Hmm.

I found that with all of those I could make out at least some, and in some cases all, of the words in the sine-wave versions without the "priming" original recording. The very first example on that page was pretty much perfectly clear to me on the first listening; others were more work.

Then, for the ones where I hadn't been able to make out all the words in the sine-wave version, I listened to the "primer" and tried again with the sine waves. In each case, I found that I could then recognize the words I'd found unclear before, but I wouldn't say they were "clear as day" or anything like; more like "well, OK, I suppose it could be interpreted that way".

The effect of priming, for me, therefore appears to be very small.

I tried the first pair on my wife, and her response appears to be the canonical one. Obviously I'm just strange. Anyone else have the same experience as me?