Introduction
Computers entered school life in the late
1950s in developed countries and have been
developing throughout the world since then
(Gunduz, 2005). Initially, they were mostly
brought to educational settings for the
purpose of processing and displaying
information and their applicability to
teaching was not greatly emphasized.
However, as Brett (1995, p. 77) states,
“increase in the speed, storage capacity and
memory size of computers, together with
developments in the sophistication of
software, now enable computers to deliver
video, sound, text and graphics”, greatly
assisting the process of teaching and making
computers part of most classrooms.
Nowadays a large amount of L2 materials
such as textbooks, dictionaries, compact
discs (CDs), and videos require computers
and technologies. And drawing on
multimedia software programs, computer
assisted language learning (CALL), an
approach to language teaching and learning
in which computers are used as an aid to the
presentation, reinforcement and assessment
of materials (Davies, 2002), is used for
learning/teaching language skills. With
CALL finding its stable floor in educational
settings, listening skill (i.e. the ability to
understand language which is used by native
speakers) is no exception in making use of
multimedia CALL.
Traditionally, second language (L2)
listening comprehension was considered as a
passive and receptive skill, meriting little
attention, and listening activities in L2
classrooms mostly consisted of listening to a
tape and repeating after the teacher or
dictation with a focus on bottom-up
processing, which made L2 classrooms
somewhat boring (Hayati& Vahid, 2012).
However, with listening as an active process
in which listeners attempt to discriminate
between sounds, understand vocabulary and
structures within the context of the
utterance, CALL programs, appropriately
selected and organized, have offered a range
of opportunities to develop L2 listening skill
and vocabulary learning; the attractive
capability of multimedia CALL in
controlling and arranging various media has
introduced audiovisual materials enhanced
with captions as a potential pedagogical tool
in helping L2 learners improve their
listening comprehension skill and
vocabulary learning. However, the use of
captions in listening materials (i.e. textual
versions of the audio dialogues displayed at
the bottom of the screen) with L2 learners at
different proficiency levels has not been
without controversy (Danan, 2004; Pujola,
2002). On the one hand, it is claimed that
captions can promote L2 learning by helping
learners visualize what they hear,
particularly if the input is a little beyond
their linguistic control level (Danan, 2004).
Besides, visual clues and soundtracks in
captioned listening materials can create an
authentic culture and language environment
in which incidental learning can take place
(Yang-dong &Cai-fen, 2007). Furthermore,
captions might be conducive to language
comprehension by facilitating additional
cognitive processes, such as greater depth of
oral-word processing (Bird & Williams,
2002). On the other hand, it is claimed that
captions are more of a distraction in natural
and meaning focused learning than help for
L2 learners, particularly for those at low
levels (Taylor, 2005). It is believed that
"misuse" of captions in listening can
potentially prevent the development of
listening strategies (Pujola, 2002, p. 252).
Creating a gap in L2 research, the above
issues are motivating enough for us to
explore the impacts of captioning and order
of its presentation on L2 listening
comprehension and vocabulary gains across
two different L2 (i.e. English) proficiency
levels through a computer multimedia
software with the hope of helping L2
teachers and material developers in the
development of more effective computer-based listening activities. This objective can
achieve more significance in the English as
a foreign language (EFL) context of Iran
where not much attempt has been made to
develop computer programs in spite of the
potential of recent computer technology in
facilitating L2 learning.
Review of literature
The arrival of personal computers in the late
1970s resulted in an increase in the
development of Computer Assisted
Language Instruction (CALI). With the use
of computers in language education,
gradually CALI changed into CALL, the
expression chosen at the 1983 TESOL
convention in Toronto (Tuncok, 2010).
Since 1980s, CALL has continued its
progress and, for the last decade or so, a
number of studies (e.g., Cushion &
Dominique, 2002; deHaan, 2011;
Jayachandran, 2007) have been conducted to
identify the effect of CALL on L2 listening
comprehension. Although there are some
studies (e.g., Chang, 2002; Dupagna, Stacks,
& Giroux, 2007) which show the negative
effect of CALL on L2 listening
comprehension, most of the studies (e.g.,
Pujola, 2002; Volle, 2005) have revealed the
positive effect of CALL on L2 learners’
listening.
For instance, Verdugo and Belmonte (2007)
examined the effects that digital stories
might have on the understanding of spoken
English by a group of Spanish learners.
