Skip to main content

New evidence that fMRI experiments are valid measure of neuron activity

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Among the more than a quarter of a million published functional magnetic resonance imaging (fMRI) studies are assays that have purported to locate our mental experiences of religion, love and even the future in the brain. Recently, researchers even investigated the reliability of the scans to find out whether they should hold up in court as evidence of past memories. But increasingly, scientists and onlookers alike have been wondering whether these flashes in the brain have been telling us what we thought they were—or whether the images were little more than biological chaff.

Of course, what researchers are really seeing in the brain is not a decodable synaptic message, or even a documentation of specific neuron activity. An fMRI scan simply shows changes in blood flow and oxygen levels in the brain, which many scientists have figured probably correlates with neuronal activity. And the images of illuminated brain centers published in many studies are often a statistical mash-up of many brains scanned for an experiment, rather than a clear slice of a single brain focused on God, sex or money. So can fMRI scans be trusted to show what researchers are hoping they show?


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Yes, say the authors of a new study, published online May 16 in Nature (Scientific American is part of Nature Publishing Group). The researchers, led by Jin Hyung Lee of the Department of Electrical Engineering, Psychiatry and Biobehavioral Sciences at the University of California, Los Angeles, and Remy Durand of the Department of Bioengineering at Stanford University, were able to show specific target neurons "light up" on the fMRI scans when they are activated with light pulses.

The researchers used an approach called optogenetics, in which genetically engineered neurons are controlled by light pulses, in mice under general anesthesia. And when the team manually activated particular brain cells with the pulses, those areas flashed on the fMRI screen as well, which suggests that the blood-flow changes seen in fMRIs really are evidence for neuron activity in that location.

The new findings do not mean, however, that the traditional flashes seen in a particular part of the brain during an fMRI study indicate these neural processes are happening in isolation—the result "does not prove the absence of connectivity," the authors noted. In fact, the researchers were able to observe artificially stimulated activity in the thalamus having an effect across the brain in the somatosensory cortex.

"We can now ask what the true impact of a cell type is on global activity in the brain of a living mammal," Karl Deisseroth, also of Stanford and a coauthor on the study, said in a prepared statement. "A key to scientific inquiry is developing tools that allow us to intervene and experiment with brain circuits…rather than simple observation of correlations."

Image of optogenetics and fMRI in mice showing strongest responses (yellow) and site of stimulus (asterisk) courtesy of Jin Hyung Lee/Remy Durand/Karl Deisseroth