外语阅读中母语的作用和影响

2008-11-18 19:38 来源: 作者: 网友评论 0 条 浏览次数 2796

外语阅读中母语的作用和影响

The Effects of L1 Cognitive Resources on

L2 Reading

 

 

作者:吴诗玉

      导师:王同顺教授

 

Abstract

The focus of this dissertation concerns the effects of L1 cognitive resources on L2 reading. Based on previous studies, L1 cognitive resources in the present study refer to L1 reading ability, the ability in L1 mental-structure building, L1 cognitive use in L2 reading, and those general cognitive mechanisms and capacities which are closely related to the ability in L1 mental-structure building, such as the mechanism of suppression, the mechanism of enhancement, and working memory capacity.

        Test-based and product-oriented as well as VPA (verbal protocol analysis)-based and process-oriented experiments are integrated. Two test-based and product-oriented experiments were designed to address several questions. Experiment 1 investigates the effects of L1 reading ability on L2 reading by testing the hypothesis that Chinese reading ability is transferable to L2 reading despite the great differences between the two languages and evaluating the presence of “mutual compensation effects” in L2 reading between Chinese reading ability and L2 proficiency. Experiment 2 investigates, within the framework of Gernsbacher’s (1990) Structure Building Theory, the cognitive processes transferred and the cognitive mechanisms involved in the transfer. In addition, the cognitive effects on L2 reading are further explored regarding the psychocognitive nature of the transfer and the “threshold effect” proposed by several investigators. The first hypothesis was tested with a sample of 229 subjects, 60 of whom were chosen to participate in the second experiment. These experiments resulted in several findings:
  (1) Chinese reading ability is transferable to L2 reading although this association is weakened by the great differences between the two languages. Chinese reading ability proved to be one of the important predictors of L2 reading success; the present study predicts 14.2% of the variance in L2 reading scores. The present study also finds the existence of “mutual compensation effects” in L2 reading between Chinese reading ability and L2 proficiency; that is, a high Chinese reading ability will compensate for low L2 proficiency, and L2 proficiency will compensate for low Chinese reading ability. The existence of “mutual compensation effects” provided further evidence of the important effects of Chinese reading ability on L2 reading, although the compensation effects of Chinese reading were lower than those of L2 proficiency. In addition, the present study also supports the existence of a “threshold effect” in L2 reading; namely, that the reader has to reach a threshold of L2 language proficiency in order to be able to transfer his or her reading skills effectively from the L1 to the L2; otherwise, insufficient knowledge of the L2 would “short-circuit” the reader’s reading systems.
        (2) The present study provides empirical evidence in support of the hypothesis proposed within Gernsbacher’s (1990) Structure Building Framework: the transfer of reading comprehension skills is linked to mental-representations of the text; it is the building of mental-representations that has been transferred by those readers who had transferred their reading comprehension skills and it is the building of mental-representations that has failed to be transferred by those who had failed to transfer their reading comprehension skills with a limited L2 proficiency.

(3) The present study maintains that both “the threshold effect” and “mutual compensation effects” result from the combined effects of linguistic factors with cognitive mechanisms and capacities such as the mechanism of suppression, the mechanism of enhancement, and working memory.

(4) The present study demonstrates that both automaticity and control are important in L2 reading. The more advanced phase of skilled performance in L2 reading is not automaticity but control.

    Altogether, 21 subjects participated in the VPA-based and process-oriented experiment. This experiment was intended to test “the threshold effect” hypothesis in L2 reading from a process perspective; secondly, it was designed to investigate the role of L1 cognitive use in L2 reading using Vygotsky’s (1986) sociocultural model.
From the process perspective, the data lends support to the existence of “a threshold effect”. Secondly, it supports Upton’s (2001) conclusion that L2 reading is not a monolingual event; L2 readers have access to their first language as they read, and many use it as a strategy to help comprehend an L2 text. It further suggests that L2 learners construct a scaffold on an intrapsychological, or cognitive plane by using their own expertise in their L1 as a means of pushing their L2 competence beyond its current level.