Results showed that computer and internet-based technology could improve English
listening comprehension. Also, in the EFL
context of Iran, Khoii and Aghabeig (2009)
and Barani (2011) investigated the effect of
using computer software on the
improvement of listening comprehension of
elementary and intermediate L2 students
respectively. Results of both studies showed
that the use of computer software could
improve the students’ listening ability, as
compared with the traditional way of
listening to tapes and answering some
questions from their book.
Captions are “on-screen text in a given
language combined with a soundtrack in the
same language” (Markham & Peter, 2003, p.
332). The processing of converting the audio
content into text and displaying it on a
screen or monitor may be a bonus in
language learning. Inspired by this claim,
Bird and Williams (2002) examined how a
bimodal presentation (aural and visual) of
novel words would impact the learning of
the words. Vocabulary was presented to
advanced learners of English in three
conditions: (a) text with sound, (b) text
without sound, and (c) sound without text.
Results demonstrated that vocabulary
presented with text and sound (i.e.
captioning) could result in better recognition
memory for spoken words when compared
to the other two presentation modalities.
Also, Pujola (2002) studied the strategies
used by Spanish-speaking ESL learners who
utilized web-based multimedia videos. She
wanted to find out whether the learners
would choose to use captions or transcripts
while watching videos. She found that those
learners with poorer listening skills used
captions more for help with comprehension.
In addition, the Spanish learners generally
had better experiences with captions than
with transcripts. Similarly, Grgurović and
Hegelheimer (2007) reported that students
who used captions in a multimedia video
environment would utilize them more
frequently and for longer periods of time
than those who used transcripts. In another
study, Markham and Peter (2003)
investigated the effects of using Spanish
(L1) captions, English (L2) captions, and no
captions on L2 students’ listening
comprehension; results revealed that the
captions groups outperformed the no
captions group. Along the same lines, Taylor
(2005) examined whether captioned video
could benefit beginning-level learners. Two
groups of Spanish learners (one in their first
year of Spanish and one with three or four
years of Spanish) viewed a video with or
without Spanish captioning. Third- and
fourth-year learners who watched the videos
with captions performed better than first-year students, but scores for those who did
not view captions did not differ regardless of
level. Also, unlike Markham and Peter's
(2003) study, Spanish first-year learners in
Taylor's study found the captions
distracting. They reported it was difficult for
them to attend to sound, image, and
captions. To strike a balance between two
sides, Guillory (1998) have reported that
captions are beneficial for beginning-level
learners when only key words are presented
as captions, rather than having entire
sentences on screen as captions (i.e., the full
text of what was spoken).
Captions can be overused (Pujola, 2002), so
it may be important to see whether listening
materials should be played once with
captions and once without, that is, whether
captioning should be in the first viewing.
Having gone beyond the comparison of
captioned versus non-captioned materials,
Winke, Gass, and Sydorenko (2010)
investigated the effects of order of
captioning during video-based listening
activities in Spanish and less-commonly
taught languages with non-Latin scripts in
the US (i.e. Arabic, Chinese, and Russian).
All the participants watched video materials
twice. The findings indicated that captioning
during the first showing of the videos was
more effective for the performance on
listening comprehension and vocabulary for
Spanish and Russian learners. They have
suggested that "learners of a language whose
orthography is closer to that of the target
language are better able to use the written
modality as an initial source of information"
(p. 80).
The above studies mostly investigated the
role of captioning in L2 listening
comprehension, but it is very difficult to
generalize findings; most of the above
studies did not group subjects by proficiency
levels; the differences might be due to
proficiency levels or the type of materials or
tests used in the study. Moreover, there has
been very little empirical research in EFL
contexts about the role of captioning and
almost none, except one (Winke et al. 2010)
about the order of captioning in L2 listening
comprehension and vocabulary. It is
important to know when EFL learners
should be exposed to captioning in audio
materials to better avoid the misuse of
captions. None of the studies have addressed
the aforementioned issues in the Iranian EFL
context. To fill this gap, the present study is
aimed to investigate the impacts of
captioning and captioning order on L2 (i.e.
English) listening comprehension and
vocabulary gains through a multimedia
computer program in an Iranian EFL
context. To these ends, the following
research questions have been developed:
1. Do captions improve L2 learners’
comprehension of English texts and
learning of English vocabulary?
2. When an English text is listened to
twice, is captioning more effective
when the first listening is with
captions or when the second listening
is with captions?
