International Journal of Language and Linguistics

2014; 2(5): 310-316

Published online September 10, 2014 (http://www.sciencepublishinggroup.com/j/ijll)
doi: 10.11648/j.ijll.20140205.14
ISSN: 2330-0205 (Print); ISSN: 2330-0221 (Online)

The effect of chunk learning on listening comprehension

Abdol-Majid Mohseni1, Amir Marzban2, Abdollah Keshavarzi3
1
Department of English language, Fars Science and Research Branch, Islamic Azad University, Fars, Iran
2
Department of English language, Ghaemshahr Branch, Islamic Azad University, Ghaemshahr, Iran
3
Department of English language, Firoozabad Branch, Islamic Azad University, Iran

Email address:
abdolmajidmohseni@yahoo.com (Abdol-Majid M.)

To cite this article:

Abdol-Majid Mohseni, Amir Marzban, Abdollah Keshavarzi. The Effect of Chunk Learning on Listening Comprehension. International
Journal of Language and Linguistics. Vol. 2, No. 5, 2014, pp. 310-316. doi: 10.11648/j.ijll.20140205.14

Abstract: The present study is done to examine the effect of chunk learning on students’ listening comprehension. Based
on the nature of the study, the null hypothesis was proposed, chunk learning has no effect on (TOEFL) listening
comprehension. In order to test the null hypothesis, a sample of 60 students was chosen. They were randomly put into two
groups (experimental and control). Members of each group were randomly assigned to one of the following conditions; first
the experimental group was given the same passages containing multiword in the passages of experimental group and the
control group that we did not give them any passages in multiword. The design used for this study was a quasi experimental
one. An English Language Test was administered at the beginning to ensure the homogeneity of the two groups in language
proficiency. Then a pre-test was done to measure the student’s knowledge of multi words. After the treatment which took
20 sessions,30 minutes per session, the same pre-test was administered as a post-test to measure the effect of the treatment.
Analysis of the calculated normality test (K-S), descriptive statistic of pre-test, paired sample t-test and independent sample
t-test for experimental and control group provided us with the judgment to reject the null hypothesis. In other words, the
result revealed that: chunk learning (multi words, verbs idioms and collocations) has a significant effect on listening
comprehension ability.
Keywords: Chunk Learning, Iranian EFL Learners, Listening Comprehension

both theory and pedagogy. Riverse (1982) asserts that

1. Introduction speaking does not of itself constitute communication unless
Chunk learning or multi units constructions consists of a what is said is comprehended by listener. Teaching the
verb plus a particle, preposition or noun. Multi word units comprehension is therefore of primary significance if the
are important and vital elements of colloquial English that communicative aim is to be reached.
no one can speak, understand, figure out, perceive and listen This study intends to investigate chunk learning:
to the conversation without fundamental knowledge of them. multiword units, collocations and its effects on Listening
It is noticed as the idiomatic expression in English and Comprehension to offer some insights into the study and
collocation expression, so idioms and collocations and application of it in English Language teaching (ELT). The
ability to use them suitably in context are depend on concept of Chunk Learning has been widely introduced and
distinguishing native-like command of English. But, why do discussed as the teaching responsibilities in Listening
not many students use multiword in speaking or Comprehension. It has been paid more attention to for the
communication? The reason is that that they have more than sake of improving Listening Comprehension. Therefore, a
one meaning, and it cannot be derived from separate words. comprehensive discussion of the concept is importance for a
It must study totally, not selectively. So if it is spoken or better understanding and application of chunk learning.
listened, it can be perceived easily. Listening One of the most widely accepted proposed explanatory
Comprehension is at the core of Second Language mechanisms is learning based on prediction.
Acquisition. So it is important for oral communication. Pioneering work in the 1940s and 1950s suggested that
Nowadays, attention to listening in second language the concept of ‘chunking’ might be important in many
development is a crucial and an important topic of study in processes of perception, learning and cognition in humans
 International Journal of Language and Linguistics 2014; 2(5): 310-316 311

