Single formant contrast in vowel identification

Subjects rated ambiguous steady-state vowels from a continuum with respect to the categories/i/ and /I/ (Experiment 1) or/ε/ and/ae/ (Experiment 2). Each target was preceded, .35 sec earlier, by one of the following precursors: (1) one endpoint from the target continuum, (2) the other endpoint, (3) the isolated first formant (F1) from (1), (4) the isolated F1 from (2), or (5), a hissing noise. Although (3) and (4) did not sound like they came from the target continuum, they produced reliable contrast in both experiments. In the /i/-/I/ experiment, contrast was as powerful from single formants as from the full vowels. These results suggest a sensory, rather than a judgmental, basis for the vowel contrast effects obtained.