3. Does English proficiency level
interact with captioning order to affect
L2 learners’ comprehension of English
texts and learning of English
vocabulary?
Method
Participants
For the purposes of this study, 200
intermediate EFL learners were selected
nonrandomly through a placement test
(OPT) from a larger sample of 240 EFL
learners from four private language
institutes (i.e. AvayeDanesh, HomayeZarrin,
Payam Parsa, PejvakeDanesh) in
Zarrinshahr, a city in Isfahan Province,
where they could be accessed by the present
researchers. They included both male (n =
82) and female (n = 118) students whose age
ranged from 18 to 24, with Persian as their
L1. They consisted of 100 high- and 100
low-intermediate learners of English.
Meanwhile, a prerequisite was that all the
participants had passed at least eight terms
in language institutes; therefore, it was
assumed that these students were familiar
with multiple-choice listening and
vocabulary tests and had an adequate
command of listening skill for the purpose
of the study.
Instruments and materials
To collect data, this study made use of
several instruments: The first instrument
was the Objective Placement Test (OPT,
2008) consisting 20 multiple-choice
listening, 20 multiple-choice reading, and 30
multiple-choice language use items. This
study used the OPT to select 200
intermediate EFL learners and place them
into two L2 ability groups (i.e., high and
low). In the current study, the Cronbach
Alpha reliability of this test was found to be
acceptable (0.80). Besides, the correlation
coefficient between the scores obtained from
the OPT and a retired paper-based TOEFL
was found to be high (see procedures). The
second one was a listening comprehension
test, consisting of 24 true/false items and 32
multiple-choice (MC) items. The
participants had to click on the choice true
or false in 15 seconds and the best
alternative in MC in 30 seconds after the
audio prompts were presented to them (see
Figure 1).
Finally, the third one was a vocabulary test,
consisting of 36 multiple-choice items with
the key target vocabulary selected from the
audio texts and no cognates. Each test had
five choices, one of which was “I knew this
word before listening to the text” (Figure 2).
Meanwhile, the validity of the listening
comprehension and vocabulary tests was
established through factor analysis, using
principle component analysis (PCA) on a
group of 100 participants. Moreover, the
reliability of the listening comprehension
and vocabulary tests as measured by
Cronbach’s alpha in the current study was
found to be 0.81 and .0.85 respectively.
The audio texts used in this study included
four English short stories, selected on the
basis of length, conceptual difficulty, and
readability from the Steps to Understanding
(Hill, 1988), presenting audio materials at
the intermediate level. Each short story,
approximately one minute in length, had a
single narrator telling the story.
Procedure
To collect the data, the OPT was given to
240 L2 students. Following guidelines of the
OPT (Hansen & Lesley, 2005), their OPT
scores were used to select 200 intermediate
(i.e., 100 high- and 100 low-intermediate)
EFL learners. Moreover, to ensure the
dependability of the data, 25 of the selected
participants (12 males and 13 females) also
answered a retired version of TOEFL
(2004), and the correlation between their
OPT and TOEFL scores was investigated
using the Pearson product-moment
correlation coefficient, which turned out to
be high (0.85). Meanwhile, the computer
software through which the listening and
vocabulary tests were administered was
piloted on a sample of 20 intermediate level
L2 learners to assess the appropriacy of the
materials, time, wordings and instruction. In
addition, the construct validity of the tests
was examined by PCA in a sample of 100
intermediate EFL students. Using Catell’s
(1966) scree test, 56 listening and 36
vocabulary items with acceptable
eigenvalues were retained for the further
data collection. To assess the potential
impacts of captioning and order of
captioning on L2 listening comprehension
and vocabulary, the selected participants
were then randomly assigned into four
groups, each with 50 EFL learners: the
caption group (CG), noncaption group
(NCG), the first caption group (FCG), and
the second caption group (SCG). For the
main trial,
1. the CG listened to the audio short
stories twice, both times with
captioning (Figure 3);
2. the NCG listened to the audio short
stories twice, both times without
captioning;
3. the FCG listened to the audio short
stories twice, first time with
captioning;
4. the SCG listened to the audio short
stories twice, second time with
captioning.
After the second listening of each audio text,
the corresponding listening comprehension
test items, followed by the corresponding
vocabulary test items, were administered to
the participants of the main study. Finally,
discrete-point scoring procedures (i.e. 0 for
false and 1 for right answers) were utilized
to obtain each participant's total listening
and vocabulary scores through the software.
