http://www.theguardian.com/science/2013/nov/25/could-a-brain-scan-diagno…
Society is intrigued by psychopaths, at least from a distance. Hollywood paints them as powerful and emotionless predators – a small few who have embraced their inner dark passengers. Whether it is Gordon Gekko, Catherine Tramell, or Dexter Morgan, the mythology of the psychopath is captivating.
Psychopathy is of course a very real disorder and a lot more complex than portrayed on film. For many years, the gold standard for diagnosing psychopathy has been various forms of behavioural assessment. But now, Californian neuroscientist James Fallon claims he can diagnose psychopathy from a brain scan. Last week the Smithsonian blog quoted him as saying:
“I was looking at many scans, scans of murderers mixed in with schizophrenics, depressives and other, normal brains. Out of serendipity, I was also doing a study on Alzheimer’s and as part of that, had brain scans from me and everyone in my family right on my desk. I got to the bottom of the stack, and saw this scan that was obviously pathological.”
What Fallon found was his own scan, which appeared to show reduced activity in a part of the brain associated with empathy. Based on this, and some genetic tests, Fallon concluded that he himself was a psychopath (just one of the “good” ones).
This isn’t the first time we’ve heard from Fallon. In addition to the fact that his claims haven’t been published in peer-reviewed journals, here are three reasons why we should take what he says with a handful of salt.
If all ravens are black then all black birds must be ravens, right?
One of the most obvious mistakes in Fallon’s reasoning is called thefallacy of reverse inference. His argument goes like this: areas of the brain called the ventromedial prefrontal cortex and orbitofrontal cortex are important for empathy and moral reasoning. At the same time, empathy and moral reasoning are lost or impaired in many psychopaths. So, people who show reduced activity in these regions must be psychopaths.
The flaw with this argument – as Fallon himself must know – is that there is no one-to-one mapping between activity in a given brain region and complex abilities such as empathy. There is no empathy region and there is no psychopath switch. If you think of the brain as a toolkit, these parts of the brain aren’t like hammers or screwdrivers that perform only one task. They’re more like Swiss army knives that have evolved to support a range of different abilities. And just as a Swiss army knife isn’t only a bottle opener, the ventromedial prefrontal cortex isn’t only associated with empathy and moral judgements. It is also involved indecision-making, sensitivity to reward, memory, and predicting the future.
If your friend walked into the room and took out an (unopened) Swiss army knife, could you tell how she was planning to use it? By the same token, changes in brain activity, on their own, tell us very little about cognitive abilities.
Brain activity can be an innocent bystander
We all know that just because two things are associated with each other doesn’t mean one of them caused the other. Otherwise we’d prosecute the witnesses of crimes and eat nothing but chocolate.
In neuroscience, confusing correlation with causation is close to unforgivable. Following Fallon’s example, suppose we were to find that psychopaths, on average, show reduced activity in a particular brain region compared with a healthy control group. What would that mean, exactly? Maybe the reduced activity caused psychopathy. Or maybe it was the symptoms of psychopathy that caused changes in that part of the brain. Or maybe the brain activity is completely unrelated to psychopathy – a mere witness to the crime. The only way to tell which is true would be to change activity in that part of the brain and see whether doing so changes psychopathic behaviours.
Seek and ye shall find
Fallon apparently began his investigation in earnest after learning that his family tree contained a number of murderers, including the infamous Lizzie Borden. He then sought evidence to confirm his belief that he inherited a psychopathic profile.
What’s wrong with doing that? As Francis Bacon put it nearly 400 years ago, “The human understanding when it has once adopted an opinion…draws all things else to support and agree with it.” If we only seek to confirm rather than falsify our beliefs then we will find that we are always right – or at least it will seem that way. This confirmation bias is one of the most powerful traps in reasoning because we all like to be right, and we prefer to be consistent. By consistently interpreting weak evidence in favour of his beliefs, Fallon’s investigation is a case study in bias.
Understanding the neuroscience of psychopathy is a fascinating and important branch of psychiatry. But as in all sciences, real advances require a self-critical mindset and a strict adherence to the scientific method. As scientists seeking publicity, it can be tempting to forget these rules and exploit logical fallacies that we think others won’t notice.
Maybe one day in the far future we will be able to make psychiatric diagnoses based on brain imaging alone. But in the meantime, neuroscientists like Fallon would do well to heed Richard Feynman’s famous warning: “The first principle is that you must not fool yourself, and you are the easiest person to fool”.