Cognitive partnerships with AI enable students to engage in theoretical and empirical reflection that the authors characterize as genuinely novel epistemic practices¶

Claim · All claims
Evidence · 1 study (2 entries) · qualitative r3 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Students use AI to generate alternative perspectives and evaluate their work against subject-specific concepts, engaging in theoretical reflection on their own work. → Dalsgaard 2026 q2 i? Students use AI to identify and explain weaknesses in their work, demonstrating empirical reflection that is difficult without external input. → Dalsgaard 2026 (2)

Evidence¶

Dalsgaard 2026¶

Dalsgaard, C., & Prilop, C. N. (2026). AI as a cognitive partner in learning: Emerging epistemic practices of reflective thought. Journal of Research on Technology in Education. https://doi.org/10.1080/15391523.2026.2695013

q2 · i? · qualitative · r3

Interpretive finding from the interview analysis: students' inspiration and improvement practices prompt theoretical reflection. The authors state these reflection processes are "difficult, if not impossible, to perform without external input"; no effect size is reported.

"Theoretical reflection occurs when students use AI to generate ideas for alternative perspectives to incorporate into their task solutions, prompting them to engage in theoretical reflection on their own work."

Dalsgaard 2026 (2)¶

Dalsgaard, C., & Prilop, C. N. (2026). AI as a cognitive partner in learning: Emerging epistemic practices of reflective thought. Journal of Research on Technology in Education. https://doi.org/10.1080/15391523.2026.2695013

q2 · i? · qualitative · r3

Interpretive finding from the same thematic analysis: students' shortcomings-identification practice demonstrates empirical reflection. The article reports this qualitatively, with no effect size.

"Students also demonstrate empirical reflection when they use AI to identify and explain weaknesses in their work or to suggest adaptations to elements of their task solutions."

Discussion¶

Related Claims¶

• LEARN reconceptualises scaffolding for GAI contexts as interactive, dialogic, and epistemically regulated interrogation of AI outputs — related
• Calibrated epistemic vigilance, given adequate prior knowledge, is the binding constraint on productive augmentation with AI — related
• Collaborative testing shows the taxonomy enables comparison between similar GenAI activities while inevitably falling short of portraying all entangled aspects — related
• An epistemological account of human-GAI engagement lays a foundation for alternative approaches to teaching, assessment, professional development and AI governance — a broader claim this one bears on
• Empirical studies across diverse learner populations show that people's epistemic aims, ideals, and processes are tailored to the information sources and tasks they engage in — related