Autistics: less biased. Researchers?

“Researchers performing fMRI scans systematically report changes in the activation of some brain regions as deficits in the autistic group – rather than evidence simply of their alternative, yet sometimes successful, brain organisation.” 

Laurent Mottron, quoted in: “Get over your bias against people with autism, scientist says.” 

LA Times. Karen Kaplan. 2011

When it comes to academic research, we presume that researchers are doing their best to be rigorous and objective, which includes avoiding bias. But when it comes to studies involving autistic subjects, are they?

We know that the way information is presented influences our judgement and decision-making: the framing effect. How we “frame” information has an impact on how we treat it, and thus on outcomes. So if your research project is based on the assumption that your subjects have a disorder or a deficit, that presumption will be reflected in both your process and your results.

What we see in research involving autistic subjects is that autism is frequently framed as “non-neurotypical”, i.e. autistics are measured against people who are non-autistic and thus end up being defined by what they aren’t. For a comparison, imagine a linguistic study of a Swedish-speaking community by French academics where the conclusion is “they can’t speak French”. If ethnocentrism is judging another culture solely by the values and standards of one's own culture, then what we often see in these papers is a kind of neurocentrism, judging another neurology by the capabilities and standards of one’s own neurology. Mottron and his research associate Michelle Dawson use the term “normocentrism”: if a non-autistic person does something it’s normal, but if an autistic person does it, it’s abnormal.

Of course, once we take such a position as a point of departure, we’re on the slippery slope of confirmation bias where, as the economist Franck Schuurmans remarked: 

“One seeks data that will validate one's bias”. 

If you’re a hammer, all you see are nails, and if autism is a deficiency, all you’ll see are deficits. Here’s an example from a study on autistic decision-making and the framing effect:

“ASD subjects show a reduced susceptibility to the framing effect.”

(De Martino et al., 2008)

The researchers explain that this is because of a failure to incorporate contextual information. So a capability (avoiding the framing effect) is presented as the result of a failure, an inadequacy.

Examples of negative framing of autism permeate this study. For example, the researchers also established that subjects on the autism spectrum “fail to incorporate emotional context into the decision-making process”. This could have been presented in a positive frame such as “succeeded in not incorporating emotional context”, or even a neutral frame e.g. “emotional context was not incorporated”. But of the three choices available, we only get the negative one. In a study on framing biases, the authors seem incapable of recognising the impact their own framing bias is having on their conclusions.

Similarly, they find that an “enhanced capability in logical consistency” is evidence of a “core neurobiological deficit”. (By the way, if you’re wondering what “enhanced capability in logical consistency” means, to paraphrase Rudyard Kipling, it’s being able to keep your head when all around you are losing theirs.) The key words in these phrases are “capability” and “deficit”, with the former being presented as evidence for the latter. Normally of course you’d expect the reverse, that it would be a diminished capability that was evidence of a deficit. But apparently not in this case. The diminished capability of the non-autistic subjects in this test (in avoiding the framing effect) passes without comment. We set up an inability as the norm, and thus see the corresponding ability as a deficit. Lewis Carroll would be pleased, we’re through the looking glass here.

This is only one example, but many of the studies into cognitive fallacies and autism take a similar tack and could be summed up as: autistics aren’t good at making the same mistakes as neurotypicals.

For another example in a similar vein, this from a study on how autistics handle the conjunction fallacy:

“The findings presented here suggest that the conjunction fallacy is less likely to occur with autistic participants.”  

(Morsanyi et al., 2009)

(If maths isn’t your thing, the conjunction fallacy isn’t easily demonstrated without a full example, but it basically says that we tend to think that adding detail to an event makes it more probable when the reverse is true. Adding detail can make an event more believable, but not more likely.)

So, autistics are better at avoiding the conjunction fallacy. But for the authors, this unique cognitive strength actually shows that autistics display “deficits in contextual processing”. 

This is another common theme in many of these studies. Autistics are ignoring context, which is giving them an advantage in situations where context gets in the way of rational decision making. But ignoring context is kind of cheating, so it’s presented as a shortcoming.

Still another example, this one from a study on managing social reputation, looking at how decision-making is influenced by the presence of others. Here we see that autistic people continue to make decisions in the same way, regardless of the social situation or whether they’re being watched. A strength? Once again, it seems not:

“The findings indicate that individuals with ASD have a specific deficit in taking into account their reputation”. 

(Izuma et al.,2011)

In other words, autistic people don’t let what others think of them influence their decision-making. But for the researchers, that’s not simply a different way of processing social information, it’s a shortcoming, a “specific deficit”. The conclusions in this case mirror what we saw in regards to ignoring context: You performed better, but you weren’t supposed to, so we’ll mark that down as a deficiency.

