Baroness Greenfield's contribution to autism science

Not so very long ago, autism was considered to be a rare condition, affecting perhaps 3 or 4 people in every 10,000. The most recent studies, however, paint a very different picture, suggesting that the rate is closer to 1 in every 110 - almost 1%.

Much of that increase can be put down to the progressive widening of the boundaries of autism. Many people with an autism spectrum diagnosis wouldn’t have received a diagnosis 20 years ago or would have been given a different diagnosis. When we compare autism rates over the past few decades, we’re not comparing apples with apples - we’re comparing historical oranges with present day citrus fruit.

Other factors may also be at play, including increased awareness of autism, decreased stigmatization, and improved access to services. And as methods for collating this kind of data improve, the numbers are only ever going to go up.

None of these mitigating factors rule out the possibility that there has also been a real increase in autism, but any increase is going to be much much smaller than the headline-grabbing statistics suggest.

What matters, however, is the public perception that we are in the throes of an autism epidemic. Riding the wave of hysteria, there’s no shortage of theories trying to explain the “epidemic”.

The argument typically goes as follows:

“Autism is increasing. X has also increased. That can’t just be coincidence”.

Well, of course, it can just be coincidence. Plenty of things have changed over the past few decades and they're not all causally related.

Latest to make the illogical leap is Baroness Susan Greenfield, neuroscience professor at Oxford University, and former director of the Royal Institution, one of the world's oldest scientific societies. In other words, someone who really should know better.

In an interview in the New Scientist magazine, she was asked for evidence to support her view that digital technology is changing our brains. Number two on the list was:

"There is an increase in people with autistic spectrum disorders."

In case we were in any doubt, she later clarified her position in the Guardian:

"I point to the increase in autism and I point to internet use. That's all. Establishing a causal relationship is very hard but there are trends out there that we must think about."

Of course, one very easy way to investigate causal relationships is to look at timing. Autism is typically diagnosed in the preschool years, but can be diagnosed reliably in the second year of life. If the internet causes autism then kids would have to be using the internet even earlier than this.

Now, my four-year-old is pretty internet savvy. He can turn on a computer, click on the swirly fox and then click on the Boowa and Kwala bookmark. But that's about it. He doesn’t have a facebook account, he doesn't tweet, and his instant messaging is limited to repeatedly typing the word "zoo". Suffice to say, he doesn't get his social interaction from the internet and I think it's fair to say that the same is true of pretty much every kid his age, including autistic kids.

In short, internet use comes after autism. Ergo, it cannot cause autism (at least not in this particular universe).

Greenfield’s comments drew a swift riposte, notably from Professor Dorothy Bishop, who wrote an open letter highlighting the dangers of casually bringing autism into the debate on internet use.

On Twitter, science writer, Carl Zimmer, proposed his own “Greenfieldism”, which made about as much sense.

"I point to the increase in esophageal cancer and I point to The Brady Bunch. That's all. #greenfieldism"

Soon everyone had a #greenfieldism.

However, while I share Bishop and Zimmer’s frustration, I do think that Greenfield might, inadvertently, have made an important contribution to autism science.

Her theory is demonstrably false. You don't need a professorship to realise that autism cannot possibly be caused by the internet. It's easy to understand that, in this case, there may be a correlation between internet usage and autism rates but there's no causal relationship.

Other theories may be less readily disproven, particularly if they refer to events in a child's life that happened before or around the same time as autism symptoms first become apparent. But, as with internet use, the fact that there's now more of whatever-it-is than there used to be is not in itself evidence that it causes autism. Using that logic is what's now officially known as a Greenfieldism.


Further reading:

• Morton Ann Gernsbacher: Three reasons not to believe in an autism epidemic [PDF]
• Autism and Oughtisms: Does Loss of Empathy make you Autistic? Continuance of the Internet-ASD Debacle.
• Martin Robbins: Has Susan Greenfield been misrepresented?
• Neuroskeptic: Susan Greenfield causes autism 
• Risk Science Blog: Is the internet dangerous? Taking a closer look at Baroness Greenfield’s concerns
• BishopBlog: Susan Greenfield and autistic spectrum disorder: was she misrepresented?
• Andrew Steele: An interview with Susan Greenfield (from 2009)