《Recent developments in second and foreign language listening comprehension research.doc》由会员分享，可在线阅读，更多相关《Recent developments in second and foreign language listening comprehension research.doc（23页珍藏版）》请在三一办公上搜索。
1、State-of-the-art articleRecent developments in second and foreign language listening comprehension researchLarry Vandergrift Institute of Official Languages and Bilingualism, University of Ottawa, Canadalvdgriftuottawa.caListening comprehension lies at the heart of language learning, but it is the l
2、east understood and least researched skill. This paper brings together recent research and developments in the field of second and foreign language (L2) listening. It begins with a brief discussion of the different cognitive and social factors that impact listening, followed by a summary of recent r
3、esearch into the development of perception skills and metacognitive knowledge. An integrated model for teaching learners how to listen is then elaborated. In addition, recent research on listening in multimedia environments, academic listening and listening assessment is presented. The paper conclud
4、es with a discussion of areas for further research, arguing that the process of listening needs more research attention with in-depth studies that probe deeper into the interaction of the processes and factors that influence successful L2 listening.Listening comprehension is an important language sk
5、illto develop.Languagelearnerswantto understand target language (L2) speakers and they want to be able to access the rich variety of aural and visual L2 texts available today via network- based multimedia, such as on-line audio and video, YouTube, podcasts and blogs. Furthermore, listening comprehen
6、sion is at the heart of L2 learning; and the development of L2 listening skills has demonstrated a beneficial impact on the development of other skills (e.g. Dunkel 1991; Rost 2002). Clearly, it is important to develop L2 listening competence; yet, in spite of its importance, L2 learners are rarely
7、taught how to listen effectively (e.g. Mendelsohn 2001,2006; Berne 2004; LeLoup & Pontiero 2007).While we have learned more about the cognitive nature of listening and the role of listening in communication, L2 listening remains the least researched of all four language skills. This may be due to it
8、s implicit nature, the ephemeral nature of the acoustic input and the difficulty in accessing the processes. In order to teach L2 listening more effectively, teachers need a richer understanding of the listening process. A narrow focus on the right answer to comprehension questions (product) does li
9、ttle to help students understand and control the processes leading to comprehension. Listening is often perceived by language learners as the most difficult language skill to learn (e.g. Hasan 2000; Kim2002; Graham 2003) and, consequently, can become a source of anxiety for L2 learners (Elkhafaifi 2
10、005a).Research into L2 listening is important because a better understanding of the process will inform pedagogy. Students who learn to control their listening processes can enhance their comprehension. This, in turn, affects the development of other skills and overall success in L2 learning.Thispap
11、erbringstogetherresearchand developments in the field of L2 listening over thepast six years. It begins with a critical overview of the methods employed to investigate listening. This is fol- lowed by an overview of recent research dealing with: the cognitive and affective dimensions of listening; a
12、pproaches to teaching L2 listening; the use of tech- nology in listening instruction; academic listening; and, finally, listening assessment. The paper concludes with a discussion of areas for further research.1. Investigating listening1.1 Investigating the product of listeningTraditionally, researc
13、h has focussed on the product of listening. Listening comprehension test scores are used to determine the success of an experiment with two groups of learners. Both groups complete a pre- test; the experimental group is exposed to a different condition (such as exposure to a different type of pre- l
14、istening activity) and then both groups complete an immediate post-test to determine the success of the experimental condition (e.g. greater comprehension than the control group). Sometimes a delayed post- test is used to assess the long-term effects of the experimental condition.LARRY VANDERGRIFT i
15、s Professor at the Institute of Official Languages and Bilingualism of the University of Ottawa, Canada, where he teaches courses in FSL, ESL and second language acquisition. He has published articles on various dimensions of L2 listening compre- hension and teaching in Language Learning, Applied Li
16、nguistics, Annual Review of Applied Linguistics, The Canadian Modern Language Review, The Modern Language Journal, Foreign Language Annals, and The French Review. He is currently a co- editor of The Canadian Modern Language Review.Lang. Teach. 40, 191210. doi:10.1017/S0261444807004338 Printed in the
17、 United KingdomOc 2007 Cambridge University Press191Listening test scores can also be correlated with the scores of other instruments that measure variables hypothesized to influence L2 listening success. For example, students may complete a questionnaire on a variable such as anxiety about listenin
18、g, the use of listening strategies, or their level of background knowledge of the topic of the oral text to which they will listen. Although positive or negative correlations may point to interesting relationships between a variable and listening success, it is not possible to prove causality. Uncov
19、ering the nature of that relationship requires careful interpretation and may be elucidated by more qualitative methods such as interviews or stimulated recalls (of questionnaire data).Listening test scores may also be used to assigna level of L2 listening proficiency to participants in studies wher
20、e L2 proficiency is a variable under investigation. While such tests may provide a more objective criterion for assigning proficiency level (rather than grade/course level or teacher assign- ment), the generalizability of the results is limited because they are often based on less reliable in- house
21、 placement tests instead of a more broad-based, standardized, objective measure (Berne 2004). These tests may also be assessing only particular aspects of listening ability (Buck 1991). The advantage of test scores is that they are not labour-intensive and tests can be easily administered to large g
22、roups.Free recall protocols constitute another, more labour-intensive, product-oriented means for invest- igating L2 listening success. When different groups listen to the same text that has been manipulated in some way for the experimental group (e.g. inserting pauses), differences in comprehension
23、 can be assessed by having students write down everything that they recall from the text (e.g. Leeser 2004). These recall protocols are then analyzed for the number of correct idea units, which will determine the level of listening success. The advantage of recall protocols is their capacity to asse
24、ss what listeners are able to retain without any question prompts.While listening test scores and recall protocolsprovide an objective measure for determining comprehension gains for research purposes, the reliability of these results is limited by the reliability of the measure or, in the case of t
