ChatGPT Impact on Brain Activity: Study Reveals Cognitive Effects

A recent study published on the preprint server arXiv reveals that using ChatGPT, a powerful AI tool developed by OpenAI, can significantly alter brain activity during writing tasks. The research, conducted by computer scientist Nataliya Kosmyna of the MIT Media Lab and her colleagues, measured brain-wave activity in university students as they wrote essays using different methods: with ChatGPT, with Google search, and with no internet access at all.

The study involved 60 students from five universities around Boston, Massachusetts. Each participant was asked to write a 20-minute essay on topics from the Scholastic Assessment Test (SAT). The students were divided into three groups: one used ChatGPT, another used Google search, and the third group wrote without any internet assistance. The researchers used electroencephalography (EEG) to monitor brain activity during the task.

The results showed that students who wrote essays without any external tools displayed the strongest and most widespread brain connectivity. These participants also performed better in recalling information from their essays when questioned later. The group that used Google search showed increased activity in brain regions associated with visual processing and memory. However, the ChatGPT group exhibited the least brain connectivity during the task.

Kosmyna emphasizes that more brain activity is not always better. It can indicate deeper engagement with a task, cognitive inefficiency, or even cognitive overload. The study does not suggest that using ChatGPT makes people 'dumber' or that their brains are 'on vacation.' Instead, it highlights the need for further research to understand the long-term cognitive effects of AI-assisted tasks.

The findings are part of a broader movement to assess whether AI is making us cognitively lazy. While the study is limited in scope and duration, it opens up important questions about the role of AI in education and cognitive development. Future research will need to explore whether habitual use of AI tools reshapes our thinking over time and how it affects brain activity in different contexts.

Kosmyna and her team are cautious about overinterpreting the results. They acknowledge that the study involved a small number of participants and a short time frame, and it does not provide conclusive evidence of long-term cognitive changes. However, the study's insights contribute to the ongoing debate about the impact of AI on human cognition and the need for responsible AI usage.