The virtual classroom: Videoconferencing for foreign language learners

The virtual classroom: Videoconferencing for foreign
language learners

Werner
Wegstein

Universität Würzburg
wegstein@mail.uni-wuerzburg.de

Derek
Lewis

University of Exeter
D.R.Lewis@exeter.ac.uk

2002

University of Tübingen

Tübingen

ALLC/ACH 2002

editor

Harald
Fuchs

encoder

Sara
A.
Schmidt

Institutional environment
In the mid-1980s the University of Exeter (Project Pallas) and the University
of Würzburg ('Linguistic Information and Text Processing') initiated courses
in Humanities Computing for undergraduate and postgraduate students. Student
and staff exchange programmes within the Erasmus/Socrates framework of the
European Union followed soon afterwards. By 1995 Exeter had launched its
ReLaTe project, to investigate the use of multimedia internet conferencing
for 'Remote Language Teaching (Buckett and Stringer 1997) together with
University College, London. Following on from this co-operation, trials with
the new medium were extended to investigate the feasibility of delivering
inter-university academic courses across national borders. A report on this
was published by the Würzburg Computing Centre in 1999 (Fahrner and Plehn
1999) and the results demonstrated at the CALL -
Challenge of Change conference at Exeter (2001). Here the authors
presented a scenario for providing and managing conversation classes
designed to allow students to practise foreign language skills with native
speakers across Europe using freely available videoconferencing tools
(Buckett et al. 2001). The study was based on work with first year students
of German at Exeter and postgraduate students of Humanities computing at
Würzburg learning English for special purposes. Building on this experience
we now outline the next stage in creating virtual classrooms based on
videoconferencing software: There are two aims. One is to extend the
conversation class setting to one in which students of English/German are
studying the language of IT and technical communication; this will enable an
investigation of the problems associated with the restrictions imposed by
the subject area in the context of 'live' videoconferencing. The second aim
is to record authentic class communication in order to build up an archive
illustrating the variety of interactions that may be exploited to teach
English/German for special purposes in the field of IT.

Impact of the new media: hardware, netware and software
Using a PC to see and talk to someone else working at a remote location
requires specific hardware (camera, graphics and audio cards, microphone and
headset) as well as software and netware tools to create and manage the
connection. Linking more than one PC at each location normally generates a
considerable load on the network. Using software for the Mbone configured to
handle multicast traffic, however, all participants can communicate with
each other efficiently and on equal terms (for details see Kumar 1996).
Online conversation with native language speakers creates a completely
different environment for language learning, compared with the standard
'schoolroom' setting of conversation classes. Normally foreign language
students, who outside class would communicate in their native language,
switch to a foreign language in order to discuss problems with their
teachers, language assistant or tutor. Using internet mediated communication
these students are able to practise naturally with native speakers, with all
the advantages which that entails: they learn to cope with regional
differences in speech, develop authentic communication strategies in the
foreign language, and become sensitive to the effect of subtle changes in
voice pitch, the function of pauses and to the vital area of non-verbal
communication. The very act of talking to native speakers (instead of each
other) is likely to stimulate the use of a quite different set of words and
syntactic structures.
In a normal face-to-face conversation class all the participants are in one
room, know each other and can draw on the physical/spatial dimensions of
speech events: sensing the direction of the sound, they can detect who is
speaking, can face him/her and use cues (such as the length of a pause) to
judge when he/she has finished in order to take a turn. Communicating via
the virtual internet, however, alters the spatial and perceptual context
appreciably: the participants are not in the same room; they do not know all
the other participants; they can hear in their headphone that some one is
speaking, although there is no direction of sound that could help them to
identify the speaker; and the audio signal in their headphones lags
perceptibly behind the video signal on screen, so that they may
misunderstand a slight pause as a signal to take over, prompting them to
interrupt the speaker. This lack of conversational coordination leads to a
kind of stop-and-go interaction in which more than one speaker starts to
take a turn, realizes that others have done the same, stops, and pauses for
quite a long time to establish who will finally dare to continue.
There is also a direct spatial relationship between the physical speech and
the visual information displayed on the computer screen. The size of the
person taking part in a face-to-face conversation class is greatly reduced
in the virtual, on-screen internet environment: here the size varies from
thumbnail (in the list of displayed participants) up to a maximum of a CIF
image. For the internet participants, who are not in the same room and do
not necessarily know each other, it is therefore indispensable to customize
the screen layout by captioning each window with the name of the speaker, so
that all the members of the conference know who he/she is.
The above factors have implications for positioning the video camera: this
should be as close to the screen the student is looking at as possible, so
that the listeners feel addressed by what the speaker is saying and the
speaker gets the impression that the others are attending to what he is
explaining (the so-called Casablanca Effect or
'Here's-looking-at-you'-principle). Ideally the camera image needs to be
large enough for the viewers to be able to see the speakers' facial
expressions (and possibly even to watch lip movements), whilst at the same
time having a sufficiently large field of vision that they can also see
gestures and general body language. The size of the conversation windows and
the screen hardware used (17 inches minimum) define the maximum number of
participants that can be handled effectively. We tested settings between
three and eight participants: with eight participants the conversation was
most lively, but not all participants took an active part in it; with three
or four participants all were really active. We measured the degree of
activity by comparing the number of sound packages sent by each participant.
We illustrate these problems and related questions by video clips from our
initial corpus, which are discussed below.

(1) An impact factor in the new medium is the relationship between the
audio and the visual dimension for the participants. Issues here
include: the size of the visual image of a speaker on the screen; the
role of direct eye contact; the ability to see more than the face (or
mouth) and the question how far the medium does intrude into the
interaction or otherwise inhibit it. Is there a relationship between the
language level (or even the language itself) and the influence of the
medium?
(2) Our experience so far suggests that the management of 'turns' or
hand-overs in the discourse is of particular interest. For instance, how
does a participant gain the attention of the group? How does one
interject? How is the participation of listeners registered if their
visual presence is reduced? Does a group need a designated manager of
the discourse or is interaction completely open? Should the group be
aware of 'rules' or procedures for managing interaction? What form
should these take?
(3) We show representative scenes that we intend to include in our
database of videoclips illustrating typical communicative interactions.
These clips will be transcribed using standard transcription systems for
discourse analysis (Edwards and Lampert 1993, Selting et al. 1998). The
materials will be used to introduce and prepare future students for
technical communication in a foreign language. In addition we plan to
use them for further research on how to enhance learning opportunities
and develop the communicative skills of students (studying, for
instance, technical translation) in an environment which offers
multimedia support.

We are aware that much research and activity are being undertaken in this
field and we will compare our experiences with the results of broader and
more general approaches to using the internet for teaching. Such approaches
may not necessarily include the use of live videocommunication in class (cf.
Warschauer 1999 and Warschauer, Shetzer and Meloni 2000).
In conclusion, our experience suggests that it is possible to enhance
interactive communication skills in the virtual internet medium but that
more research is needed to identify the optimum parameters for co-ordinating
the needs of learners in groups with the medium itself. In this paper we
focus on German-English interactions for the purposes of general
communication, with emphasis on initiating conversation, greetings,
turn-taking, eliciting information, and overcoming communication obstacles.
The corpus material will provide a searchable textbase that can be compared
with instances of classroom interaction and exploited for pedagogical
purposes.

References

J.
Buckett

N.
Datta

D.
Lewis

G.
Stringer

H.
Plehn

P.
Ruff

P.
Tscherner

W.
Wegstein

Conversation classes across Europe: A challenge for
videoconferencing

C.A.L.L. The Challenge of Change. Research and Practice

Presented by

K.
Cameron

Exeter
Elm Bank Publications
2001
169-176

J.
Buckett

G.
B.
Stringer

ReLaTe: Progress, Problems and Potential

Proceedings of CALL'97, Exeter, September 1997

1997

H.
Fahrner

H.
Plehn

Was ist Mbone?

Benutzermitteilungen Rechenzentrum Universität
Würzburg, Juni 1999

1999
24-27

V.
Kumar

Mbone. Interactive Multimedia on the Internet

Indianapolis
New Rider
1996

Mbone information in Germany is provided by "Beratungszentrum für
Videokonferenzdienste" at the Technical University of Dresden (), multicast is supported
very efficiently as standard service by the DFN consortium ("Deutsches
Forschungsnetz") in Germany; in the UK for Mbone software see .

J.
Schwitalla

Gesprochenes Deutsch. Eine Einführung.
Grundlagen der Germanistik 35

Berlin
Erich Schmitt
1997

J.
A.
Edwards

M.
D.
Lampert

Talking Data. Transcription and coding in discourse
research

Hillsdale
Lawrence Erlbaum
1993

M.
Selting

P.
Auer

B.
Barden

J.
Bergmann

E.
Couper-Kuhlen

S.
Günthner

C.
Meier

U.
Quasthoff

P.
Schlobinski

S.
Uhmann

Gesprächsanalytisches Transkriptionssystem
(GAT)

Linguistische Berichte

173
91-122
1998

M.
Warschauer

Electronic literacies: Language, culture, and power in
online education

Hillsdale
Lawrence Erlbaum
1999

M.
Warschauer

H.
Shetzer

C.
Meloni

Internet for English Language Teaching

Alexandria, Virgina
Teachers of English to Speakers of Other Languages
(TESOL), Inc.
2000