Task Design for Language Learning in an Embodied Environment - Teaching through Multi-User Virtual Environments - page 271

tice face-to-face teaching than a traditional concept
of either computer assisted language learning
(CALL) or mainly asynchronous online-learning
using a Virtual Learning Environment, so the type
of student assessment is also derived from good
face to face practice. It consisted of regular teacher
monitoring of performance in-class, recording of
work set for outside class with individual student
interviews where necessary. It is perfectly feasible
to set written tasks for completion within a time
limit although we did not use this instrument. As
products in development, clearly there was also
systematically collated feedback from students
(individual interview, focus group and web ques-
tionnaire) about their overall satisfaction which
very often brought up valuable learning issues.
Thus, assessment here differs from what is
normally referred to as computer based testing.
Though some of the tools associated with this area
such as web based multiple choice questionnaires
were available, they were generally used in survey
mode for gathering course evaluation information
rather than student performance information in a
testing mode. The MUVE aspect of assessment
was more apparent in devising metrics to capture
performance within the type of autonomous, simu-
lation tasks made possible by the MUVE. Thus
an assessment form would consist of:

Independent of the programme any student
was attending, they also had the opportunity to
meet up with an advisor to assess their progress.
It is certainly true that this was influenced by the
amount of time they were spending in Second Life
overall and, within this, the degree to which they
were seeking out informal practice and learning
opportunities. Through regular social as well as
lesson attendance some intermediate students
achieved the type of linguistic progress within
8 weeks or so which is normally associated with
140-170 hours of instruction (from CEFR level
B1-CEFR level B2).

Disadvantages of a mUVe as
Opposed to the Real Life Classroom

However engaging they may be, MUVEs are still
pictures on a screen and therefore facial under-
standing, haptics, and much body language are
not present, something which avatar movement
or animation cannot compensate for. It's there-
fore difficult to know if students are engaged or
understand the task set by the teacher. Students
need to be encouraged to ask questions.
Learners must learn to use the virtual environ-
ment before/or at the same time they learn any
language (although the MUVE could be totally
in the target language). The struggle of 'newbies'
(SL slang for new users) with the learning envi-
ronment can lead to high anxiety levels which
are counterproductive for language learning.
However, tutorials and a few insights during les-
son preparation can minimize this). There may be
technical issues, days when the central servers run
slowly which may make the class more difficult
to deliver, learners may have problems with their
voice or very bad sound which makes it difficult
to understand them or which causes disruption in
the classroom. However, technical issues can also
happen in the real life classroom just as easily.
Extended texts like pieces of student writing
or reading are more challenging to work with in
SL, so a multifaceted approach may be required

• Key language skills & competencies dis-

played in a task

• Task specific information (marked on a

scale of 1-4): task completion; grammar;
vocabulary; fluency; appropriateness;
pronunciation

• Open comments

In Second Life the camera can be detached from
the avatar and thus observe activity the equivalent
of several hundred metres away. This facility is
very useful for classroom quality control as well
as student observation. Note students are never
'spied on' anyone overhearing is visible in the
'active speakers' panel even if not proximate.