3 ms·

Scott Alexander [1] allegedly works with autistic patients and has a different impression. My impression is that autism is an epistemologically unsound category

by EdiX 11y ago

Scott Alexander [1] allegedly works with autistic patients and has a different impression. My impression is that autism is an epistemologically unsound category, as if biologists had 'flying animal' as a category encompassing some insects, some birds and bats, with flying squirrels and frogs classified "in the flight spectrum". [1] http://slatestarcodex.com/2015/10/12/against-against-autism-cures/ http://slatestarcodex.com/2015/10/12/against-against-autism-...

PepeGomez 11y ago
That's surely a possibility as well. But it's impossible to decide until medical science improves. Be it autism, depression, obesity, nearsightedness, and so on, medical research is wandering helplessly with different hypotheses being in vogue at different times while having no way to conclusively prove or disprove any of them. The problem is that medical science is dominated by a particularly insidious misinterpretation of frequentist statistics which makes most medical research for most practical purposes worthles. It goes as follows: A study that looks at one possible cause is, all things being the same, a hundred timess less likely to produce a false positive than a study that looks at a hundred different causes, so we should make studies as specific as possible to avoid false positives. The error in this reasoning is that while a study that looks at a hundred possible causes is indeed a hundred times more likely to produce a false positive, it also produces a hundred times more data, so the amount of noise is the same. So in an ideal world, a hundred small studies are perfectly equal to one big study, so it shouldn't be such a big deal after all. But we don't live in an ideal world and studies that don't produce significant results are often shelved, and studies that report results for one possible cause in fact looked at many and report only the most significant result. This makes it impossible to add up results from different studies to get a clearer picture and you only have separate studies pointing at random directions and leading nowhere, with no hope of getting more certain results untilt the way science is done changes.
- zajd 11y ago
  > But we don't live in an ideal world What a funny thing for someone to say after they just finished unironically posting > You won't find an autistic child whose parents aren't either mentally ill or autistic themselves.
  - PepeGomez 11y ago
    You completely misunderstood the first sentence you quoted then.
mhkool 11y ago
I urge you to read of few articles that Dr Perlmutter refers to. Only when you see the hard data you will be able to accept that diet and microbiome play a very important role.

by EdiX 11y ago

PepeGomez 11y ago

That's surely a possibility as well. But it's impossible to decide until medical science improves. Be it autism, depression, obesity, nearsightedness, and so on, medical research is wandering helplessly with different hypotheses being in vogue at different times while having no way to conclusively prove or disprove any of them. The problem is that medical science is dominated by a particularly insidious misinterpretation of frequentist statistics which makes most medical research for most practical purposes worthles. It goes as follows: A study that looks at one possible cause is, all things being the same, a hundred timess less likely to produce a false positive than a study that looks at a hundred different causes, so we should make studies as specific as possible to avoid false positives. The error in this reasoning is that while a study that looks at a hundred possible causes is indeed a hundred times more likely to produce a false positive, it also produces a hundred times more data, so the amount of noise is the same. So in an ideal world, a hundred small studies are perfectly equal to one big study, so it shouldn't be such a big deal after all. But we don't live in an ideal world and studies that don't produce significant results are often shelved, and studies that report results for one possible cause in fact looked at many and report only the most significant result. This makes it impossible to add up results from different studies to get a clearer picture and you only have separate studies pointing at random directions and leading nowhere, with no hope of getting more certain results untilt the way science is done changes.

zajd 11y ago
> But we don't live in an ideal world What a funny thing for someone to say after they just finished unironically posting > You won't find an autistic child whose parents aren't either mentally ill or autistic themselves.
- PepeGomez 11y ago
  You completely misunderstood the first sentence you quoted then.

mhkool 11y ago

I urge you to read of few articles that Dr Perlmutter refers to. Only when you see the hard data you will be able to accept that diet and microbiome play a very important role.