By Nicolas Hulscher, MPH
A new MIT study titled, Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task, has found that using ChatGPT to help write essays leads to long-term cognitive harm—measurable through EEG brain scans. Students who repeatedly relied on ChatGPT showed weakened neural connectivity, impaired memory recall, and diminished sense of ownership over their own writing. While the AI-generated content often scored well, the brains behind it were shutting down.
The findings are clear: Large Language Models (LLMs) like ChatGPT and Grok don’t just help students write—they train the brain to disengage. Here’s what the researchers found:
Brain Connectivity Declines with AI Use
EEG scans revealed a systematic scaling down of neural connectivity in the brain with increasing reliance on external tools: Brain-only group: strongest, most widespread connectivity. Search Engine group: intermediate. LLM group: weakest connectivity across alpha, beta, delta, and theta bands.
LLM use resulted in under-engagement of critical attention and visual processing networks, especially in Session 4 when participants tried to write without AI.
LLM Users Forget What They Just Wrote
In post-task interviews: 83.3% of LLM users were unable to quote even one sentence from the essay they had just written. In contrast, 88.9% of Search and Brain-only users could quote accurately.
0% of LLM users could produce a correct quote, while most Brain-only and Search users could.
AI Use Disrupts Memory and Learning Pathways
... continue reading