Parallel processing and initial phoneme criterion in naming words: Evidence from frequency effects on onset and rime duration

This study reports 4 experiments that investigated the locus of temporal effects of printed word frequency in speeded-naming tasks. Response latencies and onset durations are shorter for high-frequency words compared with low-frequency words, but there is no effect of frequency on rime durations. These results can only be accounted for if (a) phonemes are activated in parallel and not sequentially from left to right and (b) the criterion to initiate pronunciation is based on the initial phoneme and not the whole word. In addition, the effect of word-initial phoneme characteristics on acoustic latency was investigated. The acoustic latency of words beginning with voiceless sibilants was less than that of words beginning with plosives, a pattern opposite that reported by R. Treiman, J. Mullennix, R. Bijeljac-Babic, and E. E. Richmond-Welty (1995). This difference was attributed to the lower sensitivity of voice keys compared with measures based on digitized responses.