Unstructured LLM use risks cognitive offloading that displaces effortful thought and weakens argument construction¶

Claim · All claims
Evidence · 1 study · theoretical r2 · q1
Currency · newest study 2026 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person

Subclaims¶

q1 i? In one study the paper cites (Stadler et al., 2024), students relying on LLMs for inquiry tasks reported lower cognitive effort and produced weaker arguments than peers using traditional search. → Mireia Vendrell 2026

Evidence¶

Mireia Vendrell 2026¶

Mireia Vendrell, Samantha-Kaye Johnston. (2026). Scaffolding critical thinking with generative AI: Design principles for integrating large language models in higher education. Computers and Education: Artificial Intelligence. https://doi.org/10.1016/j.caeai.2026.100572

q1 · conceptual paper · i? · theoretical · r2

A conceptual, normative and design-oriented paper that states it does not report an empirical study. It cites Stadler et al. (2024) as comparing students who used LLMs for inquiry-based tasks with peers who used traditional search engines: the LLM users reported "lower cognitive effort and produced weaker arguments". The paper does not describe Stadler et al.'s design; the finding is that study's, reported second-hand here.

"Stadler et al. (2024) found that students using LLMs for inquiry-based tasks reported lower cognitive effort and produced weaker arguments, suggesting a trade-off between fluency and depth."

Discussion¶

Related Claims¶

• Direct GenAI-supported feedback improves immediate argument-quality gain over peer feedback, while reflective and hybrid designs outperform direct GenAI feedback — the general claim this finding bears on, with the studies behind it
• Intensive LLM use in educational contexts may lead to an accumulation of cognitive debt that weakens critical thinking and epistemic ownership — related
• Cognitive offloading to generative AI is associated with reduced learner engagement, lower critical thinking, and lower-quality reasoning — related
• Reviewed literature reports risks of overreliance, AI errors in complex tasks, and students' uncritical acceptance of AI-generated outputs — related
• Immediate LLM explanations risk bypassing the generative processes required for durable understanding — related
• An effort-based reward did not significantly reduce answer offloading or improve unaided performance — related
• Some AI tools can inhibit productive struggle by reducing cognitive effort when they automate processes students would otherwise reason through — related
• AI roles in learning communities risk passivity, diminished authenticity, and overdependence if reflective inquiry is bypassed — related
• Unstructured LLM use contributes to epistemic narrowing by privileging convergent outputs and dominant viewpoints — related
• Delegating intellectual work to generative AI risks eroding epistemic agency through reduced critical effort and cognitive offloading — possibly the same claim (merge candidate)
• Some GenAI tools can inhibit productive struggle by reducing cognitive effort (metacognitive laziness) — related
• AI takeover of metacognitive monitoring may foster illusion of competence (untested conjecture) — related
• LLM-supported socio-scientific inquiry lowered cognitive load but produced lower-quality reasoning and argumentation than conventional search — a narrower finding that bears on this claim
• Learners who construct good single arguments individually are not always able to apply that competence in comparable collaborative tasks — related