Learn English or die: The effects of digital games on interaction and willingness to communicate in a foreign language

Abstract

In recent years there has been a lot of interest in the potential role of digital games in language education. Playing digital games is said to be motivating to students and to benefit the development of social skills, such as collaboration, and metacognitive skills such as planning and organisation. An important potential benefit is also that digital games encourage the use of the target language in a non-threatening environment. Willingness to communicate has been shown to affect second language acquisition in a number of ways and it is therefore important to investigate if there is a connection between playing games and learners’ interaction in the target language. In this article we report on the results of a pilot study that investigated the effects of playing an online multiplayer game on the quantity and quality of second language interaction in the game and on participants’ willingness to communicate in the target language. We will show that digital games can indeed affect second language interaction patterns and contribute to second language acquisition, but that this depends, like in all other teaching and learning environments, on careful pedagogic planning of the activity.

Full text

DIGITAL'CULTURE'&'EDUCATION,'3(1)'
Copyright'©'2010,'ISSN'1836‐8301'
Digital Culture & Education (DCE)
Publication details, including instructions for authors
http://www.digitalcultureandeducation.com/
Learn English or die: The effects of digital
games on interaction and willingness to
communicate in a foreign language
Hayo Reinders
Middlesex University
Sorada Wattana
Dhurakij Pundit University
Online Publication Date: 15 April, 2011
To cite this Article: Reinders, H., & Wattana, S. (2011). Learn English or Die: The effects of digital games on interaction
and willingness to communicate in a foreign language. Digital Culture & Education, 3:1, 3-29.
URL: http://www.digitalcultureandeducation.com/cms/wp-content/uploads/2011/04/dce1049_reinders_2011.pdf
PLEASE SCROLL DOWN FOR ARTICLE

Reinders & Wattana'
4
Learn English or die:
The effects of digital games on interaction and willingness
to communicate in a foreign language
Hayo Reinders and Sorada Wattana
Abstract
In recent years there has been a lot of interest in the potential role of digital games in language education.
Playing digital games is said to be motivating to students and to benefit the development of social skills,
such as collaboration, and metacognitive skills such as planning and organisation. An important
potential benefit is also that digital games encourage the use of the target language in a non-threatening
environment. Willingness to communicate has been shown to affect second language acquisition in a
number of ways and it is therefore important to investigate if there is a connection between playing games
and learners’ interaction in the target language. In this article we report on the results of a pilot study
that investigated the effects of playing an online multiplayer game on the quantity and quality of second
language interaction in the game and on participants’ willingness to communicate in the target language.
We will show that digital games can indeed affect second language interaction patterns and contribute to
second language acquisition, but that this depends, like in all other teaching and learning environments,
on careful pedagogic planning of the activity.
Keywords: dig ital g am es , in terac ti on , la nguage te ac hi ng se co nd
la nguage ac qu is it ion, wi ll in gn ess t o co mm un ic ate.
Recent years have seen a growing interest in the pedagogical benefits of digital games
for language learning. Gee (2003), for example, identified 36 learning principles that he
found to be present in many of the games he investigated. An example of these is the
‘Active, Critical Learning Principle’. This stipulates that “All aspects of the learning
environment (including the ways in which the semiotic domain is designed and
presented) are set up to encourage active and critical, not passive, learning.” In other
words, computer games engage learners and involve in the tasks at hand. A second
principle is the ‘Regime of Competence Principle’ where “the learner gets ample
opportunity to operate within, but at the outer edge of, his or her resources, so that at
those points things are felt as challenging but not ‘undoable.’” (Gee, 2003, p. 36). Most
games adapt to the player’s level until they succeed, at which point new challenges
appear. These principles are intuitively appealing and grounded in educational research,
but it is not clear how and to what extent they are related to second language
acquisition. There is not much research on the effects of game play on learning a second
language and the purpose of this article is to review this small body of research before
describing the results of a study investigating the relationship between participation in
an online multiplayer gaming environment and second language interaction patters and
participants’ attitudes towards interacting in the target language (English). We limit
ourselves in this study to an investigation of the acquisition of aspects of the target
language. We acknowledge the importance of sociocultural and ecological views of
language acquisition (Larsen-Freeman & Cameron 2008; Van Lier 2004) but these were
not the focus of this study. For a “cognitive ethnography” of gaming and its effect on
literacy practices, we refer the reader to Steinkuehler (2006; 2008).

'Learn'English'or'die'
5
The effects of game play on second language acquisition
Many claims are made for the benefits of games on affective factors such as anxiety and
motivation, but few studies have directly investigated the effects of digital games on
second language acquisition. An example of such a study was conducted by deHaan,
Reed and Kuwada (2010), who investigated the effects of playing a digital game versus
watching it on immediate and delayed recall of vocabulary by Japanese learners.
Participants in the study were given a music game in which the players had to complete
parts of a song by pressing controller buttons at the correct time. Participants in this
study did not collaborate but were interacting only with the computer (Chapelle’s
human-computer interaction; 2001). An important feature of the study, and perhaps a
major limitation, is that participants did not have to understand the English in order to
play the game. The authors found that playing the game resulted in less vocabulary
acquisition than watching it (although both resulted in learning gains), probably as a
result of the greater cognitive load of having to interact with the game. A post-
experimental questionnaire revealed that there was no difference between players and
watchers in terms of their mental effort, so the effects were due only to their interaction
with the game. The authors argue that playing digital games and interactivity are
therefore not necessarily conducive to language acquisition. However, it is of course
important to understand these findings in light of the fact that the language was not a
focal part of participants’ experience and that they could complete the tasks without
attention to the vocabulary. It is therefore important that future studies investigate
gaming environments that do involve meaningful language use. Another limitation of
this study was the nature of the game that was chosen. This genre of game lacks a
detailed narrative component that requires comprehension in order to respond
appropriately, which is common in many adventure games, for example.
That noticing linguistic elements in an environment where the primary focus is not
on language is possible in a gaming environment was shown by Piirainen-Marsh and
Tainio (2009), who used Conversation Analysis to examine how two teenage boys
repeated language elements in the game to show their involvement and to make sense of
the game. Video recordings of their game interactions showed frequent repetitions both
in the form of immediate imitation but also for anticipatory use and to recontextualise
previously heard utterances, or to expand on them. The authors conclude: ‘On the
whole, repetition offers a flexible resource through which the participants display
continued attention to relevant features of the game and co-construct the collaborative
play activity’ (p. 166). This study did not investigate the effects of this repetition on
linguistic acquisition, however.
Similarly, Zheng, Young, Wagner and Brewer (2009) focused on the effects of game
play on the interaction and collaborative construction of cultural and discourse practices
between native and non-native speakers in the educational game Quest Atlantis. The
collaborative nature of the game required a deep exchange between the two dyads of
players and encouraged the development of not only semantic and syntactic, but also
pragmatic knowledge, and both from native to non-native speaker and vice versa. The
authors refer to this type of interaction as negotiation for action.
Chen and Johnson (2004) “modded” a commercial role playing game called
Neverwinter Nights (Bioware, 2002) to investigate whether a digital game simulating a
foreign language learning context could promote a state of ‘flow’ and motivate students
to practise language skills (Spanish in the case of this study) outside of the classroom.
The authors used questionnaires, video transcripts, field notes, and a post-game
interview to investigate this but realised that there were significant differences in the

