As cited second-hand in an AI-literacy framework paper: ChatGPT use improved short-term essay scores but triggered metacognitive laziness with no significant knowledge-transfer gains¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person

Subclaims¶

q2 i? The article attributes to Fan et al. (2025) a randomized experiment in which ChatGPT use improved short-term essay scores but reduced monitoring, evaluation, and orientation behaviors, with no significant gains in knowledge transfer. → Liu 2026

Evidence¶

Liu 2026¶

Liu, J. P., & Levy, R. (2026). Beyond Tool Adoption: A Practical Five-Stage Developmental Continuum for AI Literacy in Higher Education. North Carolina State University. https://arxiv.org/abs/2606.00038 doi:10.48550/arxiv.2606.00038

q2 · i? · design · r2

A conceptual framework paper (a five-stage AI-literacy continuum, with a design-based observational implementation case), not a review; it reports this finding second-hand in its discussion of Stage 1 concerns, citing Fan et al. (2025); the underlying randomized experiment was not read directly, so design details beyond those printed are not described.

"Fan et al. (2025) demonstrated in a randomized experiment that ChatGPT use improved short-term essay scores but triggered "metacognitive laziness," with students showing reduced monitoring, evaluation, and orientation behaviors and no significant gains in knowledge transfer."

Discussion¶

Learner Variables¶

• Self-Regulation — outcome: instruction changes it
• Metacognition — outcome: instruction changes it

Related Claims¶

• Evidence suggests over-reliance on generative AI may negatively affect students' self-regulated learning behaviors, potentially harming retention and transfer to novel contexts — the general claim this finding bears on, with the studies behind it
• Some GenAI tools can inhibit productive struggle by reducing cognitive effort (metacognitive laziness) — a broader claim this one bears on
• The availability of AI assistants improved the quality of submitted admissions essays, particularly mechanical aspects and especially for international applicants — related
• Robotic projected AI assistance yields 60% higher short-term learning transfer than screen-based ChatGPT once assistance is withdrawn — related
• GenAI use collapses SRL cycle transitions, with metacognitive activity restructuring around the tool — possibly the same claim (merge candidate)
• AI's direct impact on SRL concentrates on metacognitive and cognitive aspects while motivation remains underexplored — related
• Personalized ChatGPT-supported feedback in an augmented-reality quantum physics laboratory improved learning outcomes and directed visual attention — related
• Product-outcome studies report GenAI-related gains on product-quality measures but do not assess durable, transferable learning — related
• Learning outcomes differ by how students use AI: augmentation and mixed users show gains in both sessions, while automation users' Session One essay advantage is absent from their unaided essays — related
• Novice programmers' GenAI use produces a widening gap in outcomes by metacognitive skill (attributed to Prather et al.) — related
• Frequent generative AI use for essay writing can foster metacognitive laziness, reducing intrinsic motivation and learning autonomy — possibly the same claim (merge candidate)
• The review reports AI-driven personalization improves student engagement and learning outcomes, and virtual teaching assistants support diverse learning needs — related
• Reflective strategies such as cognitive forcing, pre-commitment, and uncertainty communication reduce overreliance on AI and improve decision quality, but have been evaluated only as short-term single-session interventions — related
• Evidence on GenAI in chemistry education is concentrated in short-term outcomes, with durable transfer and retention rarely tested — a broader claim this one bears on
• An efficiency paradox: well-intentioned students default to passive GenAI processes and experience cognitive laziness and overreliance — a broader claim this one bears on
• The customized agent's advantage in short-term task improvement emerges under directive feedback but not under Socratic feedback — related
• AI-assisted task performance can rise without corresponding gains in underlying knowledge, metacognitive processing, or independent reasoning — possibly the same claim (merge candidate)
• Purpose-built AI chatbot coaching is generally well received and supports short-term motivation and selected behavior change, but effects on sustained, meaningful outcomes are inconsistent — related
• GCLA does not bolster motivational aspects of SRL: traditional ChatGPT users showed higher intrinsic motivation and higher amotivation than GCLA users — related
• Documented GenAI-related challenges to active learning include cognitive debt, metacognitive laziness, unreflected acceptance, and overreliance on GenAI access — a broader claim this one bears on
• Rule-triggered pedagogical agents increased SRL activities and learning outcomes relative to untriggered agents — related
• AI-assisted practice applications improved music students' performance outcomes, self-efficacy, and self-regulated learning — related
• Socratic feedback is associated with higher task orientation and comprehension monitoring than directive feedback; on feedback regulation it has no main effect, but its effect depends on agent type — related
• GenAI's default output on a take-home exam question remained at the quantitative, multistructural SOLO phase, producing polished but pedagogically thin responses — related
• Effective AI use positively influences sustainable learning performance — related
• The paper reports that metacognition—students thinking about how they understand their own learning—supports more efficient and effective learning and improved outcomes — related
• Customized AI writing systems show benefits in writing quality, learning processes, assessment reliability, and feedback literacy — related
• The effectiveness of reflective AI-engagement mechanisms varies across users, depending on Need for Cognition, expertise, domain familiarity, and task difficulty — related
• GCLA use improves cognitive and behavioral engagement and self-efficacy compared with traditional ChatGPT use in blended learning — related
• Metacognitive analysis activities were positively associated with transfer, but metacognitive activities showed no association with essay quality — related
• Interviews show students used scores to trigger action but reflected shallowly, derived planning from previous scores, and diverged emotionally — related
• AI over-reliance can produce inversion effects, reducing cognitive processing depth and learning outcomes — possibly the same claim (merge candidate)