Results
Table 1 shows the descriptive statistics
(mean and standard deviations) of the L2
(i.e., English) listening comprehension and
vocabulary scores for the four groups across
the two proficiency levels. As demonstrated
in the table, the high-intermediate
participants in the four groups received a
higher mean score than the low-intermediate
ones on both listening and vocabulary items,
with the highest listening and vocabulary
mean scores belonging to the caption group
(M = 51.12, M = 31.48 respectively).
Moreover, the standard deviations in the
four groups did not show great variability in
the listening and vocabulary scores.
To answer the first research question, which
concerned the overall impact of captioning
on the English listening comprehension and
vocabulary scores, independent t-tests were
used, with the captioning (i.e.
captions/noncaptions) as the independent
variable and the listening comprehension
and vocabulary scores as the independent
variables in the analysis. As demonstrated in
Tables 2 and 3, there was a statistically
significant difference between the listening
mean scores of the CG (M = 48.84, SD =
2.77) and the NCG participants (M = 46.74,
SD = 3.25) at the 0.01 level, t (98) = 3.47,
*p < .01. That is, the CG participants
outperformed the NCG ones on the L2
listening comprehension test. Moreover, the
eta squared, showing the magnitude of the
mean difference, was found to be moderate
(0.10). Along the same lines, a statistically
significant difference between the
vocabulary mean scores of the CG (M =
29.22, SD = 2.72) and the NCG participants
(M = 27.52, SD = 2.96) was found, t (98) =
2.98, *p < .01. That is, the CG participants
performed better on the vocabulary test than
the NCG participants did. However, the
magnitude of the difference in the means
was not large (eta squared = .083).
The focus of enquiry in the second research
question was the effect of captioning order
on the L2 listening comprehension and
vocabulary scores. To respond, independent
t-tests were employed with the order of
captioning (first/second captioning) as the
independent variable and listening
comprehension and vocabulary scores as
dependent variables involved in the analysis.
As exhibited in Table 4, there was not a
statistically significant difference between
the listening mean scores of the FCG (M =
45.32, SD = 3.08) and SCG participants (M
= 46.28, SD = 2.97) at 0.01, t (98) = -1.58, p
= .117. That is, the FCG participants'
performance on the listening comprehension
test was not significantly different from that
of the SCG participants. Naturally, the
magnitude of the mean difference was small
(eta squared = 0.02). In line with these
results, as depicted in Table 5, no significant
difference between the mean scores of the
FCG (M = 26.86, SD = 3.28) and SCG
participants (M = 25.80, SD = 3.27) was
reported, t (98) = 1.61, p = .110. And, the
eta squared was found to be so small (0.02).
That is, the FCG participants' performance
on the L2 vocabulary test was not
significantly different from that of the SCG
participants.
The third research question explored
whether L2 proficiency level interacted with
the captioning order to impact the
participants’ L2 listening comprehension
and vocabulary scores. To respond, separate
two-way between-groups ANOVAs were
run. The results are shown in Tables 6 and 7.
The ANOVA revealed that there was a
statistically significant effect for the L2
proficiency level on the listening
comprehension scores, F (1, 192) = 487,
*p< .01. Based on Cohen’s (1988)
guidelines, the effect size for the proficiency
level was large (partial eta squared = .71).
But, the interaction effect between
captioning and proficiency level for the
listening comprehension scores was not
statistically significant, F (3, 192) = .766, p
= .514. Figure 4 displays a clear picture of
the participants' performance in the two
proficiency levels. The pattern of
performance was somehow similar in the
two proficiency level groups, with the CG
receiving a higher listening mean score and
FCG receiving a lower one. Likewise, the
post hoc test showed that the CG performed
significantly better than other groups
including the FCG and SCG. Also, the
performance of the SCG was better than that
of FCG on the listening scores even though
the difference was not significant at .01
(mean difference = .96, p = .018).
Furthermore, as depicted in Table 7, there
was a statistically significant effect for the
L2 proficiency level on the vocabulary
scores, F (1, 192) = 429 *p< .01, with the
effect size of .69. Following Cohen’s (1988)
guidelines, this effect size could be large. In
line with the results on the listening
comprehension scores, the interaction effect
between captioning and proficiency level for
the vocabulary scores was not statistically
significant, F (3, 192) = 1.20, p = .297.
