ISSN 0439-755X
CN 11-1911/B
主办:中国心理学会
   中国科学院心理研究所
出版:科学出版社

心理学报 ›› 2004, Vol. 36 ›› Issue (05): 515-524.

• • 上一篇    下一篇

词素效应到底是什么?——来自维吾尔语的证据

玛依拉·亚克甫,周晓林   

  1. (1 北京大学脑科学与认知科学中心,北京 100871) (2 北京大学心理学系,北京 100871)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2004-09-30 发布日期:2004-09-30
  • 通讯作者: 周晓林

WHAT ARE REALLY MORPHOLOGICAL EFFECTS IN LEXICAL PROCESSING? EVIDENCE FROM THE UYGHUR LANGUAGE

Mahire Yakup,Zhou Xiaolin   

  • Received:1900-01-01 Revised:1900-01-01 Published:2004-09-30 Online:2004-09-30
  • Contact: Xiaolin Zhou

摘要: 利用维吾尔语的一些特点,采用视觉-视觉启动、听觉-视觉跨通道启动、和延迟重复启动等三种启动技术,改变目标词和启动词之间在词素、词形以及语义上的多种关系,通过比较启动效应的异同,来考察词素效应是否可以还原为词形启动效应、语义效应,是否可以看成是词形和语义在实时加工中相互作用的后果,从而探讨词汇加工中词素效应的实质。实验一发现,在明确配对的视觉-视觉和跨通道启动中,启动词和目标词之间只有在具有词素加语义关系(+m+o+s / D,+m+o+s / R;其中,m 代表词素,o 代表词形,s 代表语义,D 代表目标词为派生词,R代表目标词为自由词根;+ 表示启动词和目标词之间在某个维度上相关,-表示启动词和目标词之间在某个维度上无关)时或者仅仅具有语义关系(-m-o+s / D)时,才有显著的启动效应。如果启动词与目标词之间仅仅具有词素(词形)关系,而没有语义关系(+m+o-s / D),或者仅仅具有词形关系(-m+o-s / R),它们之间的启动效应微乎其微。实验二通过使用维吾尔语中的借词和延迟启动技术,比较了词素加语义加词形(+m+o+s / D)的启动和语义加词形启动(-m+o+s / D)以及语义启动(-m+o+s / D),发现词素启动效应并没有不同于其他启动条件下的效应。两个实验都表明,语义关系是产生启动效应的关键,没有语义相关,就不可能产生词素启动,词素效应依赖于语义效应。有了语义关系,词素关系和词形关系产生的作用很小。因此,词汇加工的词素效应是词形和语义统计学规律的反应,是词形和语义加工相互作用的后果,心理词典中并不需要明确地表征词汇结构。

关键词: 词汇加工, 词素启动, 维吾尔语, 连接主义

Abstract: What is nature of morphological effects in lexical processing? Can they be reduced to lexical form effects, semantic effects, or the interaction between form and semantic processing? Are there explicit representations of morphological structures in the mental lexicon? These questions were investigated in two experiments on the Uyghur language, which is spoken in China’s Xinjiang area and is an agglutinate language with rich morphological structures. Experiment 1 used the visual-visual priming and cross-modal priming lexical decision tasks and manipulated the morphological relations between primes and targets. In both tasks, significant priming effects were observed for prime-target pairs that shared root morphemes and were semantically related as wholes, whether the targets were derived words (+m+o+s / D, or free-root morphemes (+m+o+s / R). [Here “m” denotes the morphological relation between primes and target, “o” orthographic relation, “s” semantic relation; and “+” denotes having the relation between primes and targets in the relevant dimension and “-“ having no relation.] Priming effects with similar magnitudes were also observed for words that were only semantically related (-m-o+s / D). In contrast, no significant effect was observed for prime-target pairs that were morphologically related but not semantically related (+m+o-s / D) or that were only formally related (-m+o-s / R). Experiment 2 took advantage of the fact that in Uyghur many words are borrowed from other languages and some of these words are morphologically related in the original languages but only semantically and formally related in Uyghur. Using the delayed repetition priming technique, this experiment compared the priming effects between Uyghur morphologically related words (+m+o+s / D) , borrowed words (-m+o+s / D) and pure semantically related words (-m+o+s / D). The same pattern of priming effects were found for the three types of words in the short and long lag repetition conditions. These findings were interpreted as supporting the connectionist argument that morphological structure does not need to be represented explicitly in the lexicon and morphological effects in lexical processing can be seen as the result of interaction between semantic processing and form processing.

Key words: lexical processing, morphological priming, Uyghur, connectionist