語文認知研究資訊庫的構想及試探性實驗

Information Pool for Psycholinguistic Experiments: Ideas and Pilot Studies

資訊庫是整合語料庫與實驗設計架構及資料分析工具的平台，提出此構想是為因應日新月異的語文認知研究需求。集合不同性質的工具優勢是能持續更新語料資訊，並利用新的分析方法開創新的研究課題。目前已完成的規劃為中文字字形及字音語料資訊，並搭配使用者可依實際需求調整的R函式。藉由這些資訊與功能，我們設計兩件假字實驗的試探性研究做為示範。第一件是假字產製測驗，請受試者根據聲旁挑出可以構成假字的部首，測驗結果符合中文讀者偏好左部首右聲旁的趨勢。第二件是假字朗讀研究，根據語料資訊挑選完全規則與完全例外的聲旁製造假字，並請大學生與小學六年級生進行假字朗讀實驗。此實驗發現大學生能有效運用聲旁發音反應，但是小學生多數還無法掌握。預期隨研究課題的拓展，資訊庫的功能可在現有基礎之上持續更新。

An information pool is an interface of psycholinguistic corpus, experimental design, and professional statistical analysis. The birth of this information pool aims to the coming change of psycholinguistic research. With the integrated functions, researchers are able to update the corpus and explore a new topic based on a new analytical method. Two pilot studies on Chinese pseudo-characters are presented. In the first study, participants were asked to choose a semantic radical to create a pseudo character with the assigned phonetic. The results showed that Chinese readers consistently prefer characters with a ”left semantic and right phonetic” structure. In the second study, the naming performances of college undergraduates and sixth graders were compared. Results indicated significant age differences on pronouncing the novel characters. This information pool will continue to expand as new research topics are developed.