Does cognitive offloading to AI look different in younger people, or are we just seeing less practice thinking without help?

This explores whether age changes how people hand their thinking over to AI tools. The short answer is yes, but the corpus has only two direct data points on age, and both need careful reading. The clearest one is a UK survey of 666 people. In it, frequent AI use went along with lower critical-thinking scores, and cognitive offloading (letting the tool do the mental work) explained much of that link. The pattern was stronger in younger participants Does frequent AI tool use reduce our critical thinking?. But it's a correlation from one survey, and education level also changed the effect. So 'younger' may partly stand in for 'less practised at doing this kind of thinking without help,' rather than marking something about youth itself.

The second data point looks at much younger users and turns up something surprising. In an fMRI study, six-to-seven-year-olds and adults wrote stories with ChatGPT, and the finished stories were rated equally creative. Their brains were working very differently, though: the children showed much weaker coordination in the networks adults use to steer attention and organize ideas Why do children's brains show weaker control networks during AI dialogue?. In other words, the output looked the same while the process behind it didn't. That ties to a broader argument in the collection that AI separates a polished product from the thinking that would normally produce it Does AI separate intellectual form from the thinking behind it?. For younger users, this means a good-looking result tells you very little about whether any skill is being built.

The adult research suggests why this might matter over time. A four-month EEG study found that brain connectivity dropped the more people relied on an LLM, and heavy users had trouble recalling work they had just done Does AI assistance weaken our brain's ability to think independently?. A review of 14 student studies found that over-reliance on AI chat tools pushes students toward quick gut judgments in place of careful reasoning Does relying on AI dialogue systems harm student critical thinking?. Neither study splits results by age. Together, though, they suggest offloading might do the most damage when habits are still forming.

Here is the turn you might not expect: age may not be the most important difference between people. An experiment on 'cognitive forcing functions' (design features that make people think before accepting an AI's answer) found that they mostly helped people who already enjoy effortful thinking. That means a fix meant to help everyone can widen the gap between people Do cognitive forcing functions help all decision makers equally?. On the hopeful side, a 704-person preregistered experiment found that simply telling people what offloading costs them cut their requests for answers in half and raised their scores on later tests taken without AI by 51%. A reward for effort did nothing Can metacognitive feedback stop students from offloading to AI?.

The corpus does not yet have a study that compares age groups directly, by tracking teenagers and older adults on the same task over time. What it does suggest is that the more useful question may not be 'are young people different?' It may be 'who knows what offloading costs them?', which can be taught.