and animals (Gobet et al., 2001). The result of the study by language acquisition (Feyten, 1991). Listening is a skill in
Giroux and Rey (2009), whose results are simulated in this language proficiency which can directly affect other skills
research, provides evidence that when chunks are learned, and be affected by several other strategies or techniques
the subunits making up these chunks are forgotten unless (Safarali & Hamidi, 2012). Therefore, when listeners know
they are refreshed independently (cited in French et al., the context of a text or an utterance, the process is facilitated
2011). This would imply that chunks are encoded as atomic considerably because listeners can activate prior knowledge
entities rather than as associations between their constituent and makes the appropriate inferences essential to
elements. This process, known as ‘lexicalization’, means comprehending the message (Byrnes, 1984).Research shows
that a chunk, once fully formed, is thereafter treated as an that skilled listeners use more metacognitive strategies than
indivisible word, causing any words within it to be their less-skilled counterparts (O'Malley &Chamot, 1990;
overlooked. Vandergrift, 1997a).Several studies have shown that people
Mendelsohn (1998) notes a gap between the interests of can incidentally learn nonlocal dependencies (Creel et al.,
listening researchers and classroom practitioners in that 2004; Gomez, 2002; Newport &Aslin, 2004), the results of
classroom materials do very little to develop metacognitive which would be rather challenging for chunking
knowledge through raising learners' consciousness of mechanisms.
listening processes. It is imperative to teach students how to
listen. This shifts the emphasis of listening practice from 2.2. Research Questions
product to process and the responsibility of learning from The research questions for the present study were
the teacher to the student, thereby helping students become formulated as follows:
self-regulated learners. Research has demonstrated that 1. Does a chunk word have any impacts on listening
adults spend 40-50 % of communication time listening comprehension?
(Gilman & Moody, 1984). 2. In what ways can chunk words affect listening
comprehension?
2. Review of the Related Literature
2.3. Objectives of the Study
2.1. Research Background
Comprehension is what novice students are almost always
Perruchet and Pacteau (1990) presented convincing looking for. To this end, they use chunk words amongst so
empirical evidence that simple associative learning can drive many techniques to step in professional development. The
the formation of chunks and chunk-fragments. Infant aim is to find any probable impacts of chunk word on
initially focused largely on the question of how infants listening comprehension. To pave the way for all language
segment a continuous speech stream into lexical units. The students in general, and for novice ones in particular, it is
raw auditory signal generated by human speech is hoped that this research can open a new route for English
notoriously hard to segment into words because breaks in students. In addition, the output of this study can be a good
the continuity of the signal are poorly correlated with actual ground for comparison with other researches in the same
word boundaries (Cole &Jakimik, 1980). field.
TRACX successfully models adults’ better learning of One of the major goals of EFL instruction is to prepare
word solver part words in the context of (a) differential students to understand the natural spontaneous speech of
within-word, versus between-word, forward TPs (Perruchet native speakers. It can be established through the knowledge
& Desaulty, 2008)in an artificial language with of lexicon (Celece-Murcia 2001).
frequency-controlled test words and part words; (b)
differential within-word, versus between-word, backward 3. Methodology
TPs (Perruchet &Desaulty, 2008) in an artificial language
with frequency-controlled test words and part words; (c) 3.1. Research Design
gradual forgetting of sub chunks found inside chunks
(Giroux &Rey, 2009), if these sub chunks are not The study was carried out during the second semester of
independently refreshed; (d) sentence length and the fact 2013 (summer term) in KISH language institute. The
that words become harder to extract as the length of the purpose of the study was to find out the effect of teaching
sentences in which they are found increases (Frank et al., multiword verbs on subjects’ listening comprehension. This
2010); and (e) vocabulary size and the fact that words study was an attempt to provide answer to the following
become harder to extract as the number and length of the research question “Does learning multiword verbs have any
words to be extracted increases (Frank et al., 2010). significant effect on advanced students’ listening
Shortly after Miller’s 1956 paper, Feigenbaum and Simon comprehension”? The design required random selection
(1962) date began to develop a pure and direct and random assignment of the subjects into two groups:
implementation of chunking mechanisms, known as EPAM experimental and control. A pretest was administered to
(Elementary Perceiver and Memorizer).Listening has capture the initial differences between the groups in terms
emerged as an important component in the process of second of knowledge of multiword. Due to the research questions
the researcher selected a pre test – post test equivalent
312 Abdol-Majid Mohseni et al.: The Effect of Chunk Learning on Listening Comprehension

