汉语词语的音形义在判断句义中的作用
|
摘要
句子加工的研究一直是语言心理学家们关注的焦点,影响句子加工的因素更是研究者们不断探讨的问题。
本研究首先回顾了词汇通达的二个理论,以及影响句子阅读的因素。在实验研究部分,以词语在句子中不同位置为切分点,把研究分成三个小实验,考察了音形义在判断句义中的作用,词语首尾变化对句子阅读的影响。
词汇通达理论主要有直通假设,语音转录假设,双通道假设。直通假设认为,在汉字识别过程中,字形直接激活了字义,而语音在词义通达中的作用不大,相应地,语音的获得是词义通达以后的一个附加过程。语音转录假设认为汉语阅读或汉字识别过程中,语音的激活是通达词汇的必要条件。即词义的获得必须把词形信息转换成语音,由语音激活词的意义,因此语音在语义的通达中起关键作用。双通道理论认为,汉语阅读和汉字识别过程中,词形通达词义与词形通达语音后通达语义两条道路并存,最终由哪条通道通达词义,这主要受到阅读材料,读者水平,任务要求等因素的影响。
句子理解的理论有独立模块理论和交互作用理论。交互作用理论认为句子理解的过程并不是几个模块独立作用的过程,也不存在句法分析优于语义分析,而是语义分析和句法分析之间相互作用。独立模块理论认为,句子的理解是由许多模块独立作用的结果,每一个模块负责自己相应部分的工作,以完成句子的理解。另外句子理解的三阶段认知神经模型指出句子理解的脑机制包括三个阶段,这个模型中的第一阶段和第二阶段其实就是一个句法加工的过程,第二阶段则包括句法分析和语义分析。影响句子阅读的因素主要有句法分析和语义分析,工作记忆广度,个体的句法意识、语音意识,句子类型,语境,词序等等。
本研究探讨的是不同位置的词语的音形义对句子阅读的影响,同时验证汉字音形义激活时间进程。本研究共三个实验,实验一是句首词语在判断句义中的作用，实验二是句中词语在判断句义中的作用,实验三是句尾词语在判断句义中的作用。实验结果发现：(1)汉词语音形义对句子阅读影响不一致,形近启动条件的反应时比本句启动,义近启动,音近启动个水平上的反应时都长。字形对句子阅读的影响最大,字音和字义对句子阅读的影响大小不能得到分离,可能实验材料有大。在已有的研究中发现,低频词汇中首先激活的是字形,而字音和字义的激活时同时完成的。(2)汉词语首尾变化对句子阅读的影响不一致,词语尾字变化条件的反应时比词语首字变化条件的反应时更长,也就足说词语尾字变化比词语首字变化对句子阅读的影响更大。(3)在句子中不同位置的词语对句子阅读有影响,但是句首.句中,句尾词语对句子阅读的影响不显著。也就是说在句子中的不同位置的词语对句子阅读的影响是一致的,这可能与实验材料有关,因为实验中的词语都是实词,对句子的影响是一致的。
The research of Sentence processing has been the focus of language psychologists, the factors of affecting the sentence processing is the question that researchers continue to explore.
This study first reviews three theories of the lexical access, and the factors that affect the sentence processing. In the section of the experimental study,words'different positions in the sentence for the cut-off point,the study were divided into three small test.Investigated the role oforthographic,phonological and semantic information in reading sentence, and the role of the head and tail of term in reading sentence.
Theory of lexical access mainly include:Phonological recoding hypothesis, direct access hypothesis,Dual-route views.Thedirect access theories assume the semantic of words is directly activated by the orthographic in the course of Chinese character recognition,no role of phonology in accessing themeaning of a word.phonology was not involved in thecognitive process of word identification.the phonologicalrecoding hypothesis proposes thatthe phonology activation is a necessary condition for lexical accessin the Chinese reading and character recognition process.the phonological form of a word isautomatically activated upon perceiving a word,and it mediatesaccess to the word's meaning. Both the visual route and thephonological route are responsible for accessing the mental lexicon.Different types of words are assumed to be processed by either of thetwo routes.Which channel accessible by the end, mainly by reading, the reader level, task requirements and other factors.
