Medical tests  Scans (CT)

Brain Scans Show Whatever You Want to See

By reporter Malou van Hintum
first published in the Netherlands’ daily newspaper
De Volkskrant on April 18, 2008, and edited on April 23, 2008.
translated by Mira de Vries

We thank De Volkskrant for their kind permission
to translate and publish this article on our website.

Scans appear to be objective, but actually they aren’t so unambiguous at all, says Sarah de Rijcke. She is a PhD student at the University of Groningen’s department of Theory and History of Psychology.

“A brain scan is not a photograph, it is the result of complex computations which after various processes are converted to images by computers. Which processes are used depends on the choices made by scientists. Only then is an attractive, colorful picture for publication achieved.”

Brains seem to speak for themselves, but that isn’t true. Just as the old sketches and copper engravings of our brains carry the personal stamp of those who made them, so also are brain scans the result of a series of decisions and processes made by those who present them.
De Rijcke: “In their reports on their research, scientists thoroughly explicate their actions. But when a scan is presented, incertitude recedes into the background. The audience “sees” how the brain of a person with depression or schizophrenia deviates from a “normal” brain. But of course that can only be shown when it is agreed in advance what deviance is. This should always be kept in mind when viewing such a picture.”

Full Color
One full color brain scan is always the result of multiple scans, it is an average, she explains. “Individual brains aren’t so different from each other. The brains of a particular schizophrenia patient, someone suffering from depression, a heterosexual or a man, have many more similarities than differences with respectively the brains of a healthy person, a homosexual or a woman. But if you’re determined to find differences, you’ll find them. It’s a matter of laying a whole lot of people under the scan and looking very closely.”

Those kinds of differences are what scientists love to find, and the public at large is engrossed. De Rijcke: “Increasingly we try to zoom in on the material origins of illnesses like for instance schizophrenia, depression, and autism. Seeing something in the brain forms “hard” evidence that something is wrong. It’s real, it’s been measured. “Soft” explanations based on social factors and environmental influences are literally effaced from the picture – whereas they may be involved at least as much. Perhaps such factors influence the brain to show up a certain way on the scan; and the image reflects not the origin of the problem but its result.”

People are immensely sensitive to visual representations; your eyes don't lie to you. But machines have to be adjusted, and the programs they use are thought up by humans. Objective measurements are no more than agreements that can potentially be revised. Never are all the data reflected, because that would cause too much “interference”. And who decides what is “interference” and what is “genuine”?
There’s another point regarding mental illness. “A tumor may be a tumor, no need for discussion,” says De Rijcke. “But when is someone mentally ill? That depends on the diagnosis. It’s a matter of continual agreements. Just think of the criteria laid down in the DSM, the diagnostic manual for psychiatry. Those criteria are revised every so-many years. A revision is often a matter of expanding the classification.”

With that De Rijcke isn’t saying that mental illness is being over-diagnosed. “No doubt all sorts of behaviors that used to be misunderstood are now recognized as illness. But it does illustrate that research into disease states is based on agreements that cannot be characterized as absolute.”

Neuroscientists are well aware of the complex nature of their research, says De Rijcke. “Many scan studies are incomparable due to the variation in researched areas of the brain, scanner types, and the sorts of subjects who were studied. Furthermore, some data are processed more or differently than others. So sharing research results is complicated.”

Pressure to Publish
The pressure to publish is another problem that plagues scientists. De Rijcke: “Of course you have to achieve that your pretty picture makes it to for instance the cover of Nature. In addition, money is a major factor in drawing attention to research. Expensive scanners and high research budgets have to produce attractive results.
Researchers themselves are usually quite careful, says De Rijcke. “I’m more concerned about others, who have exaggerated faith in the power of proof provided by brain scans and run away with them. Here it’s not occurring yet, but in the United States brain scans are employed in legal trials. You can bring an expert on stage who puts the picture in balance, but such a scan remains visually compelling with a persuasive power of its own. I consider that a dangerous development.

“The context in which the images are made must continue to be taken into consideration when interpreting them. It’s questionable whether someone’s mental illness or criminal insanity can be caught on camera.”