You've seen those colourful brain pictures. Red and yellow blobs lighting up when someone thinks about food, feels sad, or solves a puzzle. Scientists point to them and say "this is the part of your brain that does X." Newspapers love them. TED talks love them. They make the brain look like a neat machine with labelled buttons.
A new study says a big chunk of those pictures may not mean what we thought: https://www.nature.com/articles/s41593-025-02132-9
The common type of brain scan used in research is called fMRI. It does not watch neurons firing directly. It watches blood flow. The idea was simple: when brain cells work harder, they need more oxygen. Blood rushes in to deliver it. More blood flow = more activity. Brighter blob on the scan = that area is "working."
That assumption has been the foundation of thousands of studies for decades
Researchers in Germany looked closer. They found that in some important parts of the brain, especially the "default mode network" (the parts active when you're daydreaming, remembering, or just letting your mind wander), the neurons can ramp up their activity by simply pulling more oxygen out of the blood that's already there. Blood flow does not always increase.
In those cases the scan stays quiet even though the brain cells are busier. The study estimates this happens with roughly 40% of the signals scientists have been reading.
In other words: the machine is measuring blood movement, not brain activity itself. And the link between the two is not as reliable as everyone assumed.
A huge amount of modern neuroscience rests on these scans. Studies claiming to show which brain areas are linked to depression, anxiety, ADHD, decision-making, love, fear, or political opinions often rely on fMRI. If the blood-flow signal is missing or misleading in many cases, those maps become shaky.
This does not mean every brain study is rubbish. Some findings are solid and have been checked with other methods. But it does mean a lot of the colourful "this is your brain on X" claims need a second look. What looked like clear evidence may have been incomplete or, in some regions, simply wrong.
Science advances by finding its own mistakes. The brain is the most complicated object we know. It was always optimistic to think we could map its activity so cleanly with a machine that mainly watches blood.
The new finding is a reminder: pretty pictures are not the same as solid understanding. Next time you see a glowing brain scan in a headline, it is fair to ask how sure we really are about what it is showing.
The good news is that better tools and more careful thinking are already being developed. The bad news is that a lot of what we thought we knew about the mind may need rewriting. For science sceptics we should enjoy this humbling of the neuroscientists, a particularly arrogant bunch anyway.