"AI brain rot" and cognitive offloading
ChatGPT made research feel easier but students' reasoning was weaker than with Google
In short: University students who researched a science question with ChatGPT felt less mental effort than students who used Google, but wrote recommendations with lower-quality reasoning and justification. Their answers were not more uniform than the Google group's.
| Paper | Cognitive ease at a cost: LLMs reduce mental effort but compromise depth in student scientific inquiry |
| Authors | Matthias Stadler, Maria Bannert, Michael Sailer |
| Published | November 2024 · Computers in Human Behavior 160 |
| Study type | Randomized experiment · Peer-reviewed |
| What we read | Abstract only. Summary based on the published abstract only. |
| Summarized |
What they did
Students were randomly assigned to use ChatGPT 3.5 or Google to research whether nanoparticles in sunscreen are a concern, and then write a recommendation with justifications. Researchers measured cognitive load and rated the quality and similarity of the recommendations.
Who took part
91 university students.
What they found
- The ChatGPT group reported significantly lower cognitive load.
- The ChatGPT group's recommendations showed lower-quality reasoning and argumentation.
- The two groups' recommendations were about equally varied.
Effect size: Effect sizes are not given in the abstract.
Limitations
- One task, one topic, one session, university students only.
- Used ChatGPT 3.5; newer models and study modes may differ.
- We read only the abstract.
What this means for learners and parents
- "That felt easy" is not a sign that you understood it. Some effort is part of learning.
- If you use AI to gather information, add a step where you write your own reasoning before reading its summary, or check it against sources.
Put it into practice (free tools)
Citation
Matthias Stadler, Maria Bannert, Michael Sailer (2024). Cognitive ease at a cost: LLMs reduce mental effort but compromise depth in student scientific inquiry. Computers in Human Behavior 160. https://doi.org/10.1016/j.chb.2024.108386
Related summaries
- Unrestricted ChatGPT help raised practice scores but lowered exam scores · Hamsa Bastani et al., 2025 · Preregistered randomized controlled trial
- "Your Brain on ChatGPT": what the MIT essay-writing study did and didn't show · Nataliya Kosmyna et al., 2025 · Lab experiment with EEG
- With AI, people scored better on reasoning puzzles but overestimated how well they did · Daniela Fernandes et al., 2026 · Two online experiments