Apparently it's not something you said, since the raters were cool with the transcripts.

I appreciate the fact that they even did this kind of study. I feel like a lot of researchers (and other people) don't bother to make any cognitive distinction between "problems actually caused by the disorder" and "problems caused by other people reacting to the fact that you have the disorder." The actual results of the study are interesting too.

Although it was weird that they considered some of the rated traits to be positive or negative instead of neutral. Like "has lots of friends" is good and "spends a lot of time by themselves" is apparently a bad thing? Why? Well, it doesn't detract from the point they were making.

One thing that might be useful to note... they used the Autism Diagnostic Observation Scale (ADOS) to confirm the autism status of adult subjects. At least for diagnosing adults, a good chunk of the ADOS seems to revolve around the diagnostician's subjective opinion of how awkward you. So most unawkward autistic adults presumably would have been excluded. (Although the chart from Study 2 showed there was one outlier who was not rated awkward.) This isn't criticism of the study; since the ADOS is the so-called "gold standard" of autism diagnosis, it's pretty much what they have to use whether they like it or not. It's just something to keep in mind when considering the implications of this study.

I believe my own awkwardness level changes considerably depending on whether I'm on my ADHD meds. This is because trying to talk takes a lot more working memory capability than I naturally have.

I got a bit of a chuckle out of this:
(If I understand correctly, this means the NTs were rated as more likeable and intelligent when the raters couldn't actually hear what they were saying!)