25、he recall protocols, inter-rater reliability. Another limitation is the fact that the measurement of listening comprehension is often confounded by other skills such as reading the questions or writing the answers. Similarly, in the case of tests using aural prompts, memory becomes a confounding var
26、iable. Most importantly, although listening test scores tell us something about the product, i.e. the level of listening success, they tell us nothing about the process; i.e. how listeners arrive at the right answer or why comprehension breaks down.1921.2 Investigating the process of listeningA proc
27、ess-oriented approach to investigating L2 listening can provide useful insights into the cognitive processes reported by the listeners. Retrospection techniques such as questionnaires, stimulated recall and interviews provide opportunities for listeners to recall the listening experience at a later
28、moment in time. Questionnaires administered after a listening activity can provide insights into student awareness of the process of listening and, when used as a repeated measure, track any changes in awareness of the listen- ing process. Greater insights into changes in awareness can be achieved w
29、ith a stimulated recall where the listener is asked to reflect on any changes in response patterns over time on the same questionnaire (e.g. Vandergrift 2007). Interviews and listening diaries can provide insights into the ways in which listeners deal with strategy training (e.g. Chen 2005) or their
30、 perceptions of a new experience, such as the useful- ness of a new advance organizer, a new strategy or a new approach to listening (e.g. Goh & Taib 2006).Introspection techniques attempt to tap the thought processes of listeners as they are listening.This research method requires listeners to thin
31、k aloud during pauses in the text, while listening to a recorded oral text. This is the closest researchers can come to tapping thought processes while information is still available to the listener in short-term memory. Think-aloud data are useful for shedding light on how listeners arrive at diffe
32、rent interpretations of a text (e.g. Goh 2000), how they orchestrate cognitive and metacognitive strategies to construct meaning (e.g. Vandergrift 2003b), or how they use visual and aural information to understand videos (e.g. Gruba2004).Since listening is a covert process, observation is of limited
33、 value for investigating listening processes in one-way listening. However, when investigating bi-directional listening, observation of videotaped interviews can provide some insights into the interactions between the speakers/listeners (Farrell & Mallard 2006). The video recording can be reviewed b
34、y the researcher for evidence of the variable under investigation (e.g. number and type of clarification strategies). Review of the recording with the research participant, using stimulated recall immediately after the interview, provides an opportunity to discuss how the participant either clarifie
35、d meaning or helped the interlocutor to advance the conversation.All of the methodologies outlined above can provide greater insights into the process of listening.However, the reliability and validity of the resulting data are often called into question. In order to overcome this limitation, resear
36、chers investigating a construct as implicit as L2 listening should use multi-method assessment to collect convergent data. Reliability can be enhanced when data from more than one source are triangulated to provide a morecomplete picture of the construct under investigation; e.g. complementing video
37、taped data with a stimu- lated recall on the video recording and a questionnaire (e.g. Cutrone 2005) or test scores with questionnaire data and interviews (e.g. Vandergrift 2007). While these methods can provide greater insight into the process of listening, they are more labour-intensive.2. Cogniti
38、ve dimensions of listening2.1 Overview of listening processesUnderstanding spoken language is essentially an inferential process (Rost 2002); linguistic knowledge and world knowledge interact in parallel fashion as listeners create a mental representation of what they have heard (Hulstijn 2003). Lis
39、teners apply these knowledge sources using top-down and bottom-up processes (Lynch & Mendelsohn 2002; Rost 2002; Flowerdew & Miller 2005). Listeners favour top- down processes when they use context and prior knowledge (topic, genre, culture and other schema knowledge stored in long-term memory) to b
40、uild a conceptual framework for comprehension. Top- down processes are developed through practice in the use of compensatory strategies. Listeners favour bottom-up processes when they construct meaning by accretion, gradually combining increasingly larger units of meaning from the phoneme-level up t
41、o discourse-level features. Bottom-up processes are developed through practice in word segmentation skills.While it is generally agreed that these processesinteract in some form of parallel distributed processing (Bechtel & Abrahamsen 1991), the degree to which listeners use one process more than th
42、e other will depend on purpose for listening, learner characteristics such as level of language proficiency, and the context of the listening event. A listener who needs to verify a specific detail, for example, will engage in more bottom-up processing than a listener who is interested in comprehend
43、ing the gist of a text.The speed and effectiveness of these processes depends on the degree to which L2 listeners can efficiently process what is heard (Segalowitz 2003). Native language (L1) listeners are able to process aural input automatically and efficiently, with little con- scious attention t
44、o individual words. Beginning-level L2 listeners, on the other hand, have limited linguistic knowledge; therefore, little of what they hear can be automatically processed. They need to consciously focus on what they are listening to, and a large proportion of what they hear may be lost, given the sp
45、eed of speech and the inability of working memory to process all the information within the time limitations. How well L2 listeners cope with these limitations will depend on their ability to make use of all the available resources to interpret what they hear. Compensatory mechanisms contextual, vis
46、ual orparalinguistic information, world knowledge, cultural information and common sense are used strategic- ally by L2 listeners to compensate for their inadequate knowledge of the target language. How language learners use these information sources while listening will determine the degree of list
47、ening success.2.2 Skilled L2 listeningRecent investigations of the differences between more-skilled and less-skilled L2 listeners provide greater insights into the ways in which listeners regulate these processes. These insights have been gained through the use of introspective, think-aloud protocol
48、s.The importance of metacognitive strategies toL2 listening success emerges from these studies(OMalley & Chamot 1990; Goh 2002b; Vandergrift2003a; Chamot 2005). In a study of adolescent learners of French, Vandergrift (2003a) found stat- istically significant differences in strategy use: skilled listeners reported using about twice as