Figure 5 displays how the participants in the
two proficiency levels performed on the
vocabulary test. Again, the pattern of
performance was almost similar in the two
proficiency level groups, suggesting no
interaction between the proficiency and
captioning order; the CG received the
highest vocabulary mean score and SCG
received the lower one in the two
proficiency level groups. Also, the post hoc
test showed that the CG performed
significantly better than other groups
including the FCG and SCG. However,
unlike the listening test, the performance of
the FCG was better than that of SCG on the
vocabulary scores though the difference was
not significant at .01 (mean difference =
1.06, p = .013).
Discussion
Captioned video and audio materials for L2
learning are becoming more common.
However, there is controversy over whether
they bring more native voices into the
learning environment. The present study sets
out to investigate L2 learners’ use of
captions while listening to short stories. In
relation to the first research question, it was
found that captioned audios aided L2
listening comprehension and vocabulary
gains to a greater degree than non-captioned
audios did. In other words, the captioned
group outperformed the non-captioned in
both the listening comprehension and the
vocabulary tests. The benefit of captioning
could be due to a bimodal presentation
provided in the caption group. Perhaps, if
one of the channels (audio or visual) failed,
the other one compensated for the failure.
This is plausible given that listening
comprehension and vocabulary learning are
dependent on the multiple input modalities.
The other possible reason is that listening
twice to the audios with captions might have
reduced the difficulty of input to reach the
optimal level or, in Krashen's (2003) terms, i
+1. Besides, providing the two channels
could help in reducing the level of stress or
anxiety on the part of participants or, in
Krashen's words, lowering the affective
filter so that the participants in the caption
group could take in more comprehensible
input. Vanderplank (1993) suggests that
captions are not affected by variations in
accent or audio quality. If so, the captions
could reduce stress and positively facilitate
their aural comprehension or implicit
vocabulary learning. This justification is
also supported by the results of the study by
Bird and Williams (2002), who found that
vocabulary presented with text and sound
(i.e. captioning) could result in better
recognition memory for spoken words.
Markham and Peter (2003) also found that
captioning could improve Spanish ESL
learners' listening comprehension
effectively. Zarei's (2009) study, in which
the bimodal subtitling was reported to be an
effective mode for EFL learners for
comprehending English movies and picking
up new words, can also partially support the
above result.
The other concern of this study was to
investigate the ordering effect of caption
presentation. The results pointed out that
when a short story was listened to twice,
once with captioning and once without, the
order of viewing had no significant impact
on either L2 listening comprehension or
vocabulary. Winke et al. (2010) argued that
the order of captioning had an impact on the
overall comprehension and vocabulary
recognition. They found that Spanish and
Russian learners presented with captions in
the first viewing were better able to perform
on the listening comprehension and the
vocabulary tests than learners presented with
captions in the second viewing. They
suggested that this was due to the important
role of attention in L2 learning. The results
of the present study likewise showed the
participants had a better performance on the
vocabulary test with the captions in the first
viewing, but unlike the study by Winke et al.
(2010), its effect was not found to be so
significant since the mean score of the first
captioning group was marginally better than
the second captioning group. Besides,
second captioning group performed better
on the listening comprehension, though not
significantly. It can be argued that the
captions in the first viewing seemed to help
isolate key vocabulary that the L2 learners
were not encountered for the first time or
perceived to be important, so they might
have paid more attention to new vocabulary
in the subsequent listening or confirmed
their hypotheses on the meaning of unknown
words, hence having a better performance
on the vocabulary test. That is, the first
caption viewing might help further
information-gathering on the vocabulary
during the second listening. If so, the second
listening in the present study provided
additional confirmatory/non-confirmatory
evidence of form-meaning as regards
vocabulary. At the same time, the first
caption viewing could not be very
facilitative for L2 listening comprehension
when the second listening was presented
without captioning perhaps because most
participants in the first caption group might
have lost track of plots or the main idea in
the audio stories. And, zero captioning in the
second listening could not greatly help them
compensate for their failure. Rather, non-captioning in this context might have put
more stress on them, hence not displaying a
good performance on the listening
comprehension test in comparison with the
vocabulary one. But when the first listening
was presented without captioning, the
participants' attention might have been better
drawn to the incidents and theme of the
audio stories and the second listening with
captions could have provided additional
confirmatory/non-confirmatory evidence of
their comprehension or, at least, reduced the
anxiety associated with listening. All said,
the effect of ordering needs further studies
before making a strong statement about its
effect given that the effect of this variable
was not found to so effective in L2 listening
comprehension and vocabulary performance
at a less conservative level of significance,
and there are not enough studies in the
literature to compare and generalize the
above findings broadly.
