SRC07/E01¶
Peer-reviewed experiment showing standard interview questions disadvantage autistic candidates, with measurable improvement from adaptations.
URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC8108109/
Extract¶
Maras, Norris, Nicholson, Heasman, Remington, and Crane (2020) studied 50 adults (25 autistic, 25 non-autistic) in a controlled experiment:
Phase 1 (Standard interview questions): - Autistic interviewees received significantly lower ratings than non-autistic peers (M=3.41 vs M=3.91) - Employment experts rated autistic candidates less favorably on confidence, communication skills, likeability, and ease of working with
Phase 2 (Adapted questions — restructured for clarity, broken into sequential components, written printouts provided): - Both groups improved substantially - Autistic interviewees showed greater improvement, narrowing the performance gap - Question clarity ratings improved notably for autistic participants (3.31 to 4.40) - Both groups reported higher confidence
Key barriers identified: - Autistic individuals struggle to "infer what information the employer wants from an answer" to open-ended questions - Challenges with memory recall under time pressure and executive function demands - Reduced use of impression management tactics
Adaptations made: Requested specific information explicitly, broke multi-part questions into sequential components, provided written question printouts, clarified expected response structure.
Conclusion: "Employers should be aware that adaptations to job interview questions are critical to level the playing field."
Relevance to Hypotheses¶
| Hypothesis | Relationship | Strength |
|---|---|---|
| H1 | Supports | Controlled experiment demonstrating measurable disadvantage from standard interview formats for autistic candidates |
| H2 | Supports | The fact that adaptations (providing structure, clarity, written questions) significantly improved autistic performance suggests the problem is in the format, not the candidate — and is fixable |
| H3 | Contradicts | Statistically significant performance gap (3.41 vs 3.91) under standard conditions directly contradicts neutrality |
Context¶
This study examines employment interviews specifically (not forensic). The sample size (50) is modest but adequate for a controlled experiment. The adaptations tested are directly relevant to how STAR could be modified for neurodivergent candidates.
Notes¶
This is the strongest single piece of evidence in the collection because it is a controlled experiment with employment interviews, directly measuring the performance gap and demonstrating that structural adaptations reduce it. The finding that adaptations also helped neurotypical candidates supports universal design principles.