group design. the subjects falling between +0.5 and -0.5 standard
In this design the pretest was administered before the deviations with a mean of 64.4 were selected for the
application of the treatments and the post test at the end of experimentation. Moreover, normality test, descriptive
the treatment period. Gain scores may be compared and statistics, paired sample test and independent sample t-test
subjected to a test of the significance of the difference were applied in order to estimate the effect of the treatment
between means. Pretest score can also be used in the on students’ language proficiency and listening
analysis of statistical control for any differences between comprehension.
the groups at the beginning of the study. The results of the The result of the data analysis are prepared and
pre test and post test in this study were correlated. In case thoroughly discussed in the next section.
of a statistically significant correlation, the null hypothesis
would be refused or else confirmed. 4. Results
3.2. Participants Regarding the null hypothesis for this research which
The participants in this study were 60 advanced students states that learning multiword verb has no effect on
from among 90 who took the test. They were randomly advanced students’ listening comprehension, an attempt
selected from KISH Language Institute. In order to select a was made to test this significance.
homogeneous sample, the researcher prepared 50 multiple To do so, the result of the students’ performance on the
Questions and 50 multiword from advanced level of two sets of tests (pre test and post test) had to be compared.
passages 1, 2 and TOEFL Longman and 50 Questions from If the comparison showed that their performance differed
listening comprehension from passages 1,2 and TOEFL significantly, then the researcher would be able to claim
Longman. To select the final subjects for the study, 60 that learning multiword verb (phrasal verbs and
students whose score fell on 0.5 standard deviations above collocations) is effective in advanced students’ listening
and 0.5 standard deviations below the mean of 67.4 were comprehension. Therefore, the results of the performance
selected. They were then randomly put into two groups of of the subjects on the sets of tests were compared by using
30, one experimental and the other control which were test of normality, descriptive statistics of pre-test, paired
equal in number. sample test and independent t-test.

3.3. Data Collection Instrument 4.1. Test of Normality

In order to select the teaching materials, some books We use Kolmogorov-Smirnov to investigate about
1. English Idioms In Use. Advanced, (Felicity, 2010). distribution of quantitative data. In this research, it is used in
2. Essential Idioms In English. Phrasal verbs and order to investigate about normality test. If it is, so we can
collocations (Robert J. Dixon, 2004). use parametric statistics. The results of these tests were
3. Idioms and Phrasal verbs. (Ruth Gairn and Stuart shown in table 1 and 2.
Redman 2011) . Table 1. Kolmogorov-Smirnov Test of experimental
Were used. They consisted of some passages ten of
which were selected for teaching since these passages Pre experimental Post experimental
Listening Listening
contained multiword. They were used for the experimental
N 30 30
group. Normal Mean 39.1667 44.3667
The experimental groups’ material included passages Parameters Std. Deviation 2.73021 2.84645
containing multiword centered around common theme. Kolmogorov-Smirnov Z 1.145 .823
There were ten passages in the material, which took 20 Sig. (2-tailed) .145 .507
sessions, 45 minutes each session per week.
Regarding the obtained the result of normality test,
In this experiment, the researcher tried to show the
significant level is higher than 0.05. Thus we can say the
effectiveness of multiword on listening comprehension
normality of test in experimental group is accepted and we
ability.
use parametric statistic.
At the end of this period, a post test was given to
subjects. All of the subjects who had taken the pretest were Table 2. Kolmogorov-Smirnov Test of control group
tested regarding their performance on multiword, and after
Pre-test listening Post-test listening
this stage, the students were required to listen to a TOEFL
control group control group
Longman listening comprehension in phrasal verbs, idioms N 30 30
and collocations, which included some dialogues and texts. Normal Mean 38.6667 38.4333
What was going to be tested was their listening ability and Parameters Std. Deviation 2.84464 3.53976
the impact of the instruction on their listening skills. Kolmogorov-Smirnov Z 1.168 1.011
Sig. (2-tailed) .130 .258
3.4. Data Analysis
Regarding the obtained the result of normality test,
After the result of multiple choice item were prepared, significant level is higher than 0.05. Thus we can say the
 International Journal of Language and Linguistics 2014; 2(5): 310-316 313

normality of test in control group is accepted and we use Having established that there is a significant difference,
parametric statistic. the next step is to find out which set of scores is higher (Time
1 or Time 2). To do this, look in the first printout box,
4.2. Descriptive Statistics of Pre-Test and Post Test labeled descriptive statistics. This box gives us the Mean
Experimental Group score at Time 1 was 39.1667 and the Mean score at Time 2
The result of this test is shown in table 3. was 44.3667.

Table 3. Descriptive statistics

4.3. Paired Sample T-Test, Pre-Test and Post Test
Experimental Group
Mean N SD SEM
Pair 1 Pre.ex.li 39.16 30 2.73 .4984 The result of this test is shown in table 4.
Post.ex.li 44.36 30 2.84 .5196

Note: Pre.ex. li= Pre experimental Listening , Pos.ex.li= Post experimental

Listening.

Table 4. Paired samples test

Paired Differences

95% Confidence Interval of the difference t df Sig. (2tailed)

Mean Std. deviation Std. Error Mean
lower upper
Pair1
Pre.ex. li -5.2 1.49482 0.27292 -5.75817 -4.64183 -19.054 29 .000
Pos.ex.li

In the table labeled Paired samples test, we need to look in The result of this test is shown in table 5.
the last column, labeled Sig. (2-tailed) – this is our
probability value. If this value is less than .05 Table 5. Descriptive statistics
(e.g. .04, .01, .001), we can conclude that there is a Std. Std. Error
Mean N
significant difference between our two scores. In the Deviation Mean
example given above, the probability value is .000. This has Pr. Lis. Con. 36.6667 30 2.84464 .51936
Pair 1
Pos. Lis.Con. 38.4333 30 3.53976 .64627
actually been rounded down to three decimal places- it
means that the actual probability value was less than .0005. Having established that there isn′t a significant
This value is substantially smaller than our specified alpha difference ,the next step is to find out which set of scores is
value of .05. Therefore, we can conclude that there is a higher (Time 1 or time 2 ). Table 5 gives the mean scores for
significant difference in the Statistics Test scores at Time 1 each of the two sets of scores. In our case, the mean score at
and at Time 2. Take note of the t value and the degrees of Time 1 was 36.6667 and the mean score at Time was
freedom (df =29), as you will need these when you report 38.4333.
your results. You should also note that the Mean increase -
5.2 was with a 95% Confidence Interval stretching from a 4.5. Paired Sample T-Test, Pre-Test and Post Test Control
Lower bound of -5.75817 to an Upper bound of -4.64183. Group
4.4. Descriptive Statistics Pre-Test and Post Test Control There is a step involved in interpreting the results of this
Group analysis. The result of this test is shown in table 6.

Table 6. Paired Samples Test

Paired Differences
95% Confidence Interval of the difference t df Sig. (2tailed)
Mean Std. deviation Std. Error Mean
lower upper
Pair1
Pr. Lis. Con. .23333 2.56882 .46900 -.72588 1.19255 .498 29 .623
Pos. Lis.Con.

In the table labeled paired samples test, we need to look in isn′t a significant difference in the of statistics test scores at
the final column, labeled sig.(2-tailed) –this is probability time 1 and at Time 2. Take note of the t value 0.498 and the
value. If this value is less than. 05(e.g. .04, .01, .001), we can degrees of freedom (df=29),as you will need these when you
conclude that there is a significant difference between two report your results. You should also note that the mean
scores. In the example given above, the probability value decrease was 0.23333, with a 95 percent confidence interval
is .623. This value is substantially higher than our specified stretching from a lower bound of -0.72588 to an Upper
alpha value of .05. Therefore, we can conclude that there bound of 1.19255.
314 Abdol-Majid Mohseni et al.: The Effect of Chunk Learning on Listening Comprehension

4.6. Descriptive Statistics Pre-Test In this regard an independent t-test is run to compare the
mean of experimental and control groups on the pre-test.
The result of this test is shown in table 7. The test observed value, 0.695, at 58 degrees of freedom is
Table 7. Group Statistics
lower than of critical value of t, i.e. 2 (table 8).
It can be concluded that there is not any significant
Std. Std. Error deference between the two group mean on the pre-test. The
group N Mean
Deviation Mean
two groups are homogenous in terms of their language
1.00 30 39.1667 2.73021 .49847
PR.EX.L proficiency prior to the administration of the treatment.
2.00 30 38.6667 2.84464 .51936
The F-test run to investigate the homogeneity of the
Table 7 gives us the Mean score at Time 1 was 39.1667 variances of the two groups (F= 0.025) has probability of
and the Mean score at Time 2 was 38.6667. 0.875, indicates that the two groups are homogenous in
terms of their variances. There for the equal variance
4.7. Independent Sample T-Test, Pre-Test Experimental estimates statistics are reported (table 8).
and Control Group

Table 8. Independent samples test: pre-test

Levene's Test for Equality of Variances t-test for Equality of Means

Sig. Mean Std. Error
F Sig. t df
(2-tailed) Difference Difference
Equal variances assumed .025 .875 .695 58 .490 .50000 .71986
PR.EX.L
Equal variances not assumed .695 57.902 .490 .50000 .71986