The following findings concerning the role of L1 cognitive use in L2 reading were yielded from this process-oriented experiment:

    (1) The cognitive use of the L1 in L2 reading positively facilitates comprehension for all students regardless of their linguistic level, giving them an additional avenue for processing L2 text.
    (2) The overall use and success of L1 in aid to L2 reading is clearly determined by the readers’ overall L2 proficiency. With the increase of English proficiency, on the one hand, the readers’ reliance on Chinese (L1) declined, its supportive uses were improved. On the other hand, the proportion of the strategies (Type A strategy identified by Pressley and Afflerbach (1990)) used for identifying ideas and constructing meaning of L2 text declined, but for conscious inference-making and cognitive and metacognitive monitoring increased.
   (3) Distinctive patterns of L1 cognitive use among Chinese EFL learners were used to identify ideas and construct meaning of L2 text. The present study finds that L1 was used not merely for mental translation; it also helped readers to monitor their reading and evaluate what they were reading.
    This study provides several theoretical, methodological and pedagogical insights. By using the theoretical framework developed by cognitive psychology, it helps us to understand the multidimensional nature of L2 reading. Through the analysis of L1 cognitive use in L2 reading, this study suggests that Vygotskyan sociocultural theory, which recognizes language as the facilitator of thought, is a suitable framework around which to explore L2 reading. Methodologically, the present study suggests that L2 studies of reading require combined approaches. The adoption of both test-based and product-oriented as well as VPA-based and process-oriented methods together with multivariate statistical analyses offers an effective option. Pedagogically, it suggests that Chinese EFL learners would find it rewarding to make active use of L1 cognitive resources in L2 learning. Firstly, it suggests that such learners make active use of those reading strategies, reading skills and the relevant background knowledge acquired in L1. It further suggests that the time and effort directed to reading comprehension skills at lower intermediate levels, which many or most learners have already possessed in L1 might not be worthwhile. It also suggests that it is advisable to maintain L1 reading, especially for those adult L2 readers who have achieved a certain level of L1 literacy. Finally, teachers should emphasize the importance of students’ acquiring high L2 language proficiency to read well in L2.
    This study also raises a number of interesting possibilities for future investigation. For example, how well do the results reported here generalize to other populations? It would also be useful to collect data in order to do a path analysis on linguistic cognitive interactions to further investigate the complex nature of “the threshold effect” and “mutual compensation effects” in L2 reading and their interactions. Future research could also explore other approaches to data analysis. For example the use of “response surface analyses” as proposed by Pichette & Segalowitz (2003) to locate threshold points and values of L1 reading ability and L2 proficiency variables associated with optimal L2 reading success.

  

Keywords: L2 reading, L1 reading, transfer, the threshold effect, mental-representations, mechanism of suppression, working memory, L1 cognitive use

 
 

摘要

本论文在迁移理论、结构建造模型和社会文化理论的框架下,研究了一语认知资源对外语阅读的影响和贡献。根据已有研究及理论(如,Upton, 2001; Yamashita, 2002b; Pichette & Segalowitz, 2003),一语认知资源在本文中被定义为一语阅读能力、一语心理表征建构能力、一语在外语阅读中的认知使用等。此外还探讨了心理表征建构的中心抑制机制及工作记忆能力等对外语阅读的影响。

为了阐述研究问题,本论文综合了以测试为基础、以结果为导向 (test-based and product-oriented) 和以口头原案分析为基础、以过程为导向 (VPA-based and process-oriented)的两种实验。