The theories of sentence comprehension include the modularity theories and interactive theories.Interactive theories is that the process of sentence comprehension is not an independent role in the course of several modules.There is not that the syntactic analysis is better than sema-ntic analysis, but the interaction between semantic analysis and syntactic analysis.Modularitytheories, the understanding of the sentence is the result of many independent modules,each module is responsible for the corresponding part of their work to complete the understanding of sentence.three-phase neurocognitive model propose that sentencecomprehension consists of three functionallydistinct phases:an initial parsing phase (phase 1),which precedes processes of thema -tic assignmentbased on semantic and morphosyntacticinformation(phase 2), and a late phase of revision during whichinteraction between semantic and syntacticinformation might take place (phase 3).the phase 1 and 3 are the syntactic analysis, the phase 2 is the semantic analysis and syntactic analysis.The main factors affecting reading are sentence parsing and semantic analysis, working memory span, the individual syntactic awareness, phonological awareness, sentence type, context, word order, etc.
In this study, the role of orthographic,phonological and semantic of the sentencedifferent positions in sentence comprehension,the time course of graphic,phonological,and semantic activation in Chinese character identification.The study insist of three experiment. The first experiment is about the role of term of the first of sentence in reading sentences. The second experiment is about the role of term of the middle of sentence in reading sentences. The last experiment is about the role of term of the end of sentence in reading sentences. The results are (1)the role of graphic,phonological,and semanticon the Reading of the sentence is inconsistent, The graphic have the greatest impact.The results may be related with the material.Low-frequency words in the first activation of the graphic, phonological and semantic activate at the same time.(2)Changes in Chinese words inclusive reading of the sentence is inconsistent,the last word have the greater impact than the first word. (3)Sentences in different locations to read the words of the sentence is consistent with the impact
本研究首先回顾了词汇通达的二个理论,以及影响句子阅读的因素。在实验研究部分,以词语在句子中不同位置为切分点,把研究分成三个小实验,考察了音形义在判断句义中的作用,词语首尾变化对句子阅读的影响。
词汇通达理论主要有直通假设,语音转录假设,双通道假设。直通假设认为,在汉字识别过程中,字形直接激活了字义,而语音在词义通达中的作用不大,相应地,语音的获得是词义通达以后的一个附加过程。语音转录假设认为汉语阅读或汉字识别过程中,语音的激活是通达词汇的必要条件。即词义的获得必须把词形信息转换成语音,由语音激活词的意义,因此语音在语义的通达中起关键作用。双通道理论认为,汉语阅读和汉字识别过程中,词形通达词义与词形通达语音后通达语义两条道路并存,最终由哪条通道通达词义,这主要受到阅读材料,读者水平,任务要求等因素的影响。
句子理解的理论有独立模块理论和交互作用理论。交互作用理论认为句子理解的过程并不是几个模块独立作用的过程,也不存在句法分析优于语义分析,而是语义分析和句法分析之间相互作用。独立模块理论认为,句子的理解是由许多模块独立作用的结果,每一个模块负责自己相应部分的工作,以完成句子的理解。另外句子理解的三阶段认知神经模型指出句子理解的脑机制包括三个阶段,这个模型中的第一阶段和第二阶段其实就是一个句法加工的过程,第二阶段则包括句法分析和语义分析。影响句子阅读的因素主要有句法分析和语义分析,工作记忆广度,个体的句法意识、语音意识,句子类型,语境,词序等等。
本研究探讨的是不同位置的词语的音形义对句子阅读的影响,同时验证汉字音形义激活时间进程。本研究共三个实验,实验一是句首词语在判断句义中的作用，实验二是句中词语在判断句义中的作用,实验三是句尾词语在判断句义中的作用。实验结果发现：(1)汉词语音形义对句子阅读影响不一致,形近启动条件的反应时比本句启动,义近启动,音近启动个水平上的反应时都长。字形对句子阅读的影响最大,字音和字义对句子阅读的影响大小不能得到分离,可能实验材料有大。在已有的研究中发现,低频词汇中首先激活的是字形,而字音和字义的激活时同时完成的。(2)汉词语首尾变化对句子阅读的影响不一致,词语尾字变化条件的反应时比词语首字变化条件的反应时更长,也就足说词语尾字变化比词语首字变化对句子阅读的影响更大。(3)在句子中不同位置的词语对句子阅读有影响,但是句首.句中,句尾词语对句子阅读的影响不显著。也就是说在句子中的不同位置的词语对句子阅读的影响是一致的,这可能与实验材料有关,因为实验中的词语都是实词,对句子的影响是一致的。
The research of Sentence processing has been the focus of language psychologists, the factors of affecting the sentence processing is the question that researchers continue to explore.