This study is particularly noteworthy for another reason. In regards to these “deficiencies”, most studies usually concede at some point that “perhaps this could be a benefit in some circumstances”. But the authors of this paper couldn’t find a single redeeming feature in autistics’ capacity to keep thinking straight under changing social situations.

In regard to these research biases, it’s revealing to look at how the same academics frame outcomes differently depending on whether their studies involve autistic subjects or not. In a general study on biases and rational decision-making involving non-autistics, Benedetto De Martino had this to say:

“People who are more rational don't perceive emotion less, they just regulate it better.” 

(De Martino et al., 2006)

Sounds reasonable. But when Dr. De Martino and his team undertook an almost identical study with people on the spectrum, looking at enhanced logical consistency during decision making, he proposed that:

“although this (logical consistency) protects ASD subjects from the framing bias … it may come at a cost of the emotional and behavioural deficits that characterise the condition” 

(De Martino et al., 2008)

So, if you’re non-autistic and more rational, it’s because you’re good at regulating your emotions. But if you’re autistic and more rational, it’s because you’re deficit in recognising your emotions.

It’s worth noting that this claim is not actually established anywhere in this paper. Are people on the spectrum failing to recognise their emotions, or are they recognising them but choosing to ignore them? There’s nothing in this research to confirm the former, though this omission is characteristic of a number of these studies. As per Dr. Mottron:

“When autistics outperform others in certain tasks, their strengths are frequently viewed as compensatory of other deficits, even when no such deficit has been demonstrated.”

One last example that really struck me in seeing just how far this can go, and how unbalanced we’ve become with some of it. It’s a study of how well children are able to deceive a researcher by hiding their feelings. So, basically a test of how good you are at lying. Here again the autistic subjects failed - in trying to be dishonest:

“children with ASD were found to be less able to refrain from expressing their emotions in order to deceive an experimenter”

(Barbaro & Dissanayake, 2007)

In other words, being dishonest is presented as a capability, while being honest is presented as a deficiency, a conclusion in line with what we’ve seen in the studies above. Though such a view is perhaps not surprising given that the authors are studying:  “children with high functioning autism and Asperger's disorder”.

It’s an understatement to say that there is a bias in autism studies. Autistic people are not being well-served by this research, if we’re being served at all. In the world of academia, strict procedures are in place to ensure that impartiality is maintained: conflict of interest declarations, rules on plagiarism and ethics, independent peer-review prior to publication etc. But many of these studies are biased towards autism as a deficiency and an incapability, which has a predictable impact on the outcomes. To look at an example from another field, Professor Lin Foxhall, Head of Archaeology and Ancient History at the University of Leicester said in regards to his team’s search for the remains of Richard III:

“It is important for the integrity of our work to conduct our research unencumbered by speculation linked to a particular outcome.”

I wish we could say that in these studies with autistic subjects, pre-conceived notions about results were limited just to speculation about outcomes. If only. Much of this research is permeated from start to finish with deep-seated autism-as-deficit convictions.

Last word to Professor Mottron:

“The definition of autism itself is biased against people who have it. The condition is characterised by a suite of negative characteristics including problems with language and social interactions. Autism’s many advantages are not part of the diagnostic criteria.”

References:

Get over your bias against people with autism, scientist says. 

LA Times. Karen Kaplan. (2011)

quotes are taken from:

Changing Perceptions: The Power of Autism. 

Laurent Mottron, Nature. (2011)

Laurent Mottron is a full professor in the Department of Psychiatry at the University of Montreal and “Chercheur National” with Quebec Health Research Fund. He holds the Marcel and Rolande Gosselin research chair on cognitive neuroscience in autism of the Montreal University since 2008.[ As a clinician, he founded the specialised autism spectrum disorder without mental retardation clinic at the Rivière-des-Prairies Hospital, Montreal, Quebec in 1995, and the pervasive development disorder centre for excellence of Université de Montréal (CETEDUM) in 2007. (from https://en.wikipedia.org/wiki/Laurent_Mottron)

Explaining Enhanced Logical Consistency during Decision Making in Autism. 

Benedetto De Martino, Neil A. Harrison, Steven Knafo, Geoff Bird, and Raymond J. Dolan. (2008)

Heuristics and Biases in Autism: Less Biased But Not More Logical. 

Kinga Morsanyi, Simon Handley, Jonathan Evans. (2009)

Insensitivity to social reputation in autism.

Keise Izuma, Kenji Matsumotob, Colin F. Camerera, and Ralph Adolphsa. (2011)

Frames, Biases, and Rational Decision-Making in the Human Brain.

Benedetto De Martino, Dharshan Kumaran, Ben Seymour, Raymond J. Dolan. (2006)

Quotation from: http://economistsview.typepad.com/economistsview/2006/08/the_framing_eff.html

A comparative study of the use and understanding of self-presentational display rules in children with high functioning autism and Asperger's disorder. 

Josephine Barbaro, Cheryl Dissanayake. (2007)