The nature of phonological encoding during spoken word retrieval

In the present study, two priming experiments were performed to gain a better understanding of the planning processes that underlie phonological encoding. In Experiment 1, participants named picture primes and targets that were identical, shared the same onsets (e.g. coat-comb), or shared the same rhymes (e.g. mouse-house). In addition, the response stimulus interval (RSI) between the primes and targets was varied (650 and 1000 msec). The results revealed a facilitatory repetition priming effect that did not decrease with RSI. In addition, a reliable rhyme-related inhibitory effect was obtained, which was smaller in magnitude than the onset-related inhibitory effect. In Experiment 2, the identity condition was excluded to determine if a strategic comparison process may have produced the inhibitory effects. In addition, a new stimulus set was created and the RSI was held at 650 msec. The results replicated those of Experiment 1 by showing a reliably larger inhibitory effect for onset-related than rhyme-related targets. The results are consistent with Dell's (1988) two-stage sequential model of encoding in which there is an initial parallel activation within a lexical network followed by a sequential left-to-right selection of the intended word's phonemes.