本文選題：N400 + P300�。� 參考：《電子科技大學(xué)》2012年碩士論文

【摘要】：本文采用ERP技術(shù)來研究漢語短語搭配和句子理解的語義和句法處理過程，以及探索語義語境效應(yīng)。給被試呈現(xiàn)由漢字組成的短語或句子，要求被試判斷短語或句子搭配的正確性。短語任務(wù)包括匹配、量名失匹配和量動失匹配三種條件，句子任務(wù)包括匹配、局部匹配、量名失匹配和量動失匹配四種條件。短語任務(wù)的材料（如，一張+床單/心臟/堅(jiān)持）被包含在句子任務(wù)中，構(gòu)成了句子任務(wù)的局部語境；主語（如，媽媽）和謂語（如，鋪好）構(gòu)成了句子任務(wù)的全局語境。 行為數(shù)據(jù)表明：在兩種任務(wù)中，語義違反有更長的反應(yīng)時(shí)和更低的準(zhǔn)確率；而句法違反和匹配條件沒有顯著性差異。當(dāng)兩種任務(wù)的局部語境相同時(shí)，句子任務(wù)比短語任務(wù)有更快更準(zhǔn)的行為反應(yīng)，支持語義語境能夠易化任務(wù)的理解和執(zhí)行這一觀點(diǎn)。 腦電數(shù)據(jù)表明：量名失匹配在兩種任務(wù)的中線和右額-顳都呈現(xiàn)明顯的N400效應(yīng)；而量動失匹配僅在句子任務(wù)的頂-中位置呈現(xiàn)N400效應(yīng)。且在頂-中區(qū)域，短語任務(wù)無P600效應(yīng)，而句子任務(wù)有顯著的P600效應(yīng)。這表明漢語的語義和句法處理在時(shí)間進(jìn)程上不能完全分離，且只有在一定的語境中，句法整合過程才容易實(shí)現(xiàn)。另外，在頂-中位置，短語任務(wù)比句子任務(wù)有更負(fù)的N400，表明執(zhí)行短語任務(wù)更困難。在短語任務(wù)中，左額-顳比右額-顳激活更負(fù)的N400；而在句子任務(wù)中激活程度剛好相反。大腦后部的P300，是語義違反的一個(gè)指標(biāo)。在短語任務(wù)中P300無半球差異，而在句子任務(wù)中左半球大于右半球。N400和P300的半球差異表明左半球是語言處理的優(yōu)勢半球，，且在句子處理中，有較多右半球區(qū)域的參與。 sLORETA分析發(fā)現(xiàn)，在N400（300-500ms）和P600（500-750ms）窗口，短語任務(wù)內(nèi)各條件間激活無顯著性差異。在句子任務(wù)中，N400效應(yīng)主要定位于額-顳網(wǎng)絡(luò)，而P600效應(yīng)主要定位于額-頂網(wǎng)絡(luò)。在N400窗口，短語任務(wù)比句子任務(wù)在頂葉、額葉和顳葉等區(qū)域激活更強(qiáng)，這可能由于短語任務(wù)語境短，執(zhí)行詞匯的檢索、獲取和整合更困難；在P600窗口，句子任務(wù)比短語任務(wù)激活更強(qiáng)的額下回、顳上回、中央后回和頂上小葉等，這可能是由于句子任務(wù)有較長的語境，涉及更多晚期的語義和句法整合過程。 因此，中文并非越言簡意賅越好，本研究支持句子的語境易化效應(yīng)。
[Abstract]:This paper uses ERP technique to study the semantic and syntactic processing of Chinese phrase collocation and sentence understanding and to explore the semantic contextual effects. The subjects were presented with phrases or sentences composed of Chinese characters, and the subjects were asked to judge the correctness of the collocation of the phrases or sentences. Phrase tasks include three conditions: matching, quantification and quantitive mismatch. Sentence tasks include matching, local matching, quantificational mismatch and quantitive mismatch. The material of phrase task (e.g., a sheet / heart / persistence) is included in the sentence task, which forms the local context of the sentence task, and the subject (for example, mother) and the predicate (e.g., laid) constitute the global context of the sentence task. Behavior data show that semantic violation has longer reaction time and lower accuracy in two tasks, but there is no significant difference between syntactic violation and matching condition. When the local context of the two tasks is the same, the sentence task has a faster and more accurate behavioral response than the phrase task, which supports the view that the semantic context can facilitate the understanding and execution of the task. EEG data showed that the N400 effect was observed in the midline of the two tasks and in the right frontotemporal, while the N400 effect was observed only in the top to middle position of the sentence task. In the top-middle region, phrase tasks have no P600 effect, while sentence tasks have significant P600 effects. This indicates that the semantic and syntactic processing of Chinese can not be completely separated in the process of time, and only in a certain context, the syntactic integration process can be easily realized. In addition, in top-middle position, phrase task has more negative N400 than sentence task, indicating that it is more difficult to execute phrase task. In phrase tasks, N400, which is more negative in left frontotemporal than right frontotemporal activation, is just the opposite in sentence tasks. The P300 at the back of the brain is an indicator of semantic violations. In phrase tasks, P300 has no hemispheric differences, while in sentence tasks, the left hemisphere is larger than the right hemisphere. N400 and P300 hemispheric differences indicate that the left hemisphere is the dominant hemisphere for language processing, and in sentence processing, there are more right hemispheric regions involved. SLORETA analysis showed that there was no significant difference in the activation of each condition in the N400 (300-500ms) and P600-500-750ms) windows. In sentence tasks, the N400 effect is mainly located in frontotemporal network, while the P600 effect is mainly located in frontoparietal network. In the N400 window, phrase tasks are more active in the parietal, frontal and temporal lobes than sentence tasks, which may be more difficult to retrieve, retrieve and integrate in the P600 window because of the short context of phrase tasks. Sentence tasks are more active in subfrontal gyrus, supratemporal gyrus, posterior central gyrus and superior parietal lobule than phrasal tasks, which may be due to the longer context of sentence tasks, which involve more late semantic and syntactic integration processes. Therefore, Chinese is not as concise as possible. This study supports the contextual facilitation of sentences.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2012
【分類號】：R318.0

【共引文獻(xiàn)】

相關(guān)博士學(xué)位論文 前1條

1 徐鵬;信號的稀疏分解及其在腦電信號處理中的應(yīng)用研究[D];電子科技大學(xué);2006年

本文編號：1848367