MIT study finds ChatGPT use may reduce critical thinking and brain engagement
A new MIT Media Lab study has hit a nerve: 54 participants wrote SAT-style essays using ChatGPT, Google Search, or nothing at all, while EEG sensors recorded their brain activity.
The result? ChatGPT users showed the least brain engagement—neural signals tied to creativity, memory, and focus were far weaker than in the other groups.
Over months, ChatGPT users grew “lazy”—by the final essays, many were copy-pasting content and showed reduced brain connectivity. Meanwhile, the brain-only group maintained strong neural engagement and originality. Search users fell somewhere in the middle, signaling that using a search engine still fosters more active thinking than AI chat tools.
Alarmingly, when ChatGPT-dependent participants later had to write without AI, their brain activity remained unusually low—suggesting they hadn’t internalized or memorized the material.
Nataliya Kosmyna, the paper’s lead author, warns: we risk letting future generations “do GPT kindergarten,” sacrificing cognitive development for convenience.
Rysysth insights:
AI convenience shouldn’t come at the cost of deeper thinking. This study signals an urgent issue: while LLMs like ChatGPT optimize for speed and ease, they may also cultivate “cognitive debt”—weaker memory, creativity, and engagement.
The key isn’t to ban AI, but to design systems and learning practices that encourage active thinking alongside AI.
In practice & policy:
- Encourage mixed methods in education: supplements like Search, not substitutes
- Build educational AI that prompts user reasoning—AI tutors that ask “why?” not just “what?”
- Schools and policymakers must prioritize critical gaps between quick answers and deep comprehension before embracing AI-first instruction
If we want AI to empower rather than erode our minds, we need smarter interfaces and mindful usage. It’s not just about having better tools—it’s about keeping our brains engaged.
Until next time.