The Perception of Sub-Phonemic Phonetic Differences

Abstract

In a preliminary study, Fry (1968) discovered that he was able to discriminate between mono-morphemic and bi-morphemic words of identical phonemic shape, such as lax v. lacks. In this study, 10 such assumedly homophonous words pairs were tested for subjects' ability to identify them. When total scores for each word pair were considered, subjects were unable to identify any of the word pairs correctly at a statistically significant level. However, in the case of some individual productions of the 'homophone' pairs, subjects were extremely consistent in how they labelled the words. The productions which were most consistently labelled were analysed in order to determine what acoustic cues the subjects were employing. Apparently, subjects were able to make use of certain acoustic cues that are sometimes considered sub-phonemic. That these cues are not linguistically significant is indicated by the fact that besides correct identifications, the subjects also produced equally consistent mislabellings.