2.4 NEW TECHNOLOGIES

topsalmonAI and Robotics

Feb 23, 2014 (3 years and 1 month ago)

116 views

Draft, to appear

GELLOF KANSELAAR, TO
N DE JONG, JERRY AND
RIESSEN, & P
E-
TER GOODYEAR
1

2.4


NEW TECHNOLOGIES

1.

INTRODUCTION

Current trends dominating the field of learning and instruction are (socio)
-
constructivism
,
situ
a
tionism
, and
collaborative learning
. Respectively, these new

views on learning imply that learners are encouraged to
construct their own
know
l
edge

instead of copying it from an authority, be it a book or a teacher,
in rea
l-
istic situations

instead of decontextualised, formal situations such as propagated in
trad
i
tio
nal textbooks,
together with others

instead of on their own.

Technology

can play a major role in implementing these new trends in education.
In current research, constructive learning is supported by computer environments
such as hypermedia, concept mappin
g, simulations, and modelling tools (see
De
Jong & Van Joolingen, 1998a; DeCorte, 1990
). Realistic situations can be brought

into the classroom by means of digital video, as for example implemented in the
Jasper series (Cognition and Technology Group at Vanderbilt, 1992
). Collaborative
learning has been supported in environments such as CSILE (Scardamalia & Ber
e
i-
ter, 1996
) and Belvédère (Suthers, Weiner, Connelly, & Paolucci, 1995
) as well as
Internet based environments such as Virtual Classrooms. In this chapter we will r
e-
view some research projects in the Netherlands and a
broad, which are in line with
this evolving conceptualisation of new types of learning by using new technol
o
gies.

Before we discuss the theoretical background, we present the research topics of
this chapter in a graph (see
Figure
1
).


In learning situations in school, the information sources of a certain domain are pr
e-
sented to the student by using technological (top left corner) or social mediation
(bottom right corner). Learning is supported by social mediation (o
ther people in
face
-
to
-
face situations) or by using media such as books, computers, etc. (techn
o
lo
g-
ical mediation) (see
Figure
1
). Research on collaboration in face
-
to
-
face situ
a
tions
without focusing on technological media is dis
cussed in the chapter on “Co
l
labor
a-
tive learning” by Van der Linden, Erkens, Schmidt and Renshaw. In this chapter, we
focus on the use of new technologies and the way they influence ‘new’ learning.
Main features of new technologies are:


a)

Multiple represent
ations
. New trends in education, such as anchored instruction,
are easier to implement by using digital video. We present the use of digital vi
d-



1

The authors are grateful to Hermi Tabachnek
-
Schijf for her helpful comments.

NEW TECHNOLOGIES

3

eo and animations in learning the relation between moving objects (i.e. a car that
starts, slows down, etc.) and

the graphical representation in distance
-
time and
speed
-
time graphs in paragraph
4
. We also discuss the use of different repr
e
se
n-
tations (text, speech, and video) in learning words in a foreign language. Di
f
fe
r-
ent representati
ons of the domain knowledge can lead to different learning a
c
tiv
i-
ties and, hopefully, to more complex knowledge structures.

b)

Technological mediation

is used here to indicate learning activities that are po
s-
sible due to the interactive way the domain knowled
ge can be used in a computer
program. Examples of this idea are presented in paragraph
5
, where we discuss
discovery learning with computer simulations.

c)

Computer mediated communication.

In the last part of this chapter, paragr
aph
6
,
we focus on the possibilities of integrating the social and technological medi
a-
tion in computer mediated communication (CMC, the two arrows in the middle
of the square of
Figure
1
). Examples of
CMC are collaborative writing and a
r-
gument
a
tion.

Figure
1

Didactical

Square

The focus on media in this chapter is on new technol
o
gies.

2.

CHARACTERISTICS OF N
EW TECHNOLOGIES

First, we will define the term “new technologies”. By this
we mean electronic d
e
vi
c-
es that process information in digital form. Examples of such processes are sto
r
ing,
transporting, transforming, searching, generating and presenting of digital i
n
fo
r-
mation. Computers and the Internet are well known devices for proc
essing and
transporting; disks and CD
-
ROMs for storage and search. Digital video and audio
are ways of presenting information in varied and flexible ways.

New media process information at a symbolic level very flexibly, thanks to these
new capabilities. E.
g., one can program a computer to run a model of the behaviour
4


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

of a system and display that behaviour as needed in textual, numerical or graphical
form. Such a model can be made interactive: users can manipulate the values of p
a-
rameters in the model and ch
ange its behaviour.

