AI Summary of Peer-Reviewed Research

This page presents an AI-generated summary of a published research paper. The original authors did not write or review this article. [See full disclosure ↓]

Publishing process signals: STRONG — reflects the venue and review process. — venue and review process.

Hybrid AI feedback prompted the most student revisions

A person from behind wearing a striped shirt types on a laptop computer displaying what appears to be a document or digital interface on its screen in an indoor setting.
Research area:Computer ScienceArtificial IntelligenceStudent Assessment and Feedback

What the study found

Hybrid AI-generated feedback prompted the most revisions, compared with directive feedback and metacognitive feedback. Confidence ratings were uniformly high, and resource quality outcomes were comparable across the three feedback types.

Why the authors say this matters

The authors conclude that AI can deliver feedback that balances clarity with reflection. They suggest hybrid approaches may be especially useful because they combine elements of directive and metacognitive feedback, although more work is needed to assess broader impact.

What the researchers tested

The researchers ran a semester-long randomised controlled trial with 329 students in an introductory design and programming course using an adaptive educational platform. Students were assigned to receive directive feedback, metacognitive feedback, or hybrid AI-generated feedback that blended both approaches.

What worked and what didn't

Revision behaviour differed by condition: the hybrid feedback prompted the most revisions. Confidence ratings did not differ meaningfully, as they were uniformly high, and resource quality was comparable across conditions.

What to keep in mind

The abstract does not describe detailed limitations beyond noting that more work is required to evaluate broader impact. The findings are limited to the course, platform, and feedback conditions studied.

Key points

  • Hybrid AI-generated feedback led to the most student revisions.
  • Directive and metacognitive feedback did not outperform the hybrid approach on revision behaviour.
  • Confidence ratings were uniformly high across all feedback conditions.
  • Resource quality outcomes were comparable across the three conditions.
  • The study was a semester-long randomised controlled trial with 329 students.

Disclosure

Research title:
Hybrid AI feedback prompted the most student revisions
Authors:
Omar Ali Saleh Alsaiari, Nilufar Baghaei, Jason M. Lodge, Omid Noroozi, Dragan Gašević, Marie Bodén, Hassan Khosravi
Publication date:
2026-02-10
OpenAlex record:
View
AI provenance: This post was generated by OpenAI. The original authors did not write or review this post.