中文及英文量詞『一些』與『全部』的推論歷程：閱讀時間及事件相關電位研究

Abstract

量詞處理歷程指的是在例如像『一些，全部』的量詞表中，當說話者用『一些』這個詞在句中時，他們心理拒絕使用另一個詞『全部』來表達。在語用學的領域中，目前有兩種理論可茲說明：『預設推論理論』及『相關理論』。這兩種理論對於人們是如何處理量詞推論做出了完全不同的預測。截至目前為止，量詞處理歷程大部分仍是依靠這兩個語用學理論或是本能想法來做闡述，鮮少有研究用實驗法或其他線上實驗方式來看這個議題。基於實驗法早已被許多學科，如心理語言學，所採用，且其他非線上實驗也產生了許多相抵觸的實驗結果，本文擬提出一個兩年期的線上實驗計畫來研究這個議題。實驗分成：(1)以中文為母與者是如何處理『一些』及『全部』推論; (2)學英文的中文母語人士是如何處理英文『一些』和『全部』的推論。因為本實驗涵蓋第一及第二語言，故預估需耗時兩年始可完成。每年的預定計畫如下： 第一年： 第一年的計畫將用於研究中文使用者是如何處理量詞推論（第一語言研究）。先做第一語言的研究是因為第一語言處理歷程通常會幫助我們瞭解第二語言處理。此外，為了要瞭解哪一種理論做出的預測是正確的，我們擬採用線上實驗法來實證這個議題。計畫中的線上實驗法包含反應時間，閱讀時間和事件相關電位研究。因為閱讀時間可以反應出受試者在那個詞中停頓下來，實驗結果可告訴我們受試者在何處覺得受到阻礙。此外，先前研究發現工作記憶和認知處理有極大相關。我們也將藉由實驗探討工作記憶和量詞處理歷程之間的關係。在實驗法上，因為反應時間和閱讀時間都受限於提供其他更詳細的資訊，本計畫也擬採用事件相關電位科技來研究此議題。實驗結果將可得知大腦如何處理量詞。 第二年： 第二年的研究將著重於英語學習者對英文中”some” and “all”的處理歷程。如前提，第二語言處理與第一語言處理具有高度相關，故第二語言實驗需等第一語言實驗完成後方可進行。第二年的研究問題和第一年相似。但因為第二語言處理更為複雜，故我們預期實驗結果會更有趣，或許會對文獻提出更多不同的觀點。同樣的，第二語言研究也包含反應時間，閱讀時間，及事件相關電位研究。研究結果應可對人類語言處理歷程產生貢獻。

Scalar inferences represent the condition when a speaker uses a weaker expression such as some in a pragmatic scale like , s/he intends to reject the more stronger use of the other word like all in the utterance. Two different pragmatic theories, i.e., Default Inference and Relevance theory, have made different predictions regarding how people process scalar inferences. Up to date, how people process scalar inferences have mostly rely on these pragmatic theories or intuitions. Few studies have looked at scalar inferences from experimental approaches and even fewer studies have looked at this issue from online measures. Given that experimental approaches have been used widely in psycholinguistics and other disciplines and given that mixed results exist among these offline studies, the current proposal intends to bridge this gap by proposing a two-year online studies to investigate scalar inferences by Mandarin speakers (L1) and Mandarin learners of English (L2). Given that first language and foreign language processing of scalar inferences will be investigated, it will take two years to conduct this series of experiments. Detailed explanation of how experiments will be conducted in each year will be given below: First Year: The first year will be used to investigate scalar inferences by native Mandarin speakers. First language studies will be performed first because it will help understand second language processing. To solve the existing discrepancies among different studies, online techniques such as reaction times, reading times studies and event-related potentials, will be implemented to see which theory is correct in predicting scalar inferences. Since reading time studies can be used to reflect word-by-word processing of sentences containing scalar inferences, it will shed light on the regions where native speakers encounter difficulty. In addition, since reading time studies are limited in providing information regarding brain responses and latencies, event-related potentials will be used to further reveal the latencies and components that are associated with scalar processing. Second year: The second year will be devoted to understanding scalar processing by foreign language speakers, i.e., Mandarin learners of English. Given that language processing of a second language or a foreign language correlates highly with first language processing, studies on scalar inferences by Mandarin speakers of English will wait after the studies on first language processing are finished. The results of first language processing will be helpful in understanding scalar inferences by foreign language speakers. The same research questions will be investigated in the second year. However, given that processing a second language is more complex, we expect to observe more interesting processing phenomena in the results. Again, reaction times, reading time studies and event-related potentials will be used to shed light on the difficulties, latencies and brain responses of scalar processing and hopefully, our findings in both first and second languages contribute to our knowledge of human language processing system.