This study first reviews three theories of the lexical access, and the factors that affect the sentence processing. In the section of the experimental study,words'different positions in the sentence for the cut-off point,the study were divided into three small test.Investigated the role oforthographic,phonological and semantic information in reading sentence, and the role of the head and tail of term in reading sentence.
Theory of lexical access mainly include:Phonological recoding hypothesis, direct access hypothesis,Dual-route views.Thedirect access theories assume the semantic of words is directly activated by the orthographic in the course of Chinese character recognition,no role of phonology in accessing themeaning of a word.phonology was not involved in thecognitive process of word identification.the phonologicalrecoding hypothesis proposes thatthe phonology activation is a necessary condition for lexical accessin the Chinese reading and character recognition process.the phonological form of a word isautomatically activated upon perceiving a word,and it mediatesaccess to the word's meaning. Both the visual route and thephonological route are responsible for accessing the mental lexicon.Different types of words are assumed to be processed by either of thetwo routes.Which channel accessible by the end, mainly by reading, the reader level, task requirements and other factors.
The theories of sentence comprehension include the modularity theories and interactive theories.Interactive theories is that the process of sentence comprehension is not an independent role in the course of several modules.There is not that the syntactic analysis is better than sema-ntic analysis, but the interaction between semantic analysis and syntactic analysis.Modularitytheories, the understanding of the sentence is the result of many independent modules,each module is responsible for the corresponding part of their work to complete the understanding of sentence.three-phase neurocognitive model propose that sentencecomprehension consists of three functionallydistinct phases:an initial parsing phase (phase 1),which precedes processes of thema -tic assignmentbased on semantic and morphosyntacticinformation(phase 2), and a late phase of revision during whichinteraction between semantic and syntacticinformation might take place (phase 3).the phase 1 and 3 are the syntactic analysis, the phase 2 is the semantic analysis and syntactic analysis.The main factors affecting reading are sentence parsing and semantic analysis, working memory span, the individual syntactic awareness, phonological awareness, sentence type, context, word order, etc.
In this study, the role of orthographic,phonological and semantic of the sentencedifferent positions in sentence comprehension,the time course of graphic,phonological,and semantic activation in Chinese character identification.The study insist of three experiment. The first experiment is about the role of term of the first of sentence in reading sentences. The second experiment is about the role of term of the middle of sentence in reading sentences. The last experiment is about the role of term of the end of sentence in reading sentences. The results are (1)the role of graphic,phonological,and semanticon the Reading of the sentence is inconsistent, The graphic have the greatest impact.The results may be related with the material.Low-frequency words in the first activation of the graphic, phonological and semantic activate at the same time.(2)Changes in Chinese words inclusive reading of the sentence is inconsistent,the last word have the greater impact than the first word. (3)Sentences in different locations to read the words of the sentence is consistent with the impact
