That's the guess I'd have made, as well, but curiously enough, it wasn't:
"As can be seen in Table 2, the prediction is borne out at least in
part. The difference between AI and AA in the converging cues block is
only significant at the 5% level in the NNV WO. In the competing cues
block, on the other hand, the differences are all significant as the
table shows. This means that in adult processing of Swedish, the
interaction of converging cues only help determine the subject more
effectively in unreliable word orders." (Marianne Gulberg "Who's Doing
What to Whom -- Testing the Competition Model on Swedish" Lund
University, Dept. of Linguistics Working Papers 42 (1994), 35-48)
I'm not sure what an article by a Swede tells us about the way English
is written by native speakers, but no matter. I'd have put a comma
before "at", if only to avoid sending readers off on the wrong track,
but it's just about acceptable without.
> "The prediction is borne out (at least in part) by ..."
>
> Even if the sentence ended there, that "at" is part of "at least ...", not
> "... out at".
However, the fact that that is obvious to you and me but not to M.
Winther tells us something about his powers of reading comprehension.
>
>> If a prediction can be borne out, then a standpoint can be borne out,
>> too. This means that it stands for itself, and doesn't need a "by". It
>> can be combined with "at", just as well. No?
>
> No.
--
athel