Secondary students most frequently use generative AI for step-by-step theoretical explanation and for checking their own solutions, more than for calculations¶

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

Subclaims¶

q2 i? The most frequently reported task use is obtaining step-by-step explanations of theory (n = 265), followed by checking correctness of own solutions (n = 225) and analysing the task and solution process (n = 202), with calculations (n = 91) and algorithmic tasks (n = 79) reported less often. → Martin Lnenicka and Petr Coufal 2026

Evidence¶

Martin Lnenicka and Petr Coufal 2026¶

Martin Lnenicka and Petr Coufal. (2026). Navigating AI in STEM: what secondary students actually do with generative AI-driven tools. International Journal of STEM Education. https://doi.org/10.1186/s40594-026-00637-8

q2 · i? · associational · r2

Survey item B4 on task types in the n = 416 sample. Students reported a "strong preference for using generative AI-driven tools as conceptual and procedural learning support"; printed counts include explanation n = 265, checking solutions n = 225, analysis n = 202, calculations n = 91, algorithmic tasks n = 79, practice examples n = 90.

"The most frequently cited use was obtaining a step-by-step explanation of the theory ( n = 265), indicating that these tools function primarily as an instrumental scaffold for clarifying complex concepts in mathematics, computer science, and natural sciences."

Discussion¶

Related Claims¶

• Participants used the Flowcode LLM across all stages of their design process, most frequently asking for code explanations — related
• Student surveys show widespread AI use, desire for instructor guidance, and 'checking correctness' as a frequent use — related
• Reviewed literature reports risks of overreliance, AI errors in complex tasks, and students' uncritical acceptance of AI-generated outputs — related
• Students used visualizations to verify their own reasoning rather than as primary explanation tools, valuing control over their cognitive effort (agency) — related
• EAI effectiveness was reported across all educational stages, with most studies in secondary education and university — related
• Students found using a chatbot to solve a photoelectric-effect problem and evaluating its answer insightful, and the authors propose group checking as a way to verify chatbot answers — related
• Students most often report hybrid interaction modes with generative AI, with passive task delegation and verification-only use as smaller patterns