Biases in autism research
Before wondering whether this post has anything to do with recent discussions anywhere, please note that it was written in January and just wasn't posted until now.
I think science is a Good Thing and I think the results of scientific research can be very helpful to autistic people. That being said, much of current autism research suffers from a number of serious problems, none of which I've barely even seen being mentioned outside of the autistic community.
These problems are sufficiently severe, especially in combination, that I cannot bring myself to take large amounts of this research seriously. There is good work being done, but as I see it, it's being obscured by all the rubbish.
I will consider taking autism research in general seriously when...
- Researchers control for common causes of stress, anxiety, depression and detrimental obsessiveness before assuming that those conditions are inherent to, caused by or just mysteriously connected to autism, and without once considering the possibility that they're caused by the same things as in non-autistic people; i.e. discrimination, bullying, abuse, sensory issues, being surrounded by unpredictable people, etc.
- Researchers performing studies of autistic people in institutional settings (including places that function like but aren't called institutions) control for the results institutionalisation has on everyone, such as depression, anxiety, side-effects of medication, fear of staff, learnt helplessness, passivity, Stockholm syndrome and many more, before assuming that either those things or the further results of those are in any way due to autism.
- Researchers presume competence and purposefulness when they encounter difference. Starting with the assumption that everything not like yourself is inferior and broken and then forcing all your findings into that mold may be fine according to one's equally ableist peers, but won't impress the people who are trying to deal with the consequences of this practice.
- Researchers stop assuming that skills and deficiencies of autistic people generalise the same way as those of non-autistic people, and take this into account when designing, performing and evaluating studies. For an example of what happens when one doesn't do this, see this discussion of theory of mind. For an example of what happens when one does, see this study.
- Ethical review boards include autistic people, in a non-tokenistic manner, in the decision making process for whether to conduct a given study of autistic people in the first place. This process of inclusion will necessarily involve thinking about accessibility in ways far beyond ramps, and will have to include reimbursement for transport and loss of income for those who participate.
- Funding bodies make non-tokenistic attempts to solicit feedback and opinion from the autistic community at large. To use a (still somewhat) recent example, giving less than a month's time for feedback and conducting the inquiry during the winter holiday season, while certainly better than nothing, does not speak highly of their interest in actually receiving feedback from a wide variety of autistic people.
- Funding bodies realise the inherent dangers of having a privileged group conduct research on an unprivileged one, especially in a climate of media hysteria and relatively unhindered discrimination, and start funding autistic people to conduct our own research; both medical, psychological and sociological. For an example of just how much of a difference this can make, see the work by Michelle Dawson.
Until these issues have been addressed, I will continue to look very carefully at the methodology of any studies performed on autistic people before even considering taking their results seriously, regardless of whose opinions they support.
PS: No, I'm not back, just clearing out the post queue.
