Grouping and the verbal transformation effect: The influence of formant transitions.

Abstract

Listening to a recorded word repeated many times produces verbal transformations (VTs) to different forms; this effect may be due in part to perceptual re‐grouping of the speech sounds. Six CVC words producing strong formant transitions between the initial consonant and vowel (e.g., “short”) were recorded and monotonized at F0=130 Hz; each word was paired with another producing weak formant transitions (e.g., “fort”). Each set of six words was used to derive another in which the CV transitions were edited out and replaced with samples selected from the neighboring steady‐state portions. For each word pair, the amount of splicing applied to the weak‐transition stimulus (30–70 ms) was matched to that required for its strong‐transition counterpart, and the VTs obtained for 3‐min sequences of the edited and unedited versions were compared. Listeners reported more VT forms in the spliced than in the unedited case for the strong‐transition words, but not for those with weak transitions. Hence, the effect is not...