引文
[1]Chumbley J I, Ballota D A. Aword's meaning affects the decision in lexical decision. Memory&Cognition,1984
[2]Perfetti C A, Zhang s, Berent I. Reading in English and Chinese:Evidence for a"universal" phonological principle. In R Frost & Katz(Eds), Orthography, phonology, mophology, And meaning.Amsterdam:North-Holland,1992
[3]周晓林,语音与词义激活的相对时间进程：来自儿童发展的证据,心理科学,1998
[4]Perfetti CA, Tan LH.The time course of graph-ic, phonological,and semantic activation in Chinese character identification.Journal of Experimental Psychology:Learning,Memory, and Cognition.1998。
[5]武宁宁、舒华等.移动窗口下语音、字形信息在汉语阅读中的作用,心理学报,1998
[6]BesnerD.Phonology,lexical access in reading and articulatorysuppression:A critical review Quarterly Journal of ExpeamentalPsychology.1987,39A:467-478
[7]Taft M,Graan F V.Lack of phonological mediation in semantic categoazationtask.Journal of Memory and Language,1998,38:203-224
[8]McClelland J L,Rumelhart D E.An interactive activation model of context effects in letter perception:partⅠ.An account of basic findings. Psychological Review, 1981,88(5):375-407
[9]Paap K. R,Newsome S L,McDonald J E,et al.An activation-verification model for letter and word recognition:the word-superiority effect.Psychological Review, 1982,89(5):573-594
[10]Taft M,van GraanF.Lack of phonological mediation in a semantic categorization task.Journal of Memory and Language,1998,38(2):203-224
[11]莫雷.关于短时记忆编码方式的实验研究[J].心理学报,1986,2：166-173
[12]Leck,K.J.,Weekes,B.S.,&ChenMJ.Visual and phonological pathways to the lexicon:evidence from Chinese readers.Memory and Cognition,1995,23:468 476
[13]周晓林.语义激活中语音的有限作用.载彭聃龄等主编：汉语认知研究.山东教 育出版社,1997,159-194
[14]张武田,冯玲.汉字属性匹配与大脑半球的协同活动.心理学报,1998,4：129-135
[15]张清芳,杨玉英.汉语词汇产生中语义,字形和音韵激活的时间进程[J].心理学报,2004,1：1-8
[16]韩在柱,毕彦超.无需语音中介的阅读理解机制：来自一例汉语失语症个案的新证[J].中国科学(C辑生命科学),2009,3：279-286
[17]Rubensiein H,Lewis S S,Rubenstein M A.Evidence for Phonemic recoding in visual word recognition.Journal of Verbal Learning and Verbal Behavior,1971, 10:647-657
[18]Van Orden G C.A ROWS is a ROSE:Spelling,sound And reading. Memory and Cognition,1987,15:181-198
[19]张必隐,阅读心理学.北京师范大学出版社,2002：34
[20]Tzeng,O.J.L.,Hung,D.L.,&Wang,.W.S-Y.(1977).speech recording in reading chinesecharacters.Journal of experementalPsychology:Human Learning and Memory,3,621-630
[21]Perfetti.C.A.&zhang,S.phonologicalproeesses in reading Chinese Words.Journal of ExPerimentalPsychology:Learning,Memory and Cognition,1991,17:633-643
[22]Perfetti.C.A.&Zhang,S.very early Phonological activation in Chinese reading.Joumal of Experimental Psychology:Learning,Memory, and Cognition,1995,21:24-33
[23]张厚粲,舒华.汉字读音中的音似和形似启动效应[J].心理学报,1989,3：284-289
[24]张武田,冯玲,何海东.汉字识别中的语音效应[J].心理学报,1993,4：353-358
[25]郭桃梅,彭腆龄,祁志强,徐伦.语音的自动激活及其在汉字语义通达中的作用[J].心理学探新,2004,1：31-33
[26]闻素霞,热比古丽·白克力.语音在维—汉双语者汉字识别中的作用[J].心理与行为研究,2009,2：134-136
[27]Meyer,D.E.,&Schvaneveldt,R.W.,&Ruddy,M.G.(1974).Functions of graphemic and phonemic codes in visual word recognition.Memory and Cognition,2,309-321
[28]彭耽龄,郭德俊,张素兰.再认性同一判断中汉字信息的提取[J].心理学报,1985,3：227-234.
[29]林泳海,张必隐.中文阅读中的语音类似效果[J].心理学报,1999,1：21-27
[30]陈宝国,彭聃龄.汉字识别中形音义激活时间进程的研究(Ⅰ)[J].心理学报,2001,33(1)：1-6
[31]陈宝国,王立新,彭聃龄.汉字识别中形音义激活时间进程的研究(Ⅱ)[J].心理学报,2003,35(5)：576-581
[32]陈宝国,王立新,彭聃龄.高、低频汉字形音义激活的时间进程[J].心理与行为研究,2006,4(4)：252-257
[33]尹文刚.汉字失读的类型和意义[J].心理学报,1990,3：297-304
[34]谢秋幼,刘晓加,李卫平,吕庆文.汉语失写症的神经心理学研究及其意义[J].中国行为医学科学,2002，3：255-257
[35]Friederici A D.Towards a neural basis of auditory sentence processing [J].Trends in Cognitive Sciences,2002(6):78-84.
