Shingo Tokimoto

Disambiguation of Homonyms in Real-Time Japanese Sentence Processing: Case-Markings and Thematic Constraint

This paper experimentally examines the effects of the case-markings and the constraint on the assignments and the receptions of thematic roles in Japanese sentence processing. A self-paced reading experiment was carried out with syntactically well-controlled Japanese sentences including homonyms locally ambiguous between nouns and verbs. The results showed that the homonyms were preferably disambiguated as verbs. We interpret this disambiguation as the result of the application of the thematic constraint to the input items on the basis of the correspondence between the case-markings and the grammatical functions in Japanese. We further examined the effect of pragmatic plausibility on the interpretation of the homonyms by questionnaire, and claim that the thematic constraint is still the chief determinant for their disambiguation even with the possible plausibility effects. We also examined the effect of the verbal working memory capacity estimated by the Japanese Reading Span Test, and we demonstrate that a reader with a high score in the test comprehends sentences more accurately but spends relatively longer time for reanalyses than a reader with a low score. We discuss the relevances of our results to the parsing models for real-time Japanese sentence processing and to the studies of verbal working memory in English.

Perspective-Taking in Sentence Comprehension: Time and Empathy

This study examines the neural substrate of perspective-taking by analyzing the electroencephalographic (EEG) activity elicited by the auditory comprehension of sentences for which the comprehender had to adopt the perspective of the person described in them. Recent studies suggest that the ability of perspective-taking can be an integrative function of temporal and spatial information processing. We thus examined the independence and possible interaction of human perspective shifts and temporal perspective-taking by utilizing Japanese subsidiary verbs for giving, namely -ageru and -kureru. We manipulated human perspective shifts and temporal perspective-taking independently in experimental sentences by syntactically changing the subject and the object between the speaker and a third person, while we manipulated the tense to be past or non-past tense via sentence-final particles ru/ta (non-past/past). The EEG analyses via electrodes indicated the suppression of the β band for human perspective shifts in sentences in non-past tense and the absence of such suppression in sentences in past tense. The analyses for the clusters of independent components indicated β suppression for past tense against non-past tense in sentences without a human perspective shift. This response pattern suggests a close relationship between human perspective shifting and temporal perspective-taking. The β suppression for the human perspective shift in our experiment can be understood as a replication of the previous EEG findings observed for perspective-taking in the presentation of visual images. The preceding findings and our result suggest that the ability or the function of perspective-taking is not specific to the modality. Furthermore, the generator of the β suppression for past tense against non-past tense without human perspective shifting was localized in the precuneus, which is consistent with recent findings indicating that the precuneus is deeply involved in time perception.

Neurophysiological base of Japanese honorific expressions: Human relationship in language use

This study investigates the neurophysiological base of honorification by the event-related brain potentials (ERPs) elicited by Japanese honorific expressions. Japanese is widely known to have many linguistic forms peculiar to respect, namely, honorific expressions. Researchers have long discussed whether the rules of honorification is semantic or syntactic. We will discuss the property of honorific rules and its domain specificity by the ERPs.Japanese honorific expressions are generally classified into two types, namely, exalted and humble expressions. The former expresses the respect to a person by the words directly relevant to the person, and the latter does so indirectly by the words for the modesty of the speaker. To examine the neural bases of the two, we constructed two types of experimental sentences from common basic verbs. For example, a sentence including osshaimashita ((the person) said) and one including mooshiagemashita ((I) said) were constructed respectively as an exalted and a humble sentence from a common verb itta (said). An example for the former is ‘Sensee-ga iken-o watashi-ni osshaimashita’ (The teacher told me an opinion), and one for the latter is ‘Watashi-ga iken-o sensee-ni mooshiagemashita’ (I told the teacher an opinion).The experiment was designed with three factors: Type of honorific expression (exalted and humble), word-order (canonical and scrambled) and grammaticality (grammatical and ungrammatical). 160 sentences were constructed for the two types, and further 40 controls were included in the main session.The phraseby-phrase visual presentation of the sentences elicited negativity around 400ms after the onsets of the main verbs (N400) for the ungrammatical sentences, suggesting that the rules of honorification are semantic. Furthermore, the component was greater for the humble than the exalted. We also observed the effect of word-order. We are now collecting more data to reveal the details of the possible inteactions.

Segmentation and Selection of Appropriate Chinese Characters in Writing Place Names in Japanese

This paper explores the relation between an unknown place name written in hiragana (a Japanese syllabary) and its corresponding written representation in kanji (Chinese characters). We propose three principles as those operating in the selection of the appropriate Chinese characters in writing unknown place names. The three principles are concerned with the combination of on and kun readings (zyuubako-yomi), the number of segmentations, and the bimoraicity characteristics of kanji chosen. We performed two experiments to test the principles; the results supported our hypotheses. These results have some implications for the structure of the Japanese mental lexicon, for the processing load in the use of Chinese characters, and for Japanese prosody and morphology.