false features in text

spent the morning reading papers. there was an interesting exchange in JOV about what was probably my favorite paper of last year; the exchange strengthened my opinion of the paper, but i'm not going to go into it now, because this post is irrelevant to the actual substance of the articles in question.

here's what i saw/see (in the letter-to-editor attacking the paper in question, as displayed in my browser):

[if you want to see it, keep fixating this line, but attend to the lower part of the text below].

Do you see it? I need to ask around. I notice these things a lot, but they're never as salient as this. In case you don't see it, here's an illustration:

There you go. Go back to the first one and see if you don't see that arc. I only see it peripherally (had to draw that line in without looking at it), but scale isn't so important, and neither is field location (so long as it isn't central). The left eye sees it better than the right, and it's stronger overall with both eyes. To me, it's so salient as to be distracting. I had to do all these tests to convince myself it wasn't a retinal tear or something.

(also interestingly ironic is that the paper(s) in question are all about measuring threshold-versus-noise functions, which are shaped just like the illusory/actual line.)

It's apparently just accidental peripheral concatenation of the structure of the letters and words in the paragraph - crowding, basically - but really, if you look straight at the texture, it's damned hard to see just what is stimulating this process, and why it isn't happening everywhere. But it's neat - basically, a positive effect of crowding or lateral interaction, which is usually held up for its negative (destructive) effects.

Also, I think this qualifies as a deja trompé. Haven't had one of those in a while.