In human evaluation, abstraction preference was associated with perceived response quality, and more complex questions generally received lower ratings¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? A linear mixed-effects model with personalization factors and Bloom's level as fixed effects and evaluator identity as a random effect found abstraction preference associated with perceived quality. → Saptarshi Basu 2026

Evidence¶

Saptarshi Basu 2026¶

Saptarshi Basu, Sandeep Kakar, Ashok Goel. (2026). A Prompt-Engineering Approach to Develop Scalable, Flexible, and Real-Time Hybrid Micro-Level Personalization in a General Purpose AI Teaching Assistant. https://arxiv.org/abs/2609.03402 doi:10.48550/arxiv.2609.03402

q2 · i? · design · r2

Human evaluation study with five evaluators from the KBAI course rating 13 profiles across six questions (390 evaluations) in Qualtrics on a 0-10 quality scale; baseline excluded, leaving 12 personalized profiles. Fixed-effect estimates appear in Figure 5.

"A linear mixed-effects model with personalization factors and Bloom's level as fixed effects and evaluator identity as a random effect showed that abstraction preference was associated with perceived response quality."

Discussion¶

Related Claims¶

• Abstraction and processing preferences, but not self-assessment, were significant predictors of perceived response complexity, abstraction, and processing style — related
• Higher self-assessment, verbosity, reflective processing, and technical abstraction are associated with greater response complexity, and Evaluation and Analysis questions tend to yield more complex responses — related
• Prompt-based personalization of an LLM/RAG teaching assistant produces measurable variation in response expression, length, and complexity across personalization conditions — a broader claim this one bears on