4.8. Descriptive Statistics Post-Test

4.9. Independent Sample t-Test, Post-Test Experimental
The result of this test is shown in table 9. and Control Group

Table 9. Group Statistics In this regard an independent t-test is run to compare the
mean of experimental and control groups on the post-test.
group N Mean Std. Deviation Std. Error Mean The test observed value, 22.139, at 58 degrees of freedom is
greater than of critical value of t, i.e. 2.02 (table 10).
1.00 30 44.3667 2.84645 .51969 It can be concluded that there is a significant deference
Post.ex between the two group mean on the post-test. The two
2.00 30 28.8667 2.56949 .46912
groups aren’t homogenous. The F-test run to investigate the
homogeneity of the variances of the two groups (F= 0.578)
has probability of 0.005, indicates that the two groups aren’t
Table 9 gives us the Mean score at Time 1 was 44.3667 homogenous in terms of their variances. There for the
and the Mean score at Time 2 was 28.8667. unequal variance estimates statistics are reported (table 10).
Table 10. Independent samples test: post-test

Levene's Test for Equality of Variances t-test for Equality of Means

Sig. Mean Std. Error

F Sig. t df
(2-tailed) Difference Difference

Equal variances assumed .578 .005 22.139 58 .000 15.50000 .70011

Pos.ex
Equal variances not assumed 22.139 57.403 .000 15.50000 .70011

English language test was administered to make sure that the

5. Discussion selected subjects were all at the same level. The subjects
Due to the importance of multiword verbs in spoken and were designed randomly to the Exp Time 1 and Time 2. Also,
written language this research was conducted with the in order to ensure that the subjects are at the same level
purpose of testing the null hypothesis claiming: learning concerning the knowledge of multiword verbs, they were
multiword has no effect on advanced level students’ given a 50 multiple choice test (pretest), and then two
listening comprehension ability. To test the stated hypothesis different pamphlets were provided for 30. The subjects in
a group of 30 students were selected. The participants were the experimental group were exposed to 10 passages
advanced level students at passage 1 & 2 and TOEFL containing phrasal verb and collocations. At the end of this
Longman listening comprehension in KISH English period, learners were given a post test on multiword in order
Language Institute. Before the inception of data collection, to see the achievement of 30 students and to determine
 International Journal of Language and Linguistics 2014; 2(5): 310-316 315

whether this has a significant difference in their use of multiword verbs.

performance. 3. It is suggested to investigate the relationship between
Then, some listening comprehension questions were the knowledge of multiword verbs and learner
administered to clarify the effect of learning multiword productive skills.
verbs on listening comprehension ability of subjects. The 4. The teaching of multiword verbs in interactive
questions were based on the multiword verbs that they had situations is the researchers last suggestion for further
learned during the period of instruction. The obtained results researcher.
of normality test, descriptive statistics, paired sample test
and independent sample t-test showed that it is impossible to
reject the null hypothesis , which holds that learning References
multiword verbs has no significant effect on students’
listening comprehension. [1] Byrnes, H. (1984). The Role of Listening Comprehension: A
Theoretical Base. Foreign Language Annals 17:317-29.