The capabilities of new technologies enable changes in the representational basis
and in the processing characteristics of information (Kozma, 1991
), for example in
the fo
l
lowing ways:



non
-
li
near representation, e.g. hypertext;



multiple representations and transformation between different representations,
e.g., a spreadsheet table can be transformed into a graph;



dynamic representations, for instance, simulation of a process;



rule
-
based repres
entation of knowledge, allowing flexible implementation of
procedural knowledge;



electronic communication, both synchronous (chatting) and a
-
synchronous (di
s-
cussion forums and e
-
mail).

By using these characteristics, we can present realistic learning situa
tions in which
we can stimulate the personal construction of knowledge and the collaboration b
e-
tween students.

Although new media have wonderful processing capabilities, they were not deve
l-
oped as a response to a pedagogical imperative or need. Rather, af
ter the technology
was developed, it was co
-
opted for educational use. Sometimes, new media added
something to the old processes. Forty years ago, textbooks were the media to use to
become informed about foreign countries; nowadays, television adds realism

and
timeliness. At other times, new media takes something away


we use devices (e.g.
the symbolic calculator) that, nearly invisibly, perform all kinds of computational
processes for the learner.

Technology, initially, was not of much help in linking lea
rning theories and ed
u-
cational practices, as its development lagged far behind, and was at first only used
for drill and practice applications. But technology has developed very fast, and can
currently offer tools for intellectual partnership, to be employ
ed in constructivistical
learning environments for the manipulation and design, access and communication
of information (Salomon, 1997
). These new tools, in turn, are allowing new

co
n
ce
p-
tual
i
zations of learning and the development of new modes of use.

The relation between instruction and learning theories is not straightforward. We
don’t assume a deterministic relation between learning theories and the use of i
n-
structional technol
ogies, but learning theories are implicit in the design of learning
environments. Duffy and Jonassen (1991, p. 7) and Duffy, Lowyck and Jonassen
(1992
) argue in favour of
a firmer link between learning theories and instructional
practices.

3.

CHANGING VIEW ON LEA
RNING AND INSTRU
C
TION

In other chapters in this book attention is paid to new learning theories such as co
n-
structivism, socio
-
constructivism, situated cognition and an
chored instruction. In
this chapter, we will review only some distinctions that relate to new technologies.
Below, we first explain our view on constructivism as we use the concept here.

NEW TECHNOLOGIES

5

New learning theories have spawned a changing view on learning and in
stru
c-
tion. Constructivism and all it implies is perhaps the most important one. Constru
c-
tivism is not a single concept, but can involve the following three aspects:

a)

a set of epistemological beliefs (that is, beliefs about the nature of reality, whet
h-
er the
re is an independent reality
-

cf. Von Glasersfeld (1978)

or Cunnin
g
ham
(1992)
);

b)

a set of psychological beliefs about the nature of
mind, cognition and learning
(e.g. that learning involves constructing one's own knowledge);

c)

a set of educational beliefs about the best way to support learning (e.g., that d
i-
rect instruction through lecture methods is very limited or inappropriate; that e
n-
gaging materials, such as dramatic video cases are potentially very valuable; that
one should allow the learner to define their own learning objectives).

Some people argue that these three aspects are necessarily very tightly coupled. But
we want to argue

that there are circumstances in which one can talk about shifts in
(c) which are not significantly linked with shifts in (a) or (b). For example, one can
adopt video
-
based anchored instruction (following Cognition and Technology Group
at Vanderbilt (1992)
) without shifting in one's beliefs about whether there is, or is
not, an objective independent reality. Equally, lots of people give lectures who also
believe that learning is fundamentally about the individual construction of mea
n-
ing/knowledge. (There is

no contradiction, in principle, between these two.). Such
research and beha
v
ior, in our view, can still be constructivist.

We introduced “socio
-
constructivism” as a
new

learning paradigm. But aspect
(b) has a long tradition in Europe, e.g. Piaget (1952)
, Vygotsky (1978)

and Bartlett
(1932)
. With regard to learning mathematics, Freudenthal (1978)

has advocated
nearly the same ideas as socio
-
constructivism since the sixties. We will not restrict
ourselves to research that is only based on strong construct
ivist principles (aspect a),
but we will be broader in our consideration of empirical and other source material.

Constructivist principles, broadly defined as above, provide a set of guiding
principles to help designers and teachers to create learner
-
cente
red, collaborative
environments that support reflective and experiential processes (Jonassen,
Davