Selective adaptation of linguistic feature detectors

Abstract

Using a selective adaptation procedure, evidence was obtained for the existence of linguistic feature detectors, analogous to visual feature detectors. These detectors are each sensitive to a restricted range of voice onset times, the physical continuum underlying the perceived phonetic distinctions between voiced and voiceless stop consonants. The sensitivity of a particular detector can be reduced selectively by repetitive presentation of its adequate stimulus. This results in a shift in the locus of the phonetic boundary separating the voiced and voiceless stops.