On The Cognitive Mechanism Of Second Language Lexical Retrieval Preference

In recent years,there has been an increasing concern about the output-based language processing in the second language acquisition research.The present study focuses on the phenomenon of lexical retrieval preference(LRP)in L2 production,with an inquiry of the cognitive mechanism of the homogeneous tendency in L2 learners' retrieval of the targets from the paradigmatic relation.It aims to reveal the effect of the interaction between the learners'interlanguage system and cognitive processing system on L2 lexical retrieval.The study centers on an essential question,that is,why L2 LRP is a common phenomenon in L2 production.In order to answer this question,the thesis breaks it into another three ones as follows:(1)What is the nature of L2 LRP?(2)What kind of theoretic framework agrees with the exploration of that nature?(3)How is the framework-based account well established?According to the literature review,it is found that L2 LRP is,in nature,a systematic variation of learner language in L2 production,particularly the variation of the L2 learners'lexical selection from the paradigmatic relation.It is a homogeneous linguistic product as well as a homogeneous psychological cognitive process.However,the previous studies fail to reveal the cognitive mechanism of the process fully.It is manifested in following three aspects:the absence of a rational unified interpretation for the various manifestations of L2 LRP;the insufficient exploration of the complexity and dynamism of lexico-semantic links at the micro-level,and the less attention to the homogeneity of lexico-semantic processing.Therefore,based on the rationale of L2 acquisitional complexity,the thesis takes a full account of the interactive effect between lexical representation,lexical processing and lexical monitoring in lexical retrieval,and constructs a unified triadic interpretation model for L2 LRP.In light of the model,three primacy principles are proposed to guide the subsequent researches,including the primacy of salience principle,the primacy of efficiency principle,and the primacy of broad semantic specification satisfaction principle.In order to testify the rationality of the three principles,with Chinese-speaking learners or users of English as participants,the thesis carried out three specific studies,each with a focus on one principle and the relevant hypotheses.Study 1 examines the relation between the primacy of salience and L2 LRP.Three tasks were designed to test the relevant hypotheses,including Lexical Priming Test,Multiple Choice Test,and Sentence Cloze Test.The research results showed that the words in the mental lexicon were hierarchically distributed,in which the learners preferred to retrieve the salient ones in that hierarchy.In addition,it was found that the learners did not prefer to avoid phrasal verbs in certain task because it depended on to what extent the distinctive semantic features of the target lexical units could be created through the semantic cues and perceived by the learners.Study 2 examines the relation between the primacy of processing efficiency and L2 LRP.Two tasks were designed to test the relevant hypotheses,including Translation-induced Sentence Cloze Test and Chinese Semantic Priming Test.The research results demonstrated that both advanced and intermediate Chinese learners of English preferred the analytical processing mode to retrieve the English targets,and no significant difference between groups was found.It was also demonstrated that there existed a difficulty hierarchy in the holistic processing of the English targets of different word classes such as verbs,nouns,adverbs and adjectives;that is,the difficulty graded up from the verbs to the adjectives.In addition,it was found that the learners produced more non-lexicalized words in the obligatory context than in the optional context for holistic processing,but some targets were both unproduced in the two contexts,which indicated the incomplete attainment in the acquisition of the depth knowledge of the relevant targets.Study 3 examines the relation between the primacy of broad semantic specification satisfaction and L2 LRP.Two tasks were also designed to test the relevant hypotheses,including Chinese Lexical Priming Test and Translation-induced Sentence Cloze Test.The research results manifested that the learners were more inclined to retrieval the targets in broad semantic specification satisfaction than those in narrow semantic specification satisfaction,which has good consistency between different learner types,as well as between different task types.It was also found that there existed difficulty hierarchy in the retrieval of the targets from the paradigmatic relation with different semantic relations,which is closely related to the general developmental path of depth knowledge in form-meaning or form-meaning-function connection.The studies above have testified the relevant hypotheses and also justified the three primacy principles.That is,the more salient the stored form-meaning unit,the more ready it is to be activated into the range of the candidate words for processing.The more effortless the language processing mode is,the more ready it is to be activated for processing those words.And the broader semantic satisfaction specification the candidate word has,the more ready it is to pass the satisfaction check in a given task.The thesis reveals that L2 LRP emerges from the interadaptedness of the three primacy principles exactly through the interaction between the learners' interlanguage system and their cognitive processing.and more precisely,under time pressure,when motivated by a given lexical production task,L2 learners are inclined to adopt the effortless cognitive processing mode to retrieve the most salient target words from the alternatives of graded salience in the mental lexicon,with broader semantic specification passing through the satisfaction check in the task,thus bringing about the efficient and automatic production of the target words.As the learners share the similar graded salience in their interlanguage systems and cognitive processing systems,L2 LRP becomes a common phenomenon in L2 production.So far,the basic question of the study has been answered.In short,the triadic model formulated in this thesis has provided a unified interpretation for the different manifestations of L2 LRP.It deepens our understanding of the cognitive mechanism behind the homogenous language production.It is also believed that the study is instructive for the development of lexical complexity,accuracy and fluency in L2 production as well as for the future exploration of learners' preference of the form-meaning connection at syntactic or discoursal level in L2 production.