[gizmo686]

I'm giving up my search for it, but I once read a research paper on categorical perception.

If you compare [ta] and [da], you find that the only difference is the time between when you make the consonant, and when your vocal chords start vibrating (voice onset time). In theory, VOT is a contimum, with any value being possible. However, in English it forms a tri-modal distribution /tʰ/ /t/ and /d/. The experiment artificially edited a sound to vary between /t/ and /d/, including with VOTs between the two that do not occur in English. What they found is that people put all of the sounds in 2 boxes, and were unable to distinguish between sounds in the same box, even if their VOT varied considerably.

However, when test subjects were played the same sounds, but told they were listening to rain drops, this effect disapeared, and they were able to distinguish between sounds in the same box.

[StavrosK]

Hmm, very interesting, thank you! Is it this one?

https://sci-hub.se/https://doi.apa.org/doiLanding?doi=10.103...

Seems to be exactly what you describe.

[gizmo686]

That is along the lines of what I was thinking, but does not make the point as cleany (they end up combining the speach-primed and non speach-primed data because participants said they percieved it as speach from the start.

They also used an ABX methodology, which forces the subject to put sounds into boxes by essentially asking if X is more like A or B, not if X is different from both.

The one I am thinking of used an odd-one-out methodology, where the subject was presented with AAX, and asked to pick which sound was different from the others (where the others were genuinly the same sound). The one I am thinking of also found a priming effect, which yours apparantly didn't.