[dragonwriter]

> Our definition of autism is based on symptoms. This isn't like someone with a rash deciding it's caused by imbalanced humors. It's more like someone noting that their skin is red, itchy, and has little bumps on it, and deciding that they have a rash.

No, it's not. A rash is a symptom, not a disorder diagnosed from a complex combination of symptoms; it's more like self-diagnosis of AIDS prior to the discovery of the role of HIV, and the use of HIV testing.

[PhasmaFelis]

Our definition of autism, like many psychological conditions, is based on symptoms--behavior and development--not causes. We're learning more about the underlying reasons why brains work the way they do, so that may change eventually. But right now, if you have a large number of autistic symptoms, you most likely meet the diagnostic criteria for ASD.

In young children, the visible symptoms of autism may mimic other developmental conditions, so a professional diagnosis is important. An intelligent adult can make the distinction for themselves, given enough information. (Though they'll still need a diagnosis to seek treatment or accommodations.)

[dragonwriter]

> Our definition of autism, like many psychological conditions, is based on symptoms--behavior and development--not causes.

That's also true of many physical conditions with unknown causes, hence the comparison to pre-knowledge-of-HIV AIDS as a more apt comparison than a rash.

> But right now, if you have a large number of autistic symptoms, you most likely meet the diagnostic criteria for ASD.

By DSM 5 criteria, you need either two or three symptoms (depending on exactly where you bound así for symptom) symptoms, manifesting by a particular time, with specific social impact, and all the potential symptoms (and the impact) are defined in a way that even an intelligent layman in a position to be objective rather than applying a bias from their own position is likely to have trouble applying them well.