以测试为基础、以结果为导向的实验总共开展了两组。实验1:主要研究汉语阅读能力是否能向外语阅读迁移,同时,检测汉语阅读能力与英语水平在外语阅读中是否存在“优劣互补效应”(“mutual compensation effects”),从而阐述汉语阅读能力对外语阅读的重要影响和作用。其中还通过检测“阈值效应”(“the threshold effect”)存在与否,以研究汉语阅读能力影响外语阅读的条件。实验2:运用Gernsbacher (1990)的结构建构框架,从认知心理的角度研究了阅读理解过程中的认知过程的迁移,通过研究从中涉及到的各种认知机制,从而进一步探讨阅读技能迁移及迁移中“阈值效应”的认知心理本质。共有229名被试参加实验1的测试,从中再挑选60名被试参加实验2。这些研究和实验得到了如下重要结论:

(1)虽然这两种语言间巨大的差别对两者的关联造成了一定的影响,但是汉语阅读能力确实会向英语阅读迁移;汉语阅读能力被证明是外语阅读的重要预测因素之一,在本实验里,它可以预测外语阅读分数的14.2%的 变异。本研究也证实了汉语阅读能力与英语水平在外语阅读中“优劣互补效应”的存在,即汉语阅读能力高能补偿英语水平低的影响,英语水平高也可以补偿汉语阅 读能力低的影响。尽管实际数据表明汉语阅读能力的补偿作用比英语水平低,但这种“优劣互补效应”的存在进一步表明了汉语阅读能力对外语阅读的重要作用。另 外,实验也验证了外语阅读中“阈值效应”的存在:即要实现汉语阅读能力有效地向外语阅读迁移,阅读者的外语水平必须达到阈值的要求;否则,外语知识的缺乏 会对读者的阅读体系造成“短路”。而一旦读者的外语水平达到阈值后,迁移就会发生,汉语阅读能力对外语阅读的影响和作用也将逐步增加。

(2)从认知心理的角度看,外语阅读理解能力的迁移与心理表征建构紧密联系;外语阅读理解能力的迁移也是读者心理表征建构的迁移,成功的迁移了汉语阅读理解能力的 读者,从认知心理的角度看,也就是心理表征建构的成功迁移;反之,则是心理表征建构能力的迁移失败。由此,本研究证明,从认知心理的角度分析,外语阅读中 的“阈值效应”及“优劣互补效应”,应都是由语言水平与各种认知机制,如抑制机制、工作记忆能力等因素综合作用的结果。

(3)自动化(automaticity)与控制(control or suppression)是二语阅读熟练表现(skilled performance)的两个方面,然而二语阅读熟练表现的更高阶段不是自动化,而是控制。

本研究另外挑选了21名被试参加以口头原案分析为基础、以过程为导向的实验。这是在中国的英语学习环境下,在Upton (2001)实验基础上,做的一个后续研究。此实验意图之一即是从阅读过程的角度,进一步检验“阈值效应”存在与否;更为重要的是,本实验意图在Vygotsky (1986)的社会文化理论的框架下,研究一语在外语阅读中的作用。此实验的重要结论阐述如下:

首先,从阅读过程的角度,它进一步证实了“阈值效应”的存在。

其次,也是更为重要的是,它证实了Upton (2001)的结论:即外语阅读已不再是一单语言事件(monolingual event),用Koda (2005)的话来说即是“双语涉及”(dual-language involvement)事件。研究结论表明,外语阅读时,读者的母语通道总是开放的。阅读中,读者常常会把母语当作一种策略,来帮助理解外语阅读材料。外语学习者利用他们擅长的母语,创造一个内在的心理或认知的平台 (intrapsychological, or cognitive plane);在这个平台上(“最佳发展区”),把他们的外语能力(阅读能力)提升到超过目前的水平。关于一语在外语阅读中认知使用的作用可以概括如下:

(1)       尽管会因为读者的外语水平不同,母语对外语阅读的帮助性作用存在差异,但从总体上看,母语的使用会对读者的阅读理解起到一定的帮助作用,对学生来说,这相当于给了他们一种额外的办法来处理正在阅读的文章。

