Not Framed by the Framing Effect

“The autistics, however, were less subject to these framing biases and overall they made more rationally neoclassical choices in the experiments.”

Tyler Cowen

Imagine for a moment that you have a potentially fatal medical condition. However, all is not lost, as you are offered two different treatments that could save your life. Your surgeon explains that for the first, out of every 100 patients 90 recover fully, while for the second, out of every 100 patients, 10 die. If you opt for the first on the basis of the success rate, you’ve fallen for the framing effect. Though you’re not alone. In a 1982 study which asked this question, most people made the same choice - including doctors themselves. (McNeil et al., 1982)

The framing effect is a cognitive fallacy where we make choices differently depending on how information is presented, or framed. Glass half full versus glass half empty, 90% fat-free versus 10% fat. Anyone who’s walked up a Supermarket aisle knows it well.

One of the standard tests to demonstrate the framing effect uses a gambling model, with a gain frame where the outcome is presented as a win, and a loss frame where it’s presented as a loss. What this test shows is that people don’t always make the most rational decisions: “When given $70 in a gambling scenario, people are more likely to gamble their money if they think they are going to “Lose $50” than if they stand to “Keep $20”, even though both options are numerically equivalent.” (Nauert, 2016)

So what happens when the subjects for this test are autistic? There are only two specific studies that I know of. The first, and most often cited, is from 2008. To summarise their findings:

“Autism Spectrum Disorder subjects show a reduced susceptibility to the framing effect” and, “susceptibility to the frame manipulation was markedly reduced in the ASD group” 

(De Martino et al., 2008)

(As an aside, I’ve also looked into this particular study elsewhere here, because while the autistic participants may have been less susceptible to the framing effect, it’s questionable whether we can say the same about the researchers themselves. If you frame autism as “contains 10% fat”, you’re not going to have outcomes that show that it’s “90% fat-free”. If you frame autism as a deficit, then you’re not going to have outcomes that show that it can be a capability, even when that’s the case.)

Some years later, a team at King’s College set out to retest the above results using the same procedure. They confirmed that:

“The framing effect was, in line with previous data, significantly smaller in autistic individuals” 

(Shah et al., 2016)

Neither of these studies were limited to studying how those on the spectrum react to framing, but used the framing fallacy as the basis for further studies in other areas. In particular, they looked at the emotional response of autistic people when making decisions. To summarise this aspect of their results, people on the spectrum “don’t use emotional information to guide their decisions”, and as such “this different way of thinking may sometimes be advantageous in situations where it is it better to follow your head and not your heart.” Only sometimes?

But to return to the framing fallacy: this is another situation where autists, to paraphrase Kipling, “keep their heads when all around are losing theirs”. Where in making decisions or working through problems, autistic people tend to remain objective and rational, regardless of how information is presented or described. They’re not framed by the framing effect.

Postscript: In looking into these kinds of studies and articles, you’ll often see people on the spectrum described as “more logical”. Even one of the references I’ve included here is entitled: “Why People with Autism Are More Logical”. But in reality, this isn’t true. Autists are no better at using logic or thinking logically than anyone else. But where autists do excel, is in avoiding the biases that lead us to make irrational choices, the framing fallacy being a prime example. In other words, people on the spectrum are not more logical, they’re less illogical. 

The autism advantage.