[36]DeDe,G,Caplan,D.,Kemtes.,&Waters,G.The relationship between age,verbal working memory,and language comprehension.Psychology and Aging[J], 2004,19(4):601-616
[37]陈庆荣,邓铸,蒋波.工作记忆和句法复杂性对汉语判断单句的影响[J]心理科学,2008,31(3)：633-637
[38]JustMA,CarpenterPA,KellerT.Workingmemory:Newfontiers of evidence [J].Psychological Review,1996,103(1):773-775
[39]井世洁.中学生的阅读理解能力与工作记忆的关系[J].宁波大学学报,2003,25(4)：26-29
[40]Plaza M,& Cohen H.The interaction betweenphonological processing, syntactic awareness, and namingnspeed in the reading and spelling performance of first-grade children[J].Brain and Cognition,2003,53:287-292
[41]Muter V,HulmeC,Snowling M J et al.Phonemes,rimes,vocabulary,and grammatical skills as foundationsof early reading development:evidence from alongtitudal study[J].Developmental Psychology,2004,40(5):665-681.
[42]陈宝国,陈雅丽.小学儿童句法意识、语音意识与阅读理解成绩的关系[J].心理科学,2008,31(4)：892-895
[43]陈永明,彭瑞祥.句子理解的实验研究[J].心理学报,1990(3)：255-231
[44]朱晓平.汉语句子语境对单阅识别的效应[J].心理学报,1991(2)：145-152
[45]鲁忠义,熊伟.汉语句子阅读理解中的语境效应[J]心理学羊报,2003,35(6)：726-733
[46]Severine,M.,Rene,A.,Wales,R.,&Christophe,A.(2008).PhonologicalPhrase Boun-daries Constrain the OnlineSyntactic Analysis of Spoken Sentences.Journal of Experimental Psychology:Learning,MemoryandCognition,34,874-885.
[2]Perfetti C A, Zhang s, Berent I. Reading in English and Chinese:Evidence for a"universal" phonological principle. In R Frost & Katz(Eds), Orthography, phonology, mophology, And meaning.Amsterdam:North-Holland,1992
[3]周晓林,语音与词义激活的相对时间进程：来自儿童发展的证据,心理科学,1998
[4]Perfetti CA, Tan LH.The time course of graph-ic, phonological,and semantic activation in Chinese character identification.Journal of Experimental Psychology:Learning,Memory, and Cognition.1998。
[5]武宁宁、舒华等.移动窗口下语音、字形信息在汉语阅读中的作用,心理学报,1998
[6]BesnerD.Phonology,lexical access in reading and articulatorysuppression:A critical review Quarterly Journal of ExpeamentalPsychology.1987,39A:467-478
[7]Taft M,Graan F V.Lack of phonological mediation in semantic categoazationtask.Journal of Memory and Language,1998,38:203-224
[8]McClelland J L,Rumelhart D E.An interactive activation model of context effects in letter perception:partⅠ.An account of basic findings. Psychological Review, 1981,88(5):375-407
[9]Paap K. R,Newsome S L,McDonald J E,et al.An activation-verification model for letter and word recognition:the word-superiority effect.Psychological Review, 1982,89(5):573-594
[10]Taft M,van GraanF.Lack of phonological mediation in a semantic categorization task.Journal of Memory and Language,1998,38(2):203-224
[11]莫雷.关于短时记忆编码方式的实验研究[J].心理学报,1986,2：166-173
[12]Leck,K.J.,Weekes,B.S.,&ChenMJ.Visual and phonological pathways to the lexicon:evidence from Chinese readers.Memory and Cognition,1995,23:468 476
[13]周晓林.语义激活中语音的有限作用.载彭聃龄等主编：汉语认知研究.山东教 育出版社,1997,159-194
[14]张武田,冯玲.汉字属性匹配与大脑半球的协同活动.心理学报,1998,4：129-135
[15]张清芳,杨玉英.汉语词汇产生中语义,字形和音韵激活的时间进程[J].心理学报,2004,1：1-8
[16]韩在柱,毕彦超.无需语音中介的阅读理解机制：来自一例汉语失语症个案的新证[J].中国科学(C辑生命科学),2009,3：279-286