(2)       本实验清晰地表明,对汉语的整体使用和有帮助性的使用由读者的整体外语水平决定。随着外语水平的提高,一方面,汉语帮助理解的作用随之增加,但读者对汉语依赖性降低;另一方面,读者运用汉语识别文章思想、构建意义比重有所减少,但运用作阅读推论及监控比重有所增加。

(3)       读 者运用汉语来识别文章思想、构建意义时,对汉语的使用呈现出清晰的模式。概括起来,可能会从八个方面运用汉语来理解文章思想:其一,读者只是用汉语简单的 把一个单词或一个短语的意思翻译过来;其二,用汉语对单词或短语的意思表示不确定、猜疑,并且运用汉语帮助理解意思;其三,用汉语来概括、证实对超过词 汇、短语范畴更大语块意思的理解;其四,用汉语来猜疑、揣测或者排除超过短语范畴的更大语块的意思;其五,用汉语来试图预测或者预期文章结构和内容;其 六,运用 “这()”或“什么”等汉语词汇来代替不认识的词汇、短语或者更大语块的意思,来构建理解或者作出推论;其七,用汉语联系头脑里已有知识,以理解抽象复杂的概念、场景;其八,用汉语来维持或者保持已经对文章形成的理解,从而形成连贯的建构。

本研究在理论、方法和教学上都提供了一定的启示作用。理论上,本研究表明,运用认知心理学的理论和发现来研究外语(或二语)阅读,可能会帮助我们更加深刻本质地理解二语阅读这个复杂的、多维建构的本质。同时通过研究一语在二语阅读中的认知使用,本研究表明,Vygotsky(1986)的社会文化理论,它把语言当作思维的利器 (facilitator), 是用来研究外语阅读的一个合适的理论框架。方法上,本研究表明,综合运用以结果和以过程为导向的方法来研究外语阅读这个复杂的认知心理过程,往往能得到更 为客观全面地理解。本研究综合以测试为基础、以结果为导向和以口头原案分析为基础、以过程为导向的研究,再结合统计软件SPSS的运用,比较有效的阐明了研究问题。教学上,本研究表明,在外语阅读中,读者应该积极的运用他们在一语中所习得的有效阅读策略、技能和相关的背景知识。教师把时间花在教授读者在一语已经掌握的策略和方法上显得多余和低效率。同时,学习者应该在平时保持对一语的积极阅读,尤其是对成年的外语学习者,他们不必等到外语水平达到很高时,才有效地运用一语资源,从中受益。最后,教师还应该强调语言水平对外语阅读成功的重要性。

由于外语阅读本身的复杂性,本研究在概念、方法和其它方面还存在一定的局限性。首先,对影响外语阅读的外语水平的定义,可以包括更多的相关语言成份;其次,在样本大小方面,尤其是以口头原案分析为基础、以过程导向的实验里,本研究只选了21名被试,这对研究结论的推广会产生一定的影响;再次,在实验工具上可以得到进一步的提高和改善,也可以提供相应的补充方法。

本 研究还为未来的研究提供了方向。比如,收集具体的数据,做路径分析,以研究语言水平是如何与各认知因素及认知能力共同作用,从而产生“阈值效应”,并最终 影响阅读技能的迁移和外语阅读。另外,在研究外语阅读的影响变量时,还可以包括对更多因素的考察,比如动机等情感因素的影响等。对数据处理还可以运用其它 方法,如Pichette & Segalowitz (2003)提出的“曲面反应分析”(surface response analysis)方法来研究变量之间的关系,具体定义影响阅读技能迁移的阈值。

 

关键词:阅读,迁移,阈值效应,心理表征,抑制机制,工作记忆,认知使用

 

(注:本论文已由上海交通大学出版社出版发行,要全文者请与版主联系)

上一篇:交际策略研究及应..    下一篇:AN INVESTIGATION..

相关主题:

网友评论