An Empirical Study On Conceptual Transfer In Second Language Vocabulary Acquisition

Language transfer,or crosslinguistic influence,refers to the influence of a person's knowledge of one language on the knowledge or use of another language.When the influence mainly occurs at the conceptual level,it is called conceptual transfer.In 1998,Pavlenko and Jarvis respectively proposed the term “conceptual transfer” and relevant theories.Since then,related research has developed fast and become a central issue in the study of second language acquisition.Most of the studies were conducted within the framework of the eight foundational domains of reference but there was a lack of studies concerning lexical conceptual transfer.In Jarvis and Pavlenko's framework,there was no systematic discussion on conceptual transfer in relation to words except for a rough distinction among linguistic,semantic,and conceptual transfer.This results in a tendency of over-generalization of conceptual transfer.To solve this problem,the present research attempts to propose a lexical transfer framework featuring linguistic transfer(LT),semantic transfer(ST),and two types of conceptual transfer(CT1 and CT2),each with an original definition.The construction of the framework has borrowed insights from Evans'(2009)theory of Lexical Concepts and Cognitive Models(LCCM),which provides the grounds for the distinction among the above-mentioned four types of lexical transfer: they occur in different processes of meaning construction and through different mechanisms.To examine these hypotheses,the present study intends to answer three research questions:1.Can the hypothesized distinction be verified among the four types of transfer in the proposed framework through automaticity measurement?2.What is the relationship between learners' vocabulary size and lexical transfer?3.What is the understanding and attitude of Chinese learners of English towards conceptual system and conceptual transfer?Among the three questions,the first is the major exploration and will be broken down into subsequent questions.To answer these questions,the present study adopted a mixed methods research.Experimental automaticity measurement was conducted to collect quantitative data,and stimulated recall and interview were employed for qualitative data.Automaticity statistics showed a person's efficiency in processing and accommodation of information when responding to stimuli,whose main indicator was reaction time(RT).The present study also recorded accuracy(ACC)and coefficient of variation(CVrt).The statistics from these three aspects can indirectly reflect the mental processes of transfer.To enhance the study with more direct observations,the participants went straight into the stimulated recalls and interviews after the experiments.They reflected on their thinking during the tasks with the guidance from the researcher to unveil phenomena undetectable by automaticity data.The participants were also encouraged to discuss their meta-knowledge of conceptual system and conceptual transfer.After pilot tests and modifications,the whole study procedure was performed twice with 40 participants selected from first-year English majors in a “211” university in2019 and 2020.The experiments were conducted in an individualistic manner and the automaticity measurement was divided into a pretest and a posttest with a treatment of providing definitions of stimulus words.Overall,each participant completed 64 tasks.The stimulus words had been discussed and selected after discussion between the researcher and a native English speaker as well as some non-participant students.After the measurement,the researcher and the participant jointly performed the stimulated recall and the interview in Chinese.Live recording was made with the participants' consent.The quantitative data were collected and analyzed statistically and the qualitative data were selected and transcribed.Based on the interpretation of the results from data analysis,the answers to the three research questions are:On research question 1:(1)The proposed framework of lexical transfer was sustained.The distinction among the four types of transfer was first supported statistically in the one-way ANOVA test,especially in terms of ACC and RT.Then in the stimulated recalls and interviews,the participants' retrospection further attested the differences.(2)Tukey post hoc procedures indicated that the variance among the four types of transfer existed.LT and CT2 were more distinct while ST and CT1 followed a similar statistical pattern.This indicated that the areas of occurrence and mechanisms between ST and CT1 were close.The stimulated recalls and interviews backed this finding.(3)Between the pretest and the posttest,the variance of ACC of the four types of transfer increased drastically to significance level.However,this only proved the improvement in the participants' intellectual understanding,but not the restructuring of the conceptual system,because the CVrt between the pretest and the posttest showed little difference.(4)With regard to each type of lexical transfer: 1)LT's ACC dramatically increased in the posttest was a sign of the transfer's “superficiality” in that it was caused by the mis-matching of vehicles purely from the two symbolic systems.2)ST occurred mainly during lexical concept integration.For instance,when the multiple senses of a word were from different semantic domains,i.e.“true polysemy” that were distinct between L1 and L2 conceptual systems,transfer arose due to the mistaken “mapping”.3)CT1 might,too,occurred in the lexical concepts during interpretation process.It contained two subtypes:one caused by “pseudo-polysemy”,a L2 word's multiple senses evolved from the original one through metaphor or metonymy,the other caused by languagemediated concepts or concepts with language-induced features.These hypotheses were not supported by statistical analyses,but found their evidence through stimulated recalls and interviews.4)CT2 was caused by the languageunique concepts resulting from unique characters of history,culture,thinking patterns,emotions or other extralinguistic elements.CT2 was most elusive,stable,and therefore most difficult to avoid.This position was supported by quantitative and qualitative data with very strong evidence.(5)Finally,Pearson Correlation results indicated that there was no significant correlation among statistics from types of transfer,except that RT and CVrt were negatively correlated in certain cases.On research question 2:(1)Vocabulary size had significant predictive power of all types of statistics related to lexical transfer.It successfully predicted the ACC,RT,and CVrt from both the pretest and the posttest.(2)The above conclusion was valid only on the assumption that vocabulary size was defined as a combination of both the size of receptive vocabulary and the size of productive vocabulary.When the independent variable was set to be the intermediate and advanced groups based merely on receptive vocabulary,no prediction could be made.(3)Multiple Linear Regression indicated that the mechanism behind the joint work of receptive and productive vocabulary was complex and dynamic.On research question 3:(1)During the interviews,most participants were able to describe their rich experience in lexical transfer.Some advanced learners could even explicitly talk about their thoughts of differences between Chinese and English conceptual systems.(2)The participants' conceptual system had exhibited features of interlanguage and been mixed with characters of English conceptual system.(3)Although the participants agreed on the need of attention towards lexical transfer,many doubted the usefulness of formal instruction on it.Instead,they believed that the avoidance of negative transfer should better be achieved through long-term self-reflection and observation.(4)The habit of consulting English-English dictionaries and reading English definitions were essential to the restructuring of a conceptual system that was compatible with both Chinese and English.More contact with pragmatic scenarios of English was also indispensable.The implications by the findings for L2 learning and teaching lie in four aspects.First,the conceptual system underpins our cognition and daily use of language.It is stable but not unchangeable.Through painstaking attention and keen observation,it is possible for L2 learners to reconstruct their conceptual system to suit both L1 and L2.Second,the four types of transfer,though distinct from each other,can present equal size of obstacles and challenges.Extensive care is required in L2 learning and innovative design of teaching and training are needed to better suit cognitive mechanisms.Third,the development in lexical competence should be made through efforts of multiple abilities.Narrow understanding of lexical competence as equal to the size of receptive vocabulary is misleading.The habit of reading English definitions and learning new words with contexts are most agreed means of conceptual restructuring.Finally,one key implication is the urgency of instructing students to use mobile network and smartphones properly to sensitize themselves to the L2 concepts and conceptual system.The present study is innovative in theoretical construction,research methodology,and pedagogical implications but suffers the limitations of unitary source of sampling,brevity of the interviews,and a lack of diachronic study.Future studies are advised to make improvements in these aspects.