Finally, the participants' English proficiency
level difference did not provide any major
benefits taken from captioning order. In
clear terms, listening twice to a short story
with captions was most effective for both
high- and low-intermediate L2 participants;
that is, captioning helped both. Similarly,
listening to a short story with captions the
first rather than the second time was equally
beneficial for the vocabulary performance,
and listening to a short story with captions
secondarily was equally useful for the
listening comprehension regardless of the
L2 proficiency level. Thus, it is assumed
that captioning can be a pedagogical tool,
which aids language processing, and
function similarly for upper and lower L2
proficiency levels. Ellis (2003) states that
“learning to understand a language involves
parsing the speech stream into chunks which
reliably mark meaning” (p. 77). It can be
argued that the captions presented twice
might have helped the L2 learners see and
be able to parse patterns or chunks in the
audio listening materials. This might have
aided both high- and low-intermediate
participants in remembering and learning
from the chunks when they were repeated
(i.e. presented in written form twice).
Meanwhile, the better mean scores obtained
by high-intermediate participants, in general,
as compared with low-intermediate ones,
could be due to better L2 ability, which was
observed in all four groups of the study.
That is to say, it was regardless of caption
ordering. The investigation of patterns of
performance by the two proficiency level
groups in Figures 4 and 5 suggest that
captioning in repeated listening can be
beneficial for a range of proficiency levels,
perhaps, so long as the listening materials
are suitable in terms of content and
difficulty to L2 learners' proficiency levels.
However, in Taylor’s (2005) study, the
lower-level learners reportedly had difficulty
with attending to captions than upper-level
students perhaps because they had a harder
time with the content of the video materials;
the content might have been too difficult for
them. In line with Winke et al.'s (2010)
claim, the above results of the present study
suggest that the question over whether
lower-level students can benefit from
captions in the same way as upper-level
learners should focus more on the
appropriateness of the complexity level of
L2 listening materials rather than the
appropriateness of the captioning for L2
lower-level learners.
Conclusion
According to Hashemi and Aziznezhad
(2011), “CALL offers modern English
language teachers many facilities and novel
techniques for teaching and learning” (p.
833). Thus, the effect of CALL on listening
comprehension and vocabulary learning has
shown great consideration among language
teachers and researchers. Despite the
significance of CALL in listening
comprehension or vocabulary learning and
captioning, supported by a number of
empirical studies conducted in L1 (Bird &
Williams, 2002), there is a paucity of
research on CALL-facilitated captioning
techniques in L2 listening comprehension
and vocabulary learning, particularly in EFL
contexts. The present study then took a
further step to help fill this gap by
investigating, firstly, the impact of
captioning; secondly, the effect of
captioning order; and, finally, the effect of
possible interaction between L2 (i.e.
English) proficiency and captioning on L2
listening comprehension and vocabulary
gains.
The results indicated that captions had a
beneficial effect on both L2 listening
comprehension and vocabulary gains. They
can result in greater depth of language
processing by presenting multiple input
modalities and reducing anxiety, and assist
the implicit learning of vocabulary through
the unpacking of language chunks or
mapping form-meaning. Also, the results
revealed that the captioning order played no
significant role in the L2 listening
comprehension and vocabulary
performance. In other words, listening twice
to a short story, first with captions and then
without, did not significantly affect the L2
learners’ performance on the listening
comprehension and vocabulary tests.
However, this issue is due further
investigation since small contributions
sometimes cannot totally be ignored in
educational settings. Finally, this study did
not find that L2 proficiency level differences
would affect performance derived from
captions ordering. Constrained by the time,
this study did not explore whether additional
listening with captions or captioning order
would result in greater vocabulary and
comprehension gains. Possibly there is a
ceiling effect for captioning. Besides, the L2
participants in this study were not allowed to
toggle captions on and off in the program.
Perhaps allowing L2 learners to toggle
captions on and off can provide more
information when captions might be useful
or useless to them. Thus, future research can
transcend limitations observed in this study
in addressing captioning in a multimedia
environment.
