In psycholinguistic research, there is still considerable debate about whether the type or token count of the frequency of a particular unit of language better predicts word recognition performance. The present study extends this distinction of type and token measures to the investigation of possible causes underlying syllable frequency effects. In two lexical decision experiments, we found a dissociation suggesting that the token measure of syllable frequency adequately predicts the inhibitory effect of initial syllable frequency, whereas the type measure led to facilitation, especially when the number of higher frequency syllabic neighbours was controlled for. This specific pattern of results, suggesting the involvement of two different processes in effects of syllable frequency, provides a strong constraint for current and future models of visual word recognition.
Stored only when both papers are in the corpus — 53 citations exist upstream.