Toward a Model of Cantonese Spoken Word Production.

摘要

Five experiments were conducted to investigate how phonological information is processed in Cantonese spoken word production using the picture-word interference (PWI) paradigm. Participants were asked to name aloud individually presented pictures and ignore an accompanying auditory word distractor. In the first three experiments of the present study, the target picture names were Cantonese mono-syllables with a consonant-vowel-consonant (CVC) structure. Participants' picture naming latencies were found faster when the target (e.g., /sing1/l¹, "Star") and the distractor (e.g., /ging 2/, /sik6/, or /soeng3/) shared two identical segments (irrespective of the segments' syllable-internal position), than when they were unrelated, whereas no reliable effects were obtained when they shared only the vowel (e.g., /dim3/), the coda (e.g., /hu ng2/), or together with the tone (e.g., /bit1/ or /fung1/). Furthermore, the facilitation effect observed in the consonant+consonant+tone-related condition (e.g., /soe ng1/) was found reliably larger than that in the consonant+consonant-related condition (e.g., /soeng3/). In Experiment 4, the syllable structure of the targets was manipulated such that half of the picture names were mono-syllables with a consonant-vowel (CV) structure (e.g., / so1/, "comb") and the other half a CVC structure (/sing1/). A significant syllable (without tone) related facilitation was found irrespective of the target type, whereas no reliable effect was observed when the target and the distractor (e.g., /se6/ and /sam2/, for CV and CVC targets, respectively) shared only the onset consonant. Furthermore, Experiment 5 investigated whether phonetic factors modulate the facilitation effect observed in a PWI task. To this end, the sonority level and duration of the target's rhyme component were manipulated in Experiments 5A and 5B, respectively. A significant rhyme-related priming effect was found in both Experiments 5A and 5B, and such effect was neither affected by the sonority level nor duration of the target's rhyme component. These results are in line with the notions that (1) a single segment does not have a significant impact on Chinese spoken word planning, (2) sub-syllabic constitutes such as rhyme is an important processing unit, (3) the lexical tone has a unique role to play during phonological encoding, and (4) segmental and tonal retrieval are operated in an interactive manner. Based on all the available results from Cantonese PWI research, an interactive model of Cantonese spoken word production is proposed.;¹The transcriptions for Cantonese syllables presented in this paper are based on the Cantonese Romanization Scheme proposed by the Linguistic Society of Hong Kong (Chinese Character Database: With Word-formations, 2003). The number besides each syllable marking denotes the lexical tone.