| Lexical inferencing is a strategy of tackling unknown words encountered in the process of language use. The present study explores the lexical inferencing process of university English majors in China. It is aimed at finding out what factors contribute to the success of lexical inferencing, under what circumstances learners feel difficult to infer the meaning of an unknown word, and what mental activities are involved in the lexical inferenicng process. Specifically, the present study addresses four questions:(1) What is the relationship between leamer-related factors and lexical inferencing? (2) How do text-related factors affect lexical inferencing? (3) What mental activities are involved in L2 lexical inferencing process? (4) What is the nature of L2 lexical inferencing?The design of the research is descriptive and exploratory. It is intended to describe relationships among factors involved in L2 lexical inferencing process and reveal cognitive patterns as reflected in learners’ lexical inferencing behavior. Participants of the study were 159 university English majors, including 81 sophomores and 78 seniors. Four reading passages were selected,2 from Readers’Digest (one more familiar, one less familiar, for the sophomores), and 2 from Time (one more familiar, one less familiar, for the seniors). The participants were required to read the articles, infer the meanings of unknown words, and then indicate the degree of difficulty in inferring the meaning of each target word on a 6-point scale.Learner-related factors examined in the study included learners’ L2 proficiency and background knowledge specific to the topic in a target reading passage, referred to as topic familiarity. Instruments used to measure these factors consist of the Vocabulary Levels Test, Word Associates Test, Passage Sight Vocabulary Test, Reading Comprehension Test, and a 5-point topic familiarity scale. SPSS 13.0 was used as the analytical tool. More specifically, to find out the relationship between learners’ L2 proficiency and lexical inferencing, Pearson Correlation was conducted. Topic familiarity was regarded as a within-subject variable, and to determine its impact on lexical inferencing, one-way ANCOVA was performed.Two text-related factors, i.e. morphological transparency and strength of contextual support, were investigated in the study. The Morphological Transparency Scale and the Contextual Support Scale were the measurement instruments. Based on the former scale, all the target words were divided into two groups:1= morphologically opaque words,2= morphologically transparent words; according to the latter scale which measured the richness of clues in the surrounding context, all the target words were classified into three categories:1= words with nondirective context,2= words with general context,3= words with directive context. In this way, each target word has two values, one indicating its degree of morphological transparency, the other showing how revealing its surrounding context is in terms of its meaning. The statistical method, Univariate Analysis of Variance, was used to determine the main effects and interaction of the two factors. Multiple Regression Analysis was also conducted to find out their combined and individual contribution to the variance in lexical inferencing.A retrospective study was also conducted after the main study. Twenty of the students participated in the retrospection immediately after the lexical inferencing task. The process was recorded and transcribed. Retrospective protocols were developed and analyzed to identify knowledge sources used in the process of lexical inferencing and to reveal frequency patterns of use of these knowledge sources.The results of the study show that there is significant correlation between learner-related factors and lexical inferencing success, and this relationship remains consistent at different levels of L2 proficiency. But the relationship between the learner-related factors and the feeling of ease/difficulty in lexical inferencing is not the same for sophomores and seniors. This variance may be caused by the higher difficulty level of the reading materials for the senior students.Of the two text-related factors examined in the study, i.e. morphological transparency of target words and the strength of contextual support, only the latter significantly affects lexical inferencing success. This is true for both sophomores and seniors. However, the impact of these text-related factors on lexical inferencing difficulty is not consistent for the two groups of students. For the sophomores, contextual support affects their lexical inferencing difficulty rating significantly, but for the senior students, it is morphological transparency that has a significant impact on lexical inferencing difficulty ratings. Further analyses show that this phenomenon may be caused by the overall difficulty level of the reading materials for the seniors. When the language is complex in the surrounding context of an unknown word, recognizable word parts will become the most important knowledge source available to help deciphering the meaning of the word, so their presence or absence will determine students’perceived difficulty in lexical inferencing.The analysis of the retrospective data shows that students use a variety of knowledge sources in inferring meanings of target words, ranging from linguistic to non-linguistic. Several major frequency patterns of knowledge source use are revealed:(a) Linguistic sources are more frequently used than non-linguistic sources; (b) Intra-lingual sources are more often employed than inter-lingual sources; (c) Sentence level sources are used most, followed by word level sources, and then discourse level ones; (d) Sentence meaning is most frequently used, followed by word morphology, discourse meaning, world knowledge, and sentence grammar. The analysis of the retrospective protocols also suggests that lexical inferencing is a hypothesis-generating-and-testing process and the outcome of this process is the lemma part of a word.The present research has both theoretical and pedagogical values. Theoretically, it provides a tentative framework for future research and also sheds new light on the cognitive process of lexical inferencing. Pedagogically, it has the following implications:1) it is necessary to establish a thorough vocabulary-learning program that integrates extensive exposure to the target language with direct and systematic vocabulary instruction; 2) building up an extensive network of background knowledge should be a never-ending job of language learners; 3) it may be advisable for teachers to de-emphasize morphological analysis and to instruct their students to verify their guesses by checking the context; 4) it is essential to understand how to use the established LI conceptual system that exists in the L2 learners’mind as a facilitating factor rather than a hindrance. Keywords:lexical inferencing, L2 proficiency, topic familiarity, morphological transparency, strength of contextual support... |
PDF Full Text Request