Individual differences in speech fusion: Methodological and theoretical explorations

Day (Note 1) constructed pairs of speech sounds such as “lanket” and “banket” from words (e.g., blanket) by omitting either the first or second consonant. When these two components were presented about the same time, one to each ear, some people appeared to fuse them, reporting they heard the word. Other people rarely fused. Moreover, if asked to report the first letter they heard, a judgment of temporal order, subjects that tended to fuse also reported what would be the first consonant of the word, even when it was not first in the pair of components. From this and other evidence, Day concluded that the perception of some individuals might be “language-bound.” We tested a strong version of the language-boundness view of fusion and were able to reject it. We also examined several other possible causes of individual differences in fusion rate. Our second experiment provides evidence that a more general failure to discriminate temporal order of even nonlinguistic material may be involved.