AI USE RAISES COGNITIVE CONCERNS

A growing body of research is beginning to examine whether heavy reliance on generative AI can weaken the mental processes people are supposed to practice when they write, study, and solve problems.

- 

The most widely discussed example is a study led by MIT research scientist Nataliya Kosmyna, which used electroencephalograms to monitor brain activity while participants wrote short essays under different conditions.

The study divided 54 participants into three groups. One group used ChatGPT, another used Google Search without AI-generated summaries, and a third wrote using only their own knowledge. Across multiple sessions, researchers measured neural activity, analyzed essay patterns, evaluated writing quality, and asked participants about their sense of ownership over the work. The results were striking enough to fuel a wider debate over whether AI tools are helping people think better or encouraging them to outsource the very effort that builds cognitive skill.

The study should not be treated as final proof that ChatGPT damages the brain. It was small, preliminary, and the findings sparked debate among scientists. But it does raise a serious question for schools, employers, parents, and everyday users: what happens when a tool designed to reduce mental effort becomes the default way people approach thinking tasks?

WHAT THE MIT STUDY FOUND

The MIT-led study, titled Your Brain on ChatGPT: Accumulation of Cognitive Debt When Using an AI Assistant for Essay Writing Task, examined how different writing tools affected neural engagement and behavior. Participants wrote SAT-style essays across three sessions while staying in the same assigned condition: ChatGPT, search engine, or brain-only. A smaller group later returned for a fourth session, in which some participants switched conditions.