Jenn Yeu Chen

Proximate Units in Word Production: Phonological Encoding Begins with Syllables in Mandarin Chinese but with Segments in English.

In Mandarin Chinese, speakers benefit from fore-knowledge of what the first syllable but not of what the first phonemic segment of a disyllabic word will be (Chen, Chen, & Dell, 2002), contrasting with findings in English, Dutch, and other Indo-European languages, and challenging the generality of current theories of word production. In this article, we extend the evidence for the language difference by showing that failure to prepare onsets in Mandarin (Experiment 1) applies even to simple monosyllables (Experiments 2-4), and confirm the contrast with English for comparable materials (Experiments 5 and 6). We also provide new evidence that Mandarin speakers do reliably prepare tonally unspecified phonological syllables (Experiment 7). To account for these patterns, we propose a language general proximate units principle whereby intentional preparation for speech as well as phonological-lexical coordination are grounded at the first phonological level below the word at which explicit unit selection occurs. The language difference arises because syllables are proximate units in Mandarin Chinese, whereas segments are proximate in English and other Indo-European languages. The proximate units perspective reconciles the aspiration toward a language general account of word production with the reality of substantial cross-linguistic differences.

Masked Priming of the Syllable in Mandarin Chinese Speech Production

The syllable’s role in Chinese (Mandarin) speech production was investigated using masked syllable priming combined with a word naming task. In three experiments, the prime (shown in the form of a Chinese character but masked) shared the first few segments with the beginning of the disyllabic target word. The segmental overlap either corresponded to a syllable or it did not. Aside from a neutral prime, there were two types of related primes: a CV (e.g., ba3) or a CVG (e.g., bay3) syllable, and two types of targets: a CV-GV (e.g., ba4-ye4) or a CVG-CVX (e.g., bay4-ley4) word, where numbers stand for tones. In Experiment 1, when the prime-target overlap corresponded to a syllable, the tones and the orthographies were also the same (i.e., the overlap in effect corresponded to the entire character). In Experiment 2, the syllable overlap included the tone but not the orthography. In Experiment 3, the syllable overlap was restricted to the syllable only, the tone and the orthography being different. A crossover interaction of the prime by target type was observed not only in Exp. 1 but also in Exp. 2 and Exp. 3, the effect being larger in Exp. 1 and about the same in Exps. 2 and 3. The crossover interaction effects from the last two experiments support the view that the syllable (lacking the tone) is a stored phonological chunk and plays an independent role as a planning unit in Chinese word production.

A Cross-Linguistic Study of Phonological Units: Syllables Emerge from the Statistics of Mandarin Chinese, but not from the Statistics of English

This study explored the statistical patterns of English and Mandarin Chinese sound sequences, by comparing their learning in a simple recurrent network. Experiment 1 showed that vivid syllable structure emerged from the sound sequence of Mandarin Chinese. Experiment 2 further demonstrated that the emerged syllable structure of Mandarin Chinese is considerably more salient than that of English. We claim that the more salient syllable structure in Mandarin Chinese inputs is one reason why syllable units are particularly emphasized in its processing in comparison to English.

Form encoding in Chinese word production does not involve morphemes

Speaking a word can be started faster when all the words in a given block share the initial portion (e.g., syllable) than when they do not (known as the form preparation effect). Two experiments employed the task to examine the role of morphemes in Chinese word production. In Experiment 1, the disyllabic target words were monomorphemic or bimorphemic. They shared the initial syllables or did not. In the bimorphemic condition, the shared syllables were of the same character and of the same morpheme. The form preparation effects were similar in the two conditions. In Experiment 2, the disyllabic target words were bimorphemic. They shared the initial characters or did not. A substantial character preparation effect was observed, but the effect was similar whether the shared characters corresponded to the same or different morphemes. The results of the two experiments support the conclusion that form encoding in Chinese word production involves only the syllable, not the morpheme, nor the orthography.

Word form encoding in Chinese word naming and word typing

The process of word form encoding was investigated in primed word naming and word typing with Chinese monosyllabic words. The target words shared or did not share the onset consonants with the prime words. The stimulus onset asynchrony (SOA) was 100ms or 300ms. Typing required the participants to enter the phonetic letters of the target word, which correspond roughly to the onset and the rhyme of the words syllable. Regardless of SOAs, response times were shorter in the related condition than in the unrelated condition (an onset priming effect) for word typing, but were similar for word naming. The results suggest that naming and typing in Chinese may involve somewhat different word form encoding processes (syllable driven in naming, but segment driven in typing) even though both tasks require accessing the phonological codes. It appears, then, that the kind of outputs a production system is designed to produce can flexibly and adaptively alter the way the system is organized and operates.

Differential Sensitivity to the Gender of a Person by English and Chinese Speakers

Can a linguistic device of a language orient its speakers to a particular aspect of the world and result in increased sensitivity to that aspect? The question was examined with respect to the biological gender marker in English and the lack of it in Chinese. In Experiment 1, English and Chinese participants listened to stories and answered gender and non-gender related questions immediately after. It was found that, relative to the non-gender-related questions, the English participants were much faster and more accurate than the Chinese participants in answering the gender-related questions. In Experiment 2, English and Chinese participants were asked to determine which of two pictures matched the sentence shown immediately before. Relative to the non-gender-related sentences, the English participants were less slower and more accurate than the Chinese participants in responding to the gender-related sentences. The findings support the view that language can have an effect on information processing in human cognition.

Linguistically directed attention to the temporal aspect of action events in monolingual English speakers and Chinese–English bilingual speakers with varying English proficiency *

Chinese and English speakers seem to hold different conceptions of time which may be related to the different codings of time in the two languages. Employing a sentence–picture matching task, we have investigated this linguistic relativity in Chinese–English bilinguals varying in English proficiency and found that those with high proficiency performed differently from those with low proficiency. Additional monolingual English data, reported here, showed further that high-proficiency bilinguals performed similarly to the English monolinguals, suggesting that Chinese speakers’ sensitivity to the time of an action event might be modifiable according to the extent of their experience with a tensed language.

Keynote Speech of the 42(superscript nd) Annual Convention of the Chinese Psychological Association Word Form Encoding in Chinese Speech Production

Research in speech production has almost exclusively focused on Indo-European languages. The present article summarizes recent work in Chinese speech production and points out how it contributes to our understanding of speech production in the language as well as in general. The Levelt-type model of speech production and its computer implementation, WEAVER++, were chosen as the target for investigation. Specifically, the models postulations about the word form encoding processes were scrutinized to see if they hold valid in Chinese. A series of experiments revealed that (1) morphological encoding is at best minimally involved in Chinese speech production, and (2) phonological syllables (segmental syllables) in Chinese may be stored linguistic units which are retrieved during word production.