A Study Of Task Involvement Loads In Incidental L2 Lexical Acquisition

The facilitative effect of reading on word learning has been an issue that invites abundant research efforts from a lot of scholars. When some researchers hold positive opinion on the effect of reading on vocabulary learning, based on the importance of internalization of L2 lexical knowledge, some others are doubtful on the ground that when learners are able to guess an unknown word successfully, they are less likely to learn the word because they are able to comprehend the text without knowing that word.While researchers diverge on the effects of reading on lexical retention, reading plus tasks is unanimously accepted to be an effective approach in L2 lexical acquisition. Second language research is rich in literature that holds tasks after reading can significantly facilitate retention. But still, how are these tasks different from each other in the way that they enhance vocabulary retention? Many scholars have proposed the concept of "depth of processing", but it actually is a vague concept with hardly any criteria to measure at all.This study aims to set up an objective criterion for evaluating the effects of different tasks in L2 lexical acquisition.The Task Involvement Components Model proposed by Laufer and Hulstijn in 2001, which is composed of three elements facilitating lexical acquisition, need, search and evaluation, is the first endeavor to take the challenge of classifying instructional tasks in terms of their vocabulary learning effectiveness. Need is the necessity to consult for and obtain the word meaning, including both an emotional need like motivation and attitude, and a cognitive need to figure out the word's meaning; search is the process of looking up new words in a dictionary; and evaluation is the process of choosing one item from all the word's meanings or usages that fits a specific language or environmental context. Their embodiment decides the effectiveness of different tasks.Hulstijn and Laufer's model is two dimensional, affective and cognitive. This study re-defines factors in their model and restricts components to cognitive category. A major contribution of this research is that it proposes and tests two other cognitive factors that are supposed to exert significant facilitative effects on L2 lexical acquisition: inference and generation.Drawing inferences from what people observe is fundamental to thinking, and the same principle can be used in the reading process. When learners search for and ascertain the relevant cues in the text and relate these cues to encyclopedic knowledge, and then combine the cues into a unified mental representation of the meaning of the word, word forms as well as semantic and other lexical information should be better retained.Generation in output activities is also supposed to be conducive to lexical retention. Learning a new word involves an ongoing elaboration of knowledge about the word and the ability to use it. Not only the meaning but the part of speech of the new word, the countability, the number or the tense and voice that it should be used in are all factors involved in the acquisition of new L2 vocabulary. In output activities, learners move focus of their attention towards not only the semantic features of new words, but also towards collocation with other words, and other specific syntactic features. In this way, more lexical information is integrated into existing knowledge through learner-initiated use of the new knowledge in production and some later encounters that help the lexicon mature.To test the effects of proposed components, four hypotheses are proposed as follow:Hypothesis 1: Inference has a significantly positive effect on lexicalretention.Hypothesis 2: Generation has a significantly positive effect on lexicalretention.Hypothesis 3: Generation exerts a more significant effect on lexicalretention than inference.Hypothesis 4: The facilitating effect on lexical retention of inference plusgeneration is greater than inference or generation alone.Three experiments are conducted to test the above four hypotheses.Experiment 1: Comparison between Task 1 (control group) that involves three components: need, search, and evaluation and Task 2 (experimental group) with need, search, evaluation and inference.Experiment 2: Comparison between Task 3 (control group) that involves three components: need, search, and evaluation (Task 3 = Task 1), and Task 4 (experimental group) with need, search, and generation.Experiment 3: Comparison between Task 5 with need, search, evaluation and inference (Task 5 = Task 2), Task 6 with need, search, and generation (Task 6 = Task 4), and Task 7 with need, search, evaluation and inference plus generation.Subjects are 210 students from Fudan University, containing 63, 69, and 78 students for three experiments respectively, randomly assigned to task groups. After each task, two vocabulary tests, Word Recognition Test (WRT) that tests word form knowledge and Vocabulary Knowledge Scale (VKS) that tests word meaning and usage, are conducted for each task group to test their mastery of the target words. All test papers are read and graded by independent evaluators. None of these tasks is presented as a vocabulary-learning task since it tests the effect of incidental lexical acquisition rather than the intentional lexical acquisition.Experiment results and statistical analyses are consistent with hypotheses of this study: both inference and generation have a positive effect on the acquisition of vocabulary, with effects from generation stronger, and inference plus generation exerts a much stronger positive effect than any of the two standing alone.Inference helps connecting words in memory to one another like nodes in the network. When words are hooked together with episode in which they are first encountered, their form and semantic features are better consolidated in memory. However, the effect of inference is positive in VKS test but not very remarkably strong and WRT does not show significant difference. It is reasonable to conclude that even if inference leads to successful guesses and comprehension, as for word retention, it may lead to the integration of some words into the learner's mental lexicon. Having contextual clues available makes it easy for the learners to understand the word but at the same time it does not encourage learners to put more efforts into making a stronger cognitive connection between the word-form and its meaning which could be applied in some other contexts. Without other encounters or immediate practice of usage, the new word's impression might fade or lose once it has served its immediate text comprehension purpose.The experiments in this study also prove that generating a word item successfully increases the chance that that word will be remembered. Generative tasks require learners to process information at semantic and syntactic levels and to integrate new information with acquired knowledge. Learners who spend effort linking the form of the word with its related concepts have a greater chance of learning the word by being pushed to deploy existing grammatical knowledge more accurately.A New Task Involvement Loads Model is established with empirical support from this research. It contains five major components: need, search, evaluation, inference and generation. They are believed to be conducive to lexical acquisition. Then the new model can be applied to analyze various tasks to compare their relative effects.The present research has both theoretical and pedagogical implications for L2 vocabulary learning. Theoretically, it clarifies which variables affect the probability that particular words will be acquired. When the construct of involvement components can be operationalized and investigated empirically, researchers could devise tasks with different involvement loads and compare them with regard to their effects on incidental vocabulary learning. Pedagogically, the new model can be employed to enhance L2 vocabulary learning practice. Teachers should try not to just present the meaning of a word to learners, but let learners work it out for themselves, and especially production activities shall be encouraged for their great effectiveness in L2 vocabulary learning process.Yet due to limitations of time and other conditions, this research only focuses on the one-time encounter with new words, though much involvement load during the first encounter will not in all likelihood induce long-term retention. For future research, it would be better to conduct experiments on how the combination of the effects of task involvement loads and the effects of multiple exposures would affect lexical acquisition. Factors like word retrieval methods after the first encounter, the number of later encounters, and the time lag between different encounters, are worth investigating. Further researches on these factors, together with the task involvement loads, will better contribute to the field of L2 lexical acquisition research.