[17]Rubensiein H,Lewis S S,Rubenstein M A.Evidence for Phonemic recoding in visual word recognition.Journal of Verbal Learning and Verbal Behavior,1971, 10:647-657
[18]Van Orden G C.A ROWS is a ROSE:Spelling,sound And reading. Memory and Cognition,1987,15:181-198
[19]张必隐,阅读心理学.北京师范大学出版社,2002：34
[20]Tzeng,O.J.L.,Hung,D.L.,&Wang,.W.S-Y.(1977).speech recording in reading chinesecharacters.Journal of experementalPsychology:Human Learning and Memory,3,621-630
[21]Perfetti.C.A.&zhang,S.phonologicalproeesses in reading Chinese Words.Journal of ExPerimentalPsychology:Learning,Memory and Cognition,1991,17:633-643
[22]Perfetti.C.A.&Zhang,S.very early Phonological activation in Chinese reading.Joumal of Experimental Psychology:Learning,Memory, and Cognition,1995,21:24-33
[23]张厚粲,舒华.汉字读音中的音似和形似启动效应[J].心理学报,1989,3：284-289
[24]张武田,冯玲,何海东.汉字识别中的语音效应[J].心理学报,1993,4：353-358
[25]郭桃梅,彭腆龄,祁志强,徐伦.语音的自动激活及其在汉字语义通达中的作用[J].心理学探新,2004,1：31-33
[26]闻素霞,热比古丽·白克力.语音在维—汉双语者汉字识别中的作用[J].心理与行为研究,2009,2：134-136
[27]Meyer,D.E.,&Schvaneveldt,R.W.,&Ruddy,M.G.(1974).Functions of graphemic and phonemic codes in visual word recognition.Memory and Cognition,2,309-321
[28]彭耽龄,郭德俊,张素兰.再认性同一判断中汉字信息的提取[J].心理学报,1985,3：227-234.
[29]林泳海,张必隐.中文阅读中的语音类似效果[J].心理学报,1999,1：21-27
[30]陈宝国,彭聃龄.汉字识别中形音义激活时间进程的研究(Ⅰ)[J].心理学报,2001,33(1)：1-6
[31]陈宝国,王立新,彭聃龄.汉字识别中形音义激活时间进程的研究(Ⅱ)[J].心理学报,2003,35(5)：576-581
[32]陈宝国,王立新,彭聃龄.高、低频汉字形音义激活的时间进程[J].心理与行为研究,2006,4(4)：252-257
[33]尹文刚.汉字失读的类型和意义[J].心理学报,1990,3：297-304
[34]谢秋幼,刘晓加,李卫平,吕庆文.汉语失写症的神经心理学研究及其意义[J].中国行为医学科学,2002，3：255-257
[35]Friederici A D.Towards a neural basis of auditory sentence processing [J].Trends in Cognitive Sciences,2002(6):78-84.
[36]DeDe,G,Caplan,D.,Kemtes.,&Waters,G.The relationship between age,verbal working memory,and language comprehension.Psychology and Aging[J], 2004,19(4):601-616
[37]陈庆荣,邓铸,蒋波.工作记忆和句法复杂性对汉语判断单句的影响[J]心理科学,2008,31(3)：633-637
[38]JustMA,CarpenterPA,KellerT.Workingmemory:Newfontiers of evidence [J].Psychological Review,1996,103(1):773-775
[39]井世洁.中学生的阅读理解能力与工作记忆的关系[J].宁波大学学报,2003,25(4)：26-29
[40]Plaza M,& Cohen H.The interaction betweenphonological processing, syntactic awareness, and namingnspeed in the reading and spelling performance of first-grade children[J].Brain and Cognition,2003,53:287-292
[41]Muter V,HulmeC,Snowling M J et al.Phonemes,rimes,vocabulary,and grammatical skills as foundationsof early reading development:evidence from alongtitudal study[J].Developmental Psychology,2004,40(5):665-681.
[42]陈宝国,陈雅丽.小学儿童句法意识、语音意识与阅读理解成绩的关系[J].心理科学,2008,31(4)：892-895
[43]陈永明,彭瑞祥.句子理解的实验研究[J].心理学报,1990(3)：255-231
[44]朱晓平.汉语句子语境对单阅识别的效应[J].心理学报,1991(2)：145-152
[45]鲁忠义,熊伟.汉语句子阅读理解中的语境效应[J]心理学羊报,2003,35(6)：726-733
[46]Severine,M.,Rene,A.,Wales,R.,&Christophe,A.(2008).PhonologicalPhrase Boun-daries Constrain the OnlineSyntactic Analysis of Spoken Sentences.Journal of Experimental Psychology:Learning,MemoryandCognition,34,874-885.