In an earlier post I had estimated that on an IQ scale where the general U.S. population has a mean of 100 and a standard deviation of 15, autistics have a mean of 72 with an SD of 23. So although the average autistic is much less intelligent than the average neurotypical, their greater variability makes them over-represented among the greatest minds.
However my analysis was based on diagnosed autistics. It’s entirely possible that there’s a large population of autistics who escape diagnosis, and this group likely has a much higher average IQ than the diagnosed autistics because they were too high functioning to be diagnosed with a mental problem. In other words, diagnosed autistics are not a random sample of autistics.
How could one get a random sample of autistics? Find a random sample of Americans and give all of them a brain scan. Only those who scored autistic on the brain scan would be considered autistic, regardless of any diagnostic or behavioral history. If this were done, we would likely find the average autistic IQ to be much higher than 72 since such a sample might include many high income computer geeks, university professors and many scientists.