6. Conclusion and Implication [2] Celce-Murcia, M. (2001). 3rd. Teaching English as a second
or foreign language. Boston: Heinle and Heinle and
Vocabulary learning is a continuous task and it is a key publishing Co.
component of language learning. The work of the scholars [3] Cowie, A.P, & Mackin, R. (1983). Oxford Dictionary of
and practitioners shows that there has been a renewed current Idiomatic English. NolI . Ox ford university press.
interest in learning and teaching vocabulary (Maftoon,
[4] Creel, S. C., Newport, E. L., &Aslin, R. N. (2004).Distant
Hamidi & Sarem, 2012). Words and particles can join melodies: Statistical learning of nonadjacent dependencies
together to make phrasal verbs and idioms. Therefore, this in tone sequences. Journal of Experimental Psychology:
study tried to investigate whether the EFL students who Learning Memory, and Cognition,30, 1119–1130.
learn multiword verbs, in addition to common verbs, phrasal
[5] Gobet et al. (2001). Chunking mechanisms in human
verbs, collocation and idioms improved their listening learning. Trends in Cognitive Science, 5 (6), 236-243.
comprehension. Because of the wide spread use of
multiword units in everyday communication. This study is [6] Feigenbaum, E.A. (1963) The simulation of ver bal learning
highly important to show the effect of multiword on advance behavior. In Computers and Thought (Feigenbaum, E.A. and
Feldman, J., eds),pp. 297–309, McGraw-Hill
level listening comprehension. So, the researcher hopes the
findings of this study can help students improve the level of [7] Feigenbaum, E.A. and Simon, H.A. (1984)EPAM-like
teaching English and promote knowledge of listening models of recognition and learning. Cognit. Sci. 8, 305–336
comprehension ability. [8] Feyten, C. M. (1991). The Power of Listening Ability: An
The use of multi words verbs (idioms, collocations and Overlooked Dimension in Lan guage Acquisition. The
phrasal verbs) appears to be avoided in speech addressed to Modern Language Journal, 75,173-80.
learner. On the other hand, idiomatic expressions aboard in
[9] Frank, M. C., Goldwater, S., Griffiths, T. L., &Tenenbaum, J.
movies, and on television. However, input without B. (2010).Modeling human performance in statistical word
intersection and exercise is not sufficient for opportunities to segmenta tion.Cognition, 117, 107–125.
clarify meaning and relieve feedback on use, which are doi:10.1016/j.cognition.2010.07.005
essential for learning.
[10] French, RM., Addyman, C., & Mareschal, D. (2001).
Thus, learners exposure to multiword verbs and idiomatic TRACX: A Recognition-Based Connectionist Framework for
expression appears to be primarily in non-interactive Sequence Segmentation and Chunk Extraction.
situations, where there is no opportunity for negotiation of Psychological Review, 4, 614-636.
meaning. Most learners have little exposure to multiword
[11] Giroux, I., & Rey, A. (2009).Lexical and sublexi cal units in
verbs or idiom in interactive situation which are necessary speech perception. Cognitive Science: A Multidisciplinary
for their learning. Multiword verbs have received inadequate Journal, 33, 260– 272.
attention in our text books and most of the students would
like to resort to the avoidance strategy when confronted with [12] Gomez, R. L. (2002). Variability and detection of invariant
structure. Psychological Science, 13, 431–436.
them. However, multiword verbs are an integral part of the
English language. So, the results obtained from this study: [13] Maftoon, P., Hamidi, H., & Sarem, S. N. (2012). The effects
multiword verbs are highly effective in learners listening of CALL on vocabulary learning: A case of Iranian
comprehension ability. intermediate EFL learners. Broad Research in Artificial
Intelligence and Neuroscience, 3 (4), 19-30.
This study investigates the impact of multiword verbs on
students’ listening comprehension ability, these findings of [14] Mendelsohn, D. (1998). Teaching Listening. Annual Review
this study led to several research areas. of Applied Linguistics 18:81- 101.
1. It will be interested to compare different techniques in [15] Miller, G.A. (1956) The magical number seven, plus or
teaching and mastering multiword verbs. minus two: some limits on our capacity for processing
2. A comparison of native speakers’ speech and EFL information. Psychol. Rev.63, 81–97
learner speech can be done to see the frequency of the
316 Abdol-Majid Mohseni et al.: The Effect of Chunk Learning on Listening Comprehension

[16] Newport, E. L., &Aslin, R. N. (2004). Learning at a distance: [20] Safarali, S. K., & Hamidi, H. (2012). The impact of videos
1. Statistical learning of non-adjacent dependencies. presenting speakers’ gestures and facial clues on Iranian EFL
Cognitive Psychol ogy, 48, 127–162. learners’ listening comprehension. International Journal of
Ap plied Linguistics & English Literature, 1 (6), 106-114.
[17] O'Malley, J. M. & A. U. Chamot (1990).Learning Strategies
in Second Language Acquisition. Cambridge: Cambridge [21] Vandergrift, L. (1997b). The Cinderella of Communication
University Press. Strategies: Receptive Strategies in Interactive Listening.
Modern Language Journal 81:494-505.
[18] Perruchet, P., & Desaulty, S. (2008). A role for backward
transitional probabilities in word segmentation? Memory [22] Vandergrift, L. (2002). 'It was nice to see that our
and Cognition, 36, 1299–1305. predictions were right': Developing Meta cognition in L2
Listening Comprehension. Ca nadian Modern Language
[19] Rivers, w. m. (1982).Teaching For eign-Language Review , 58, 555-75.
Skills(2nd) . Chicago: the univer sity of Chicago.