The real finding in "Your Brain on ChatGPT" is buried in the final round of testing — and it changes everything

This article surfaces something the headlines missed: the MIT study's most important finding is not that AI harms thinking. It is that when you reach for it determines everything.In the final round of testing, the researchers switched the groups. Students who had built their own cognitive framework first — then brought AI in — showed stronger memory recall and re-engaged the brain regions tied to synthesis and judgment. Students who had offloaded from the start never owned the work, even when the machine was taken away.The study is a preprint — peer review pending — but the methodology is rigorous enough to take seriously.That is not an AI problem. That is a sequence problem.I spent over 14 years in Title I bilingual classrooms before moving into research. The struggle has always been part of learning — the dead ends, the tiredness, the breadcrumb trail that leads nowhere. That frustration is not a bug. It is the mechanism. What AI changes is the cost of the search phase. When used after the cognitive wrestling, it is an amplifier. When used instead of it, it is a shortcut that quietly empties the work of ownership.My recommendation: teach the order of operations explicitly. From day one. Even Pre-K. Their reasoning potential is entirely intact — what they need is context and guardrails, not restricted access.A five-year-old with access to AI will build a logical case for why donuts are the most important meal of the day. I have seen the reasoning. It is airtight. That is...