Detecting categorical perception in continuous discrimination data

Abstract

We present a method for assessing categorical perception from continuous discrimination data. Until recently, categorical perception of speech has exclusively been measured by discrimination and identification experiments with a small number of different stimuli, each of which is presented multiple times. Experiments by Rogers and Davis (2009), however, suggest that using non-repeating stimuli yields a more reliable measure of categorization. If this idea is applied to a single phonetic continuum, the continuum has to be densely sampled and the obtained discrimination data is nearly continuous. In the present study, we describe a maximum-likelihood method that is appropriate for analysing such continuous discrimination data.