Reinders & Wattana'
6
amount of experience the participants had with playing games, and that this strongly
affected their ability to play the game successfully. For example, the one participant who
did have previous game-playing experience felt more comfortable in the game, spent
less time accomplishing the tasks, and self-reported a higher level of enjoyment and
flow in the game than the other participants. This study thus highlighted the importance
of sufficient training, both to encourage greater success in playing the game, and to
minimise the possibility of differences between students acting as a confounding factor
in subsequent analyses.
A key element in the studies above, and in discussions of computer games in
education in general, is that learners more actively participate in the activity at hand (see
also Garcia-Carbonell, Rising, Montero, & Watts, 2001). In language learning, this
means that games are suggested to encourage more interaction in the second language.
We therefore now briefly discuss the importance of interaction on second language
acquisition.
The role of interaction in second language acquisition
Interaction is the term used to refer to the interpersonal activity that takes place both
face-to-face and electronically between people or between people and computer, as well
as the intrapersonal activity that occurs within our minds (Chapelle, 2001). Interaction in
the foreign language has been found to contribute to language acquisition. Interaction
helps generate comprehensible input (Krashen, 1985), encourages negotiation of
meaning (Pica, 1994), facilitates noticing (Schmidt, 1990), produces negative feedback
(Schmidt ibid), and encourages output (Swain, 1985). Swain’s Output Hypothesis (1985)
posits that for successful second language acquisition to occur comprehensible input
alone is insufficient but learners must also be given opportunities to try out new
language and produce comprehensible output during interaction, which, in turn enables
them to develop competence in the target language.
For comprehensible output to be produced, however, learners have to be pushed in
their language production. Pica (1994) claimed that negotiation of meaning helps
learners make input comprehensible and helps them modify their own output, and, in
turn, provides opportunities for them to acquire new language. Similar claims for the
benefits of negotiation have been made by Long (1996) in his Interaction Hypothesis.
According to Long, negotiation of meaning during interaction contributes significantly
to second language comprehension and the negative feedback received through
negotiation facilitates second language development, particularly for vocabulary,
morphology, and syntax. Negotiation also provides opportunities for learners to focus
their attention on linguistic form and to notice aspects of the target language. Noticing
has been considered important because when input is noticed, it can become intake, i.e.
input that the learner has comprehended semantically and syntactically, which facilitates
acquisition (Schmidt 1990). In addition, noticing pushes learners into a more syntactic
processing mode that will help them to internalise new forms and improve the accuracy
of their existing grammatical knowledge.
All of the above, however, assumes that learners are not only given opportunities to
produce the target language, but are also willing to make use of this opportunity. The
crucial aspect of “willingness to communicate” is therefore discussed below.
Willingness to communicate
Many claims have been made for the benefits of games on lowering affective barriers
and encouraging learners to interact within a target domain. In the area of second

'Learn'English'or'die'
7
language acquisition this can have potentially important implications as decades of
research have convincingly shown that exposure to second language input affects
second language acquisition (cf. Ellis, 2002). However, even if learners have potential
access to input, this does not mean that they are willing or able to interact with that
input. Willingness to Communicate as an second language acquisition concept emerged
from previous research on “predispositions toward verbal behavior” (Mortensen,
Arntson, & Lustig, 1977), “shyness” (McCroskey & Richmond, 1982 ), and
“unwillingness to communicate” (Burgoon, 1976). Studies were initially mainly done on
first language acquisition. When applied to second language learning, willingness to
communicate was used to explain why communicative competence alone is necessary
but not sufficient for effective communication in the target language; situational
influences affect willingness to initiate or engage in communication (MacIntyre,
Dörnyei, Clément, & Noels, 1998). In this view, willingness to communicate in a second
language is defined as “readiness to enter into the discourse at a particular time with a
specific person or persons, using a L2 [second language]” (MacIntyre et al., 1998, p.
547).
Willingness to communicate has been found to influence the frequency with which
learners engage in second language communication (Clement, Baker, & MacIntyre,
2003; Yashima, 2002), which in turn is related to the development of second language
communication skills. Willingness to communicate is also regarded a crucial factor in
ultimate proficiency levels in second language production (Kang, 2005); the more use of
the second language, the more likely that second language proficiency will develop
(although of course, proficiency does not necessarily extend to grammatical accuracy or
native-like language use, as demonstrated by Swain and others). Consequently,
willingness to communicate has been proposed as a fundamental goal of second
language learning and instruction in line with the emphasis on authentic second
language communication as an essential part of second language learning and to increase
the likelihood of learners actually using the target language, not only in class, but also in
more naturalistic settings (MacIntyre, Baker, Clément, & Conrod, 2001; MacIntyre,
Baker, Clément, & Donovan, 2003; MacIntyre et al., 1998)
Figure 1 Heuristic Model of Variables Influencing Willingness to Communicate (MacIntyre et al., 1998,
p. 547).

Reinders & Wattana'
8
MacIntyre et al. (1998) conceptualized willingness to communicate in the second
language as a layered pyramid model in which a range of different variables influence
second language learners’ eventual second language use (See Figure 1). The authors
propose that willingness to communicate is influenced by both situational influences
(Layers I, II, III) and enduring influences (Layers IV, V, VI). As the learner moves up
the pyramid, the learner has more control over the act of communicating in a second
language.
Past studies into willingness to communicate have demonstrated its positive effect
on second language acquisition; a willingness to communicate is clearly related to the
likelihood of students improving their second language skills, particularly in productive
skills. Major findings from willingness to communicate studies indicated that learners
who demonstrate willingness to communicate interact in the target language actively,
which, in turn, contributes to increased frequency and greater amount of second
language use (Clement et al., 2003; Freiermuth & Jarrell, 2006; Hashimoto, 2002;
MacIntyre & Charos, 1996; Yashima, Zenuk-Nishide, & Shimizu, 2004). Second
language researchers have recognized that language learners who are more active with
second language use have a greater potential to develop language proficiency as a result
of having more opportunities to communicate with others. Language learners who are
more willing to communicate have been found to have more potential to practice in an
second language (MacIntyre, Baker, Clement & Conrad, 2001), improve their
communicative skills (Yashima et al., 2004), acquire language fluency (Derwing, Munro,
& Thomson, 2008), and generally achieve greater language proficiency (MacIntyre et al.,
2001; MacIntyre et al., 1998; Yashima, 2002). Clearly, an important aim of second
language instruction should be to improve willingness to communicate, and for second
language acquisition research to investigate how this can best be done.
Games can affect some aspects of the above variables influencing second language
willingness to communicate. As games are considered by learners as “fun” and engaging,
they generally create low anxiety environments. Intergroup attitudes within gaming
environments are based on expectations of constant interaction and the social situation
is frequently one that is non-hierarchical and inclusive, and one in which the (second
language) participant has a genuine desire to communicate. In addition, games,
particularly MMORPGs (online role-playing games such as Blizzard Entertainment’s
2004 game, World of Warcraft), provide opportunities for authentic interaction and social
support increasing exposure to authentic second language input and opportunities for
second language output. Such social support is found crucial for developing levels of
willingness to communicate especially outside the classroom (MacIntyre et al., 2001).
The Study
Willingness to communicate has been shown to affect second language acquisition and
the use of computer games is thought to facilitate L second language communication. In
this study we are interested in the relationship between participating in MMORPG
games and second language interaction and therefore pose the following questions:
1) What effects does playing a MMORPG have on a) the quantity and b) quality of
second language interaction?
2) What effects does playing a MMORPG have on learners’ willingness to
communicate?

'Learn'English'or'die'
9
Methodology
Participants. The participants were 10 male and 6 female fourth-year undergraduate IT
students, between 21-26 years of age, at a university in Thailand. All of them indicated
that they had played MMORPGs before and played digital games on average 27 hours
per week. In addition, all males and 3 of the females had played Ragnarok Online
(Gravity, 2002), the game used in this study, before. We therefore did not expect
participants to be unduly affected by novelty and training effects. Their English
proficiency levels ranged from beginning to intermediate as indicated by their grades
from a previous language course, as well as their test scores on the university test of
English proficiency. The pre-survey result revealed that the participants had no other
contact with English other than during the class and that 14 of them considered their
English communication skills only as ‘fair’, and all of them had previous experience
playing MMORPGs.
The game and how it was adapted. Using a commercial game has the advantage of
enabling students to use a high-end and attractive product. It is, however, difficult to
find the right type of game with a design and with content that are appropriate for
second language learners and that match the desired learning outcomes. Consequently,
we decided to modify an existing game called Ragnarok Online, the most popular
MMORPG in Thailand. We obtained permission from its Thai distributors to use the
game in the computer-assisted language learning lab of a university, using a private
server, and to modify the game in order to ensure its appropriateness to the second
language learning context, as well as its alignment with our learning activities and
objectives.
Although the game contains a variety of authentic scenarios and tasks (similar to
those that players may need to achieve in real life), the content of the original game was
considered less than ideal as a computer-assisted language learning environment in the
sense that the opportunities for target language exposure and “language learning
potential” (Chapelle, 2001) were limited. This was due to the fact that the game was
created for Thai native speakers as a form of entertainment, not education. The
international version available from Ragnarok Online’s servers was considered, but it was
not possible to obtain permission to use it for our study. Also, the international version
may not be suitable in terms of the language level used, which could easily be too
advanced.
Another important reason for modifying the game was that the original in-game
quests were considered to be too long for the study participants to complete during
class time. The modification in this study, as a result, meant creating new quest events
relevant to the course that the participants were on, for application of language skills at
the level appropriate. The modification also meant inserting language learning content
inside the game activities to use in ways perceived to be meaningful to students.
Generally, the modified version of Ragnarok Online had some differences in the number
of players, gameplay, language, and game server as shown below.

Reinders & Wattana'
10
Table 1: A Comparison between Official and Modified Ragnarok Online.
Original Ra g n aro k O n l ine
Modified Ra g nar o k O n lin e
Description
Ragnarok Online is set in a fantasy and adventure world inspired by a Norse mythology
and a popular Korean comic (Manhwa) series by Lee Myung-jin. The game provides
collaboration and social interaction, allowing players to interact with others, undertake
quests, and combat computer-controlled enemies.
Genre
Massively multiplayer online role-playing
game (MMORPG)
Multiplayer online role-playing game
In the study, access was limited to study
participants only.
Environment
The game environment is divided into a series of maps with their own unique terrains
and native monsters. Game events take place in different game locations. Players are
free to explore the game environment in a non-linear manner, although they are all
required to start in the same part of the online world.
Players can use the keyboard and the mouse to control most of the basic functions.
Players can create several characters per account but they can only control one
character at a time. Players begin as “Novices” and then gain new skills and abilities as
they specialize in the job they choose and earn experience points and level up. In
general, basic gameplay involves:
• Finding the starting non-player character (NPC) and then accepting a quest for
special rewards required to progress to the next one.
• Interacting with NPCs through controlled dialogues.
• Engaging in non-violent combat by attacking monsters wandering around each
game area in order to collect the items required for quest completion and gain
experience and level up.
However, a few differences in game play for regular players and study participants are:
Gameplay
and Tasks
• Collaborating and communicating
with other players in their native
language via an in-game
communication tool.
• Undertaking optional series of
quests from non-player character
(NPC) for job advancements.
• Collaborating and communicating
with other players in English via both
an in-game communication tool and
additional voice communication
software.

'Learn'English'or'die'
11
Quest
Description
Quests in original Ragnarok Online are for
item hunting and NPC searching. During
and after completing the quest, players
will receive special rewards required to
advance to other quests.
The original game quests were modified to
encourage participants to use the language
and practice the vocabulary and language
skills they learned in class in a fun way.
To complete the quests, participants were
asked to talk to NPCs, read texts, listen to
NPCs’ dialogues, and talk to other players
via either text or oral chat. Participants
were asked to collaborate as they planned
game strategies, discussed maps, solved
problems, made decisions, helped each
other, and exchanged information to
progress throughout the quests.
During and after game events participants
were offered certain in-game rewards such
as base and skill levels, top headgears, and
items required to proceed to the next
quests. In addition, participants were given
immediate and continuous in-game
feedback when they chose wrong
responses while communicating with
NPCs through controlled dialogues.
Language
The regional (Thailand) version of
Ragnarok Online is in Thai.
The entire game is available in English
only.
Game server
• The official game is run and
managed by authorized game
servers.
• Players have to pay by a monthly
subscription or by pre-paid card.
• The modified version of Ragnarok
Online is implemented on a private
server administered by the
researchers, allowing only the
participants of this study to play
the game together free of charge.
• This server is carried out for the
purpose of this study only.
Ragnarok Online was chosen because it is one of the most popular games in Thailand.
Students could therefore be expected to be more likely to know about it or be interested
in playing it. We also chose it because this game allowed us to make partial
modifications and extensions to the original game and host the game on our own server,
giving us more control. The most important reason, however, was that Ragnarok Online
requires participants to communicate in order to progress in the game more quickly, and
for this reason it seemed particularly suited as a medium for our study.
In this study, digital games were used as computer-assisted language learning and
integrated as part of the regular language course. Overall, there were two objectives for
computer assisted language activities in the form of digital games. The first objective
was to give students opportunities to review the course material through play, or
“plearn”, for “play and learn”, which is also the Thai word for “enjoy” (Samudavanija,
1999). The term Plearn is one of the most important concepts in Thai education,
stressing that learning should be an enjoyable activity and students should gain
knowledge through their play. As part of playing newly modified quests, students had
opportunities to learn and practice the vocabulary and language skills they studied in
class in a fun way. By lowering the affective barrier, the intention was to encourage
students to relax and learn in a more natural way (Aoki 1999). The other objective of

Reinders & Wattana'
12
using computer games was to encourage more participation. Thai students are
notoriously reticent and generally avoid interaction in English classes
(Kamprasertwong, 2010). By encouraging students to work together in a non-
threatening environment, the aim was to encourage them to become more actively
involved in the learning process.
Three new quests (i.e. the missions that players are assigned to accomplish in order
to get items and progress throughout the game) were specifically designed in agreement
with the learning activities and objectives of the three lessons with which digital games
activities were integrated, as well as students’ English proficiency at a university level.
The development of new quests was in alignment with particular lessons in ways that
the quests purposefully contained scenarios, language, and lexical items related to what
students previously studied in class. Table 2 shows an example of how learning
objectives were mapped to activities in Ragnarok Online. Figure 2 shows an example of a
screenshot of the modified quest for this study. The tasks in the quests constantly gave
students instantaneous feedback and gradually increased in difficulty as the game
progressed to encourage more communication. Each quest was designed to fit into a
90-minute teaching session in a computer-assisted language learning lab. Each quest
itself lasted approximately 40 minutes but the remaining class time was devoted to the
integration of supporting activities such as preparation and debriefing. Only data for
quests 1 and 3 was recorded and is reported here.
Table 2: A Mapping of Leaning Objectives of Unit 2 to Gaming Activities.
Unit
The objectives for this Unit
are for the students to:
Computer game activities are:
2. Computer
Architecture
1. Use names and types of
computers and computer
features
2. Understand computer
advertisements
3. Understand computing
terms and abbreviations
4. Make inquiries and give
answers about computer
specifications
5. Describe the function of an
item
6. Give instructions
Quest Event:
Looking for a computer shop sales assistant
Quest Description
Students were required to pass the test about
computer knowledge (i.e. names and types of
computers 1, computing terms and
abbreviations 3). They were asked to
complete several tasks (i.e. reading computer
ads 2, describing computer specs 4, using
computing terms and abbreviations 3,
describing functions 5) in order to be
employed as a new sales assistant in a
computer shop.
Students needed to talk to the Shop Manager
to start the quest, help a Customer, hunt for
required items, exchange ideas and
information, listen to each other, and help
each other to progress in the quest. This help
could include giving their friends instructions
on how to complete each task in the quest.

'Learn'English'or'die'
13
Figure 2: A screenshot of Quest Event called ‘Looking for a computer shop sales assistant’.
Procedures. Students were requested to participate in three game sessions. Before
starting each session, a 15-minute briefing was given which included linguistic
preparation (i.e. giving students planning time to write down the questions that might be
asked during gameplay and to learn from each other about words, phrases, grammar,
and language functions that could help them during the game), and familiarizing
participants with the quests (i.e. discussing quest information such as which NPC they
needed to interact with to accept the quest, what kinds of tasks needed to be completed,
etc). Students were clearly briefed on what they were expected to do and when and why.
The expectations involved length of class time students had to complete the quest,
ground rules for communication (e.g. what is good ‘netiquette’?) and collaboration (e.g.
what is collaboration and what is cheating?), as advised by Whitton (2010, p. 81), things
they could and could not perform as players (e.g. do’s and don’t’s), and benefits they
could obtain from the activity. Students were also instructed to focus on collaboration
rather than competition, to play the game not only for fun but also for learning, and to
try to use the target language for communication in the game. During game play,
students were randomly divided into either a text-based chat or a voice-based chat
group and were instructed to collaborate and communicate synchronously with other
playing characters (PCs) in order to progress throughout the game. After each game
session ended, students were asked to complete a willingness to communicate
questionnaire and finally a collaborative debriefing took place during which participants
were asked to discuss in small groups about their experience, success and failure in the
game and how they had communicated synchronously.
Data collection and analysis. Two types of data were collected: (a) transcripts of
students’ produced discourse were recorded using Skype Chat Recording or Skype Call
Recording for evidence of their interaction and language use as they worked on
computer game activities and (b) students’ responses to a questionnaire for evidence of
their WTC. The questionnaire was adapted from previous studies on willingness to

Reinders & Wattana'
14
communicate (Cao & Philp, 2006; Freiermuth & Jarrell, 2006; LÈger & Storch, 2009;
MacIntyre et al., 1998; MacIntyre et al., 2001) and was specifically modified to focus on
communication situations that commonly take place during game play. The transcripts
were analyzed using (a) discourse analysis to describe the syntactic and functional
characteristics of the language produced during game play and (b) interaction analysis
for tabulating the number of words and the number and length of turns. Descriptive
statistics were calculated for the mean and frequency of the responses to Likert scale
items on the questionnaire data, revealing to what extent students accept each statement
in mean scores and percentage points, and responses to open-ended questions were
grouped according to recurrent themes.
Results
What effects does game play have on the quantity and quality of second
language interaction?
Below we first look at the amount of interaction that took place during text and voice
chat and compare them. In the following section we look at the type of interaction that
took place.
Quantity of second language interaction. The results showed that game play had
positive effects, compared with anecdotal observations of students’ communicative
behaviour during face-to-face interaction, on the quantity of second language interaction
via text-based chat and oral-based chat during the three computer game sessions, as
measured by the number of words and length of turns. Figure 3 and Table 5 show that
students took 528 and 607 turns in text chat in session 1 and 3 respectively. Individuals
ranged from 35 to 121 turns with both the average and the minimum and maximum
number of turns increasing between the two sessions. The average number of turns per
student in session 3 (M = 75.88, SD = 20.518) was greater than the average number of
turns per student in session 1 (M = 66, SD = 18.174) when they had just started and
were not yet used to the game. A paired t-test was performed to determine if this
difference was statistically significant. As shown in Table 3, this was found to be the
case (t = 3.837, p = .006; p < .05) with a medium effect size (d = 0.49), meaning that
the amount of interaction increased from session 1 to 3.
Table 3: Paired Samples Test for Average Number of Turns via Text-Based Chat per Student in Session
3 and Session 1.
95% Confidence
Interval of the
Difference
t
df
Sig.
(2-tailed)
Mean
Difference
Lower
Upper
Pair 1
Session 3 – Session 1
3.837
7
.006
9.875
3.790
15.960
In voice-based chat, participants took 348 turns during the first and 408 during
the third session (see Table 4), again showing a similar pattern of increasing averages
and a higher minimum and maximum number of turns. As shown in Figure 3, the
average number of turns per student in session 1 was M = 43.50 (SD = 10.170) and in

'Learn'English'or'die'
15
session 3 M = 51 (SD = 9.957). According to Table 4, this difference was found to be
significant (t=8.1, p = .000; p < .05) with a medium (effect size (d = 0.75).
Table 4: Paired Samples Test for Average Number of Turns via Voice-Based Chat per Student in
Session 3 and Session 1.
95% Confidence
Interval of the
Difference
t
df
Sig.
(2-tailed)
Mean
Difference
Lower
Upper
Pair 1
Session 3 – Session 1
8.101
7
.000
7.500
5.311
9.689
Figure 3: Average number of turns per student, communicating via text-based chat and voice-based chat
while working on computer game activities.
Table 5: Number of Words and Number and Length of Turns in Text-Based Chat and Voice-Based
Chat during Game Play.
Session 1
Session 3
Written
(n=8)
Oral
(n=8)
Written
(n=8)
Oral
(n=8)
Number of turns
528
348
607
408
Length of turn
• Single word
• Phrase
• Incomplete T-units
• Complete T-units
49 (9.3%)
27 (5.1%)
58 (11%)
394 (74.6%)
19 (5.6%)
20 (5.7%)
62 (17.8%)
247 (71%)
53 (8.7%)
31 (5.1%)
61(10%)
462 (76.1%)
41 (10%)
21 (5.2%)
65 (15.9%)
281 (68.9%)
Total words
1,881
1,064
2,456
1,249
English-only total words
1,875
1,054
2,455
1,245
Not surprisingly, a slightly higher proportion of incomplete T-units—the
shortest form of a sentence that is still grammatical—(17.82% vs. 10.98% and 15.93 %

Reinders & Wattana'
16
vs. 10.05%) was found in a voice-based chat (See Table 5) while a greater proportion of
the number of words produced (235 vs. 133 and 307 vs. 156 average words per student)
was found in a text-based chat between the two sessions (See Figure 4). However, the
increased number of words produced seemed unstable. Finally, it was found that there
were a higher number of English words used through written interaction than in oral
interaction (1,875 vs. 1,054 and 2,455 vs. 1,245 in sessions 1 and 3 respectively) (See
Table 5).
Figure 4: Average number of words per student in text-based chat and voice-based chat.
Quality of second language interaction. Although a large quantity of the target
language was produced, second language interaction during game play did not seem to
pose a development of accuracy and complexity of learners’ produced discourse,
probably due to the demands for simultaneous communication flow. However, second
language interaction during game play did, indeed, encourage a variety of discourse
functions, which is summarized in Table 6. The oral-based chat transcripts showed
more use of greetings than did the text-based chat transcripts (16 vs. 8 and 24 vs. 6 in
sessions 1 and 3 respectively). Especially in the third session of oral interaction where
students seemed to have stronger interpersonal relationships, greetings were found to be
more formulaic and interaction contained more turns and small talk, while students
engaged in written interaction spent less time greeting each other and initiated their
conversations directly (See Example 1).
Example 1
Example of voice-based chat
Hunna: Hello Momman
Momman: Hello Hunna. How are you today?
Hunna: I’m fine. Thank you. And you?
Momman: I’m fine too. Today is last game session. I so sad.
Hunna: Why?
Momman: Because I can’t talk to you in game again. I wanna cry.
Hunna: Don’t cry now, Momman. We should start quest or we can’t finish.
Momman: OK
Example of text-based chat
[15:09] ManN: hi
[15:11] LKAK:
[15:17] r u ok?
[15:19] ManN: yes
[15:29] have u find NPC “Newton”?
[15:30] LKAK: no

'Learn'English'or'die'
17
Use of clarification requests was present frequently through both written and oral
interaction. Not surprisingly, more clarification requests were made via the oral oral-
based chat than the written-based chat (21 vs. 10 and 23 vs. 4 in sessions 1 and 3
respectively). Probably because of problems with pronunciation, a lack of preparation
time, the use of recording equipment and audio quality issues, clarification requests were
often produced during the voice-based chat. An example of how clarification requests
were used in both voice- and text-based chat is shown in Example 2.
Example 2
Example of voice-based chat
YEEHAAA: What’s the function of mouse?
Innoker: control cursor
YEEHAAA: I don’t understand. Say it again.
Example of text-based chat
[16:51:56] Number1: Why you not wearing the hat?
[16:52:14] Coopy: what do you mean?
As comprehension was required to proceed to other game tasks, a large number of
confirmation checks were present throughout the oral interaction (18 in session 1 and
17 in session 3), while none were present in the written communication. However, self-
corrections were more frequent in the written interaction than in the oral interaction (17
vs. 6 and 18 vs. 11 in sessions 1 and 3 respectively). The fact that participants engaged in
the former medium could read on-screen messages and that they had time to think and
prepare would easily allow them to reflect and correct their messages before and after
posting. Example 3 gives an example of confirmation checks taking place during voice-
and text-based chat.
Example 3
Example of voice-based chat
Example of text-based chat
PzMaxGate: West is lift yes or no?
BadlyAG: Eh?
PzMaxGate: Sorry.
West is left or right?
BadlyAG: Left
[16:34:58] Burn Zero: find another NPC
[16:35:02] where in NPC?
[16:35:05] where is* NPC?
[16:35:39] Zerotz: north of town
In addition, questions were more frequent in the written communication than in the
oral communication. However, most questions asked in both mediums were incomplete
(e.g. where?) and ungrammatical (e.g. “is Burn Zero download something?”). Both text
and voice-based chat transcripts bore evidence of few wh-questions (i.e. what, why,
how, when, or where), many yes/no questions, many uninverted questions (e.g. “u find
Professor?”), and no-tag questions (e.g. questions such as “you come with me?”, as
opposed to “come with me, will you?”). Moreover, in the voice-based chat questions
were usually interrupted by another interlocutor (e.g. “Have you finish the…” “Yes”)
Table 6: Discourse functions of clauses in written and oral interaction during gameplay.
Session 1
Session 3
Written
(n=8)
Oral
(n=8)
Written
(n=8)
Oral
(n=8)
Greeting
8
16
6
24
Clarification requests
10
21
4
23

Reinders & Wattana'
18
Confirmation checks
0
18
0
17
Self-corrections
17
6
18
11
Questions
• WH-questions
• Yes/no questions
• Tag questions
22
27
0
20
24
0
23
29
0
22
21
0
Linguistic features. Learners’ linguistic features are summarized in Table 7. Overall,
students engaged in written interaction were found to pay more attention to
grammatical accuracy than those communicating orally via voice-based chat during
game play. Lexical accuracy was also found to be generally high during written
interaction (1,327 in session 1 and 1,611 in session 3) although some words were
misspelled deliberately in the written interaction as an approach to saving typing time.
In voice-based chat, participants did not have pronunciation problems with simple
words but did with long, difficult, unfamiliar ones. Finally, use of native language words
was more frequent in the oral interaction than in the written interaction, particularly in
the first session (10 vs. 6), but was rare in both written and oral interaction in the last
session. An example of use of native language in voice- and text-based chat is illustrated
in Example 4.
Example 4
Example of voice-based chat
Example of text-based chat
Panzil: do u know ans about how david use
computer in free time?
Funzy: no
my ans yoong yai laew1
Note1: “my ans young yai laew” means “my
answer is now confusing” in English.
[15:57:31] Burn Zero: i have a problem
[15:58:03] LKAK: serious?
[15:58:13] don’t worry
[15:58:21] IKAK: yeah drink!!
[15:58:29]
[15:58:31] LKAK: krean!!!1
Note1: “Krean” is a slang Thai word commonly
used in an online game community to refer to a
misbehaved gamer
Table 7: Linguistic features of the learners’ language production via written and oral communication
during computer game activities.
Session 1
Session 3
Written
(n=8)
Oral
(n=8)
Written
(n=8)
Oral
(n=8)
Grammatical accuracy
Tense (unit of analysis = clause)
• Present Simple
• Present Continuous
• Present Perfect
• Future Simple
301
65
2
26
203
33
0
11
378
48
7
29
233
35
0
13
Lexical accuracy (Spelling)
1,327
-
1,611
-
Pronunciation
-
967
-
1026
Use of native language words
6
10
1
4
Use of simplified or reduced registers. Simplified or reduced registers here included
(1) “leets” commonly used among chat and online game communities where letters are
replaced by numbers and symbols (i.e. emoticons) and words are commonly misspelled,

'Learn'English'or'die'
19
(2) omission of articles, (3) contractions, and (4) abbreviations. Examination of text-
based chat transcripts revealed use of emoticons to exhibit facial expressions and use of
exclamation marks to represent tone of voice. Obviously, the presence of these features
was to compensate for the absence of paralinguistic features found in face-to-face oral
communication. Besides, many omissions of articles and contractions were found in
both text- and voice-based chat, which made it easier and faster for the delivery of the
messages. Abbreviations were also frequently posted in text-based chat but not much
found in voice-based chat so that messages became more comprehensible to the
interlocutor. With the use of simplified or reduced registers, the researchers could
observe learners’ increased capacity to quickly read, comprehend, and produce messages
in English while communicating during game play. Use of simplified or reduced
registers, particularly ”leets”, might be considered inappropriate (See Example 5).
However, it was found necessary for gamers to interact with others quickly so that they
could complete the game quest within the time allotted.
Example 5
Example of text-based chat
[15:20:31]Coopy: hi Masumoto
[15:20:55]MasumoTo: hi Coopy
[15:20:55] have u started test yet dood1?
[15:21:31] it’s very hard
[15:21:38]Coopy: yes!!!!!!!!!!!
[15:21:57] hard 4 me 2 eiei2
[15:22:22]Number1: How many bytes are in megabyte? What’s your ans?
[15:22:57]MasumoTo: i don’t know T_T
[15:23:00]Coopy: 555
[15:23:07] die
[15:23:14] we die!!!!!!!!!!!!!
[15:23:46] T must kill us
[15:23:49]MasumoTo: lol
Note1: dood is a deliberately inaccurate spelling for dude.
Note2: eiei is the textual representation of laughter, which is like “heehee”, “huhu”,
“haha”, “hoho” etc. in Korean laughter expressions
What effects does digital game play have on learners’ willingness to
communicate?
After finishing each digital game session, participants were given a questionnaire to
complete which asked them to rate their willingness to communicate in English on a
scale from 1 (“absolutely not willing”) to 5 (“very willing”) in a range of situations
normally encountered during game play. The obtained Chronbach Alpha coefficient
was .72, which was rather high and indicated acceptable internal consistency among the
five willingness to communicate items. The results revealed that participants were
generally willing to communicate in English (mean 4.52) and generally showed positive
changes in their willingness to engage in communication situations between the two
sessions using second language. Particularly when participants were confused about the
quest, they were increasingly willing to use English to ask for explanations from other
players (mean 4.00 and 4.81 in the first and third session respectively). As shown in
Table 8, the mean score of participants’ willingness to communicate in session 3 (M =
4.84, SD = .13) was higher than the mean score of participants’ willingness to
communicate in session 1 (M = 4.19, SD = .34) and this was found to be statistically
significant (t = 5.921, p = .004; p < .05), showing that participants became more willing

Reinders & Wattana'
20
to interact in the second language over time. The actual difference (Cohen’s d = 1.34)
revealed a very large effect according to Cohen’s d scale of magnitudes of correlation,
which means that computer game playing had very practical importance in willingness
to communicate improvement among second language learners.
Table 8: Willingness to Communicate.
Session 1
Session 3
Total
Situations
Mean
SD
Mean
SD
Mea
n
SD
Give game instructions to other players
4.63
.62
4.88
.34
4.76
.18
Ask for explanations from other players when
you are confused about the quest you must
complete.
4.00
.82
4.81
.4
4.41
.57
Talk to other players about the quest.
4.19
.75
4.88
.34
4.54
.49
Read/Listen to other players’ conversations
attentively.
4.38
.5
5.00
.01
4.69
.44
Read/Listen to Non-player characters’
dialogues actively.
3.75
.86
4.63
.5
4.19
.62
Total Mean
4.19
.34
4.84
.13
4.52
.23
The second section of the questionnaire dealt with participants’ feelings about
communicating in English during game play such as apprehension, excitement,
motivation, and self-confidence, which are considered important factors contributing to
second language willingness to communicate. The obtained Alpha score was .69,
indicating that the four items on this construct were fairly reliable. Again, each item was
rated on a 5-point scale, with the anchors ‘strongly disagree’ (1) and ‘strongly agree’ (5).
Overall, participants reported their positive feelings about using English to interact with
others during game play (mean 4.26), which in turn suggested that they had high levels
of willingness to communicate in English. However, the results showed that participants
gave a low score to ‘confidence’ (mean 3.69). As not all of the participants were
confident in their second language communication, a low level of willingness to
communicate could be present. Table 9 shows that the average score of session 1 was
3.89 and the average score of session 3 was 4.63. This difference between sessions 1 and
3 was statistically significant (t=6.301, p = .008; p < .05). The effect size (Cohen’s d) of
1.15 was large and showed a huge magnitude of the impact of computer games on
second language learners ‘feelings about communicating in English during game play.
Table 9: Participants’ feelings about communicating in English in a computer game context.
Session 1
Session 3
Total
Statements
Mean
SD
Mean
SD
Mean
SD
I feel relaxed communicating in English
during game play.
3.75
1
4.56
.46
4.15
.57
I find communicating in English during
game play challenging.
4.50
.89
5.00
.01
4.75
.35
It is fun communicating in English during
game play.
3.94
.85
4.94
.25
4.44
.7

'Learn'English'or'die'
21
I feel confident when communicating in
English during game play.
3.38
.89
4.00
.97
3.69
.44
Total Mean
3.89
.47
4.63
.46
4.26
.45
The third section examined participants’ reflection on their communication
behavior and second language use in a digital game context. The five items showed a
fairly good level of internal consistency, with a Cronbach’s alpha of .68. In general, what
the participants thought about communicating in the second language was found to be
closely related to their actual communication behavior and discourse produced during
game play. Responses given on a 5-point scale with the anchors “never” (1) and
“always” (5), revealed that participants often used English only to communicate with
other players (mean 4.13) and often made an effort to communicate in English (mean
4.04). Therefore, it was not surprising to find from the transcripts that the participants
gradually reduced the use of their native language and became willing to use only
English as a medium of communication. As shown in Table 7, the mean score of
session 3 (M = 4.05, SD = .67) was also higher than the mean score of session 1 (M =
3.24, SD = .39) and again, this was significant (t = 4.866, p =.008; p < .05). The effect
size (Cohen’s d = 0.75) was medium (Becker, 2002) according to Cohen’s d scale of
magnitudes of correlation.
Table 10: Learners’ reflection on their communication behavior and second language use in a digital
game context.
Session 1
Session 3
Total
Statements
Mean
SD
Mean
SD
Mean
SD
I communicate in English
fluently (with little hesitation
and pauses).
3.13
.62
4.06
.57
3.60
.66
I communicate in English
clearly (i.e. understandable to
the interlocutors).
3.19
.83
4.13
.62
3.66
.67
I request repetition or
clarification when I do not
understand what other players
are saying in English.
2.69
.6
2.94
.68
2.82
.18
I use English only to
communicate with other
players.
3.75
.58
4.50
.52
4.13
.53
I make an effort to
communicate in English.
3.44
1.15
4.63
.62
4.04
.84
Total Mean
3.24
.39
4.05
.67
3.65
.52
The fourth section containing both Likert-scale and open-ended questions was
designed to elicit participants’ experience communicating in a digital game context, as
well as their comments on their progress on second language communication over the
three digital game sessions. Participants were required to complete this section once
they had finished the last digital game session only.
Regarding participants’ experience communicating in English while working on
digital game activities, they gave a favorable rating to their overall experience (M = 3.8,
SD = .4), thus suggesting a high level of willingness to communicate. Most of them
claimed that they liked communicating in a gaming environment because of the fact that
playing and having opportunities for language use went together and enabled them to
communicate without anxiety or embarrassment. In addition, all of the participants

Reinders & Wattana'
22
realized that game play helped them improve their second language communication in a
number of ways. Here are some of the reasons given by the participants.
Playing computer games contributes to my comprehension development. If I make no effort to
understand the quest assignment, NPCs’ dialogues, and contributions from other players, I
cannot complete the game tasks successfully.
I learn more vocabulary from the game and other players. Besides, I have a chance to use a
variety of new words and language functions and therefore increase my language practice.
Playing games is fun, thus developing more confidence and motivation to use English for
communication.
Playing computer games require instant reaction and communication, so it enhances the
development of my language fluency.
Playing computer games provides me the opportunity for second language communication outside
the classroom.
However, what the majority of the participants disliked about communicating in English
during game play was the use of abbreviations, emoticons, smileys, simple words, and
ungrammatical sentences to communicate because they felt that over-use of these would
not contribute to their accuracy and complexity in language production.
There were some changes in the way participants rated their communication skills
(see Figure 5). In other words, while only 6.2 percent of participants considered their
English communication skills as “good” before taking part in the study, 44% claimed
that their English communication skills had improved over the three digital game
sessions because game play had made them feel relaxed, confident, and in turn more
willing to use the target language. Similarly, the observation of the actual
communication behavior of this group showed their improvement in both the quantity
and quality of their language production. More interestingly, students who were
normally shy in face-to-face class tended to become less reluctant, increase participation
levels, and express themselves in a different way through the games. However, the
participants who still considered their communication skills as poor after digital game
participation reported that they did not achieve any improvement because they were
unable to stay focused on second language communication but tended to concentrate
on the game itself. When observing their transcripts, it was found that this group of
students hardly initiated any conversations, often delayed their responses to other
players, and always used only simple words, abbreviations and emoticons.

'Learn'English'or'die'
23
Figure 5: Percentage of learners’ perceived competence in their English communication skills before and
after participating in gaming activities.
Discussion and Conclusions
The results above reveal some interesting findings. Firstly, the differences between voice
and text-based chat in computer games are quite marked, as could be expected. Voice
chat is generally considered difficult and more demanding while text chat might be
preferred by learners as it gives them more time to read others’ and prepare their own
answers (see for example, Sykes, 2005; Thorne, Black & Sykes, 2009). Participants
communicating using voice chat communicate less and less often, produce a greater
proportion of incomplete t-units (the shortest form of a sentence that is still
grammatical), and fewer words and make more clarification requests. Voice chat
resulted in discourse that was, in some ways, more similar to face-to-face
communication, especially in greetings, probably because this modality offers an
environment which is real-time in nature and creates more authentic communicative
situations for interaction. These findings support the results of other chat studies which
also reported more communication (Kern, 1995) and more formal and complex
discourse (Warschauer, 1996) among learners when engaging in text chat than in voice
chat and face-to-face interaction. Similar results were also reported in Jepson (2005)
which found that in the voice chat environment learners were more willing to negotiate
for meaning and used significantly more repair moves than they did in text chat.
Perhaps more interesting is the overall high number of turns for participants.
Although there was considerable variation between speakers, generally participants
communicated quite freely in this online digital game, and increasingly so from session 1
to session 3. However, language production was quite inaccurate and neither complexity
nor accuracy improved from session 1 to session 3. The most likely explanation for this
is that participants were unable to pay attention to both form and meaning. At first
glance, this finding seems to contradict those made by Piirainen-Marsh and Tainio
(2009), however, in their study participants did not converse in the game, only about the
game and were found to repeat input derived from the game. In our study participants
met as characters in the game and therefore had to actively participate in a
communicative exchange. This poses considerable cognitive demand on learners’
language systems (Levelt, 1989) and prevents learners from allocating limited resources

Reinders & Wattana'
24
to both exchanging meaning and paying attention to grammar. This seems to be
confirmed by the fact that participants produced fewer mistakes in text than in voice
chat. Similar results were found by Ortega (2009) for text chat.
When investigating participants’ willingness to communicate, the results showed
that on the whole students were quite prepared to speak in English, and this was
confirmed by their actual participation in the computer quests. This result is more
impressive than it might seem; students in Thailand are notoriously reticent when it
comes to communicating in English (Kamprasertwong, 2010) and teachers frequently
report having great difficulty encouraging language production in class. The digital game
environment clearly posed less of a barrier to them, and – an important finding – their
willingness to communicate improved significantly from session 1 to session 3. In other
words, not only did the game provide an attractive environment to participants, but
communicating in that environment led them to become more willing to communicate
in that environment with the proportion of first language use diminishing over time.
This is particularly interesting when comparing this with participants’ self-evaluation
of their communication skills; seven participants felt their skills had improved. Students
who were normally shy in face-to-face classes tended to become less reluctant, increased
participation levels, and expressed themselves quite freely in the game. However, those
participants who still considered their communication skills to be poor after
participating in the game did not show any improvement. When observing their
transcripts, it was found that they hardly ever initiated a conversation, often delayed
their answers and tended to use only simple words, abbreviations and emoticons. When
asked about this, they said they found it difficult to focus on the language while also
having to learn to use the game.
Limitations and future research
This exploratory study has a number of limitations. Firstly, due to the small sample size
and descriptive nature of this study, statistical analysis is challenging and the results may
not be directly generalizable to other populations. Secondly, some of the data used in
the study was self-reported and this poses a number of well-known challenges
(Crockett, Schulenberg & Petersen 1987). Finally, and perhaps most importantly, we do
not know if participants would have increased their amount of interaction as much if
they had participated in a more traditional, face-to-face class. In other words, we cannot
be certain that the above findings can be attributed to the computer environment
(although participants’ responses to the questionnaires do make this likely). Our aim was
not to compare different learning environments, simply to investigate what happens
when second language learners are asked to play and communicate at the same time.
There was considerable variation between players in terms of their participation
levels and their willingness to communicate, as shown by the large standard deviations.
This variation should not be ignored and further investigation is needed to identify the
impact of individual differences and to determine if there are certain (types of) learners
for whom the use of games is more suitable than others, or ways of ensuring that games
are used in a way that is more inclusive.
Perhaps most importantly, this study took one, admittedly limited, approach to the
students’ interaction; it focused predominantly on their language production and did not
take a more holistic, socio-cultural view of the learning experience of the students in this
particular online environment. Future studies should investigate all aspects of the
learning process to complement the more narrowly interactional perspective taken here.
An example of such a study, also drawing on chat data, is Zheng et al (2009, p. 489),
who applied an iterative and multilayered analysis of the chat data to show how

'Learn'English'or'die'
25
“meaning emerges when language is used to coordinate in-the-moment actions”. This
type of analysis is promising in that it can provide a deeper view on the data.
It is also important future studies further explore the effect of learners’ willingness
and readiness for interacting in games (cf. Chen & Johnson, 2004) and the type and
amount of teacher preparation necessary to be able to make use of games.
Implications
This study has a number of implications, the most obvious of which is perhaps that
commercial games can be adapted for use in second language learning and teaching,
thus removing one important barrier for teachers who may not have the time and
resources to develop their own games. The results have also shown that students
increase their interaction when playing games, which of course is an encouraging
finding. The differences between voice and chat-based text show that students write
more than they speak, and make fewer mistakes in writing, but that in both oral and
written interaction their participation increases; teachers can thus choose one or the
other modality depending on their educational objectives.
Interestingly, however, when students were asked about changes in their confidence in
communicating in English, they were not as positive. It is possible that playing
computer games on three occasions only was simply not enough to help them build that
confidence, but it is also likely that participants felt safe in the confined space of the
game, access to which had been restricted from the general public. This shows that it
may be important for teachers to be careful about gradually opening up communication
and not to withdraw the perceived safety of the closed environment immediately, and to
provide ample support and feedback when doing so.
More generally, the above shows once again the importance of affective factors in
second language acquisition, especially the potentially negative impact of anxiety. Aoki
(1999, p. 149) suggests the development of a “psychologically secure environment”.
Clearly, participants in this study found such a secure environment in the digital game.
They confirmed this in their responses on the questionnaires, which showed they felt
they did not have to be embarrassed or anxious to make mistakes within the game.
Participants also felt that playing the game and communicating in English went together
and that they were therefore less conscious of themselves. Participants mentioned other
benefits too, such as the ability to develop their vocabulary, the fact that they had to
respond quickly and therefore became more fluent, and that they could practice English
outside the classroom.
What this study has shown, then, is that games are able to increase student
enthusiasm, lower anxiety, and improve willingness to communicate. These are of
course very valuable for supporting the second language acquisition process and digital
games may deserve a more frequent place in the second language curriculum.
References
Aoki, N. (1999). Affect and the role of teachers in the development of learner
autonomy. In J. Arnold (Ed.), Affect in language learning (pp. 142-154). Cambridge:
Cambridge University Press.
Burgoon, J. K. (1976). The unwillingness to communicate scale: Development and
validation. Communication Monographs, 43(1), 60-69.

Reinders & Wattana'
26
Cao, Y., & Philp, J. (2006). Interactional context and willingness to communicate: A
comparison of behavior in whole class, group and dyadic interaction. System,
34(4), 480-493.
Chapelle, C. A. (2001). Computer applications in second language acquisition: Foundations for
teaching, testing, and research. Cambridge: Cambridge University Press.
Chen, M., & Johnson, S. (2004). Measuring flow in a computer game simulating a
foreign language environment. Retrieved March 8, 2011, from
http://www.markdangerchen.net/pubs/flow_in_game_simulating_fle.pdf
Clement, R., Baker, S. C., & MacIntyre, P. D. (2003). Willingness to communicate in a
Second Language: The effects of context, norms, and vitality. Journal of Language
and Social Psychology, 22(2), 190-209.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale:
Lawrence Earlbaum Associates.
de Haan, J., Reed, M., & Kuwada, K. (2010). The effect of interactivity with a music
video game on second language vocabulary recall. Language Learning & Technology,
14(2), 74-94.
Derwing, T. M., Munro, M. J., & Thomson, R. I. (2008). A longitudinal study of ESL
learners' fluency and comprehensibility development. Applied Linguistics, 29(3),
359-380.
Ellis, N. (2002). Frequency effects in language processing: A review with implications
for theories of implicit and explicit language acquisition. Studies in Second Language
Acquisition, 24(2), 143-188.
Freiermuth, M., & Jarrell, D. (2006). Willingesss to communicate: Can online chat help?
International Journal of Applied Linguistics, 16(2), 189-212.
Garcia-Carbonell, A., Rising, B., Montero, B., & Watts, F. (2001). Simulation/Gaming
and the acquisition of communicative competence in another language.
Simulation & Gaming: An Interdisciplinary Journal of Theory, Practice and Research,
32(4), 481-491.
Gee, J. P. (2003). What video games have to teach us about learning and literacy. Basingstoke:
Palgrave Macmillan.
Hashimoto, Y. (2002). Motivation and willingness to communicate as predictors of
reported L2 use: The Japanese ESL context. Second Language Studies, 20(2), 29-70.
Jepson, K. (2005). Conversations--and negotiated interaction--in text and voice chat
rooms. Language, Learning & Technology, 9(3), 79-98.
Kamprasertwong, M. (2010). Willingness to communicate in English speech as a second language:
A study of Thai, Chinese, and Dutch Samples. Unpublished MA dissertation,
University of Groningen, the Netherlands.
Kang, S.-J. (2005). Dynamic emergence of situational willingness to communicate in a
second language. System, 33(2), 277-292.
Kern, R. G. (1995). Restructuring classroom interaction with networked computers:
Effects on quantity and characteristics of language production. The Modern
Language Journal, 79(4), 457-476.
Krashen, S. (1985). The Input Hypothesis: Issues and implications. London: Longman.
Larsen-Freeman, D., & Cameron, L. (2008). Research methodology on language
development from a complex systems perspective. The Modern Language Journal,
92(2), 200-213.
Lèger, D. d. S., & Storch, N. (2009). Learners' perceptions and attitudes: Implications
for willingness to communicate in an L2 classroom. System, 37(2), 269-285.
Levelt, W. J. M. (1989). Speaking: From intention to articulation. Cambridge: The MIT Press.

'Learn'English'or'die'
27
Long, M. H. (1996). The role of the linguistic environment in second language
acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of language
acquisition. Vol. 2: Second language acquisition (pp. 413-468). New York: Academic.
MacIntyre, P. D., Baker, S. C., Clément, R., & Conrod, S. (2001). Willingness to
communicate, social support , and language-learning orientations of immersion
students. Studies in Second Language Acquisition, 23(03), 369-388.
MacIntyre, P. D., Baker, S. C., Clément, R., & Donovan, L. A. (2003). Talking in order
to learn: Willingness to communicate and intensive program. Canadian Modern
Language Review, 59(4), 589–607.
MacIntyre, P. D., & Charos, C. (1996). Personality, attitudes and affect as predictors of
second language communication. Journal of Language and Social Psychology, 15(1), 3-
26.
MacIntyre, P. D., Dörnyei, Z., Clément, R., & Noels, K. A. (1998). Conceptualizing
willingness to communicate in a L2: A situational model of L2 confidence and
affiliation. The Modern Language Journal, 82(4), 545-562.
McCroskey, J. C., & Richmond, V. P. (1982 ). Communication apprehension and
shyness: Conceptual and operational distinction. Central States Speech journal,
33(4), 458-468.
Mortensen, D. C., Arntson, P. H., & Lustig, M. (1977). The measurement of verbal
predispositions: Scale development and application. Human Communication
Research, 3(2), 146-158.
Ortega, L. (2009). Interaction and attention to form in L2 text-based computer-
mediated communication. In A. Mackey & C. Polio (Eds.), Multiple perspectives on
interaction in SLA: Research in honor of Susan M. Gass (pp. 226-253). New York:
Routledge.
Pica, T. (1994). Research on negotiation: What does it reveal about second-language
learning conditions, processes, and outcomes? Language Learning, 44(4), 493-527.
Piirainen-Marsh, A., & Tainio, L. (2009). Other-Repetition as a Resource for
Participation in the Activity of Playing a Video Game. The Modern Language
Journal, 93(2), 153-169.
Samudavanija, C. (1999). Plearn Puea Roo (Enjoyment for learnng). Bangkok: Vajiravudh
College.
Schmidt, R. W. (1990). The role of consciousness in second language learning. Applied
Linguistics, 11(2), 129-158.
Steinkuehler, C. (2006). Massively multiplayer online videogaming as participation in a
Discourse. Mind, Culture & Activity, 13(1), 38-52.
Steinkuehler, C. (2007). Massively multiplayer online gaming as a constellation of literacy
practices. eLearning, 4(3), 297-318.
Swain, M. (1985). Communicative competence: Some roles of comprehensible input
and comprehensible output in its development. In S. Gass & C. Madden (Eds.),
Input in second language acquisition (pp. 235-253). Rowley, MA: Newbury House.
Sykes, J. (2005). Synchronous CMC and pragmatic development: effects of oral and
written chat. Calico, 22(3), 399-432.
Thorne, S. L., Black, R. W., & Sykes, J. (2009). Second language use, socialization, and
learning in Internet interest communities and online games. The Modern Language
Journal, 93, 802-821.
van Lier, L. (2004). The ecology and semiotics of language learning: A sociocultural
perspective. Norwell: Kluwer Academic Publishers.
Warschauer, M. (1996). Comparing face-to-face and electronic discussion in the second
language classroom. CALICO, 13(2&3), 7-26.

Reinders & Wattana'
28
Whitton, N. (2010). Learning with digital games: A practical guide to engaing students in Higher
Education. New York: Routledge.
Yashima, T. (2002). Willingness to communicate in a Second Language: The Japanese
EFL context. The Modern Language Journal, 86(1), 54-66.
Yashima, T., Zenuk-Nishide, L., & Shimizu, K. (2004). The influence of attitudes and
affect on willingness to communicate and Second Language communication.
Language Learning, 54(1), 119-152.
Zheng, D., Young, M. F., Wagner, M. M., & Brewer, R. A. (2009). Negotiation for
action: English language learning in bame-based virtual worlds. The Modern
Language Journal, 93(4), 489-511.
Biographical Statements
Dr. Hayo Reinders is Head of Learner Development at Middlesex University in London.
He is also Editor of Innovation in Language Learning and Teaching, and Convenor of
the AILA Research Network for CALL and the Learner. Hayo’s interests are in CALL,
autonomy, and out-of-class learning. He is a speaker for the Royal Society of New
Zealand. His most recent books are on teacher autonomy, teaching methodologies, and
second language acquisition and he edits a book series on ‘New Language Learning and
Teaching Environments’ for Palgrave Macmillan.
Homepage: www.innovationinteaching.org
Sorada Wattana is a lecturer at Dhurakij Pundit University, Thailand and a PhD student
at the University of Canterbury, New Zealand. Her enthusiasm for the role of computer
games in transforming the way people learn inspires her to investigate their potential for
language learning. She is currently devoting her time to her thesis writing under the
supervision of Dr. Michael Grimley, Professor Niki Davis, and Dr. Hayo Reinders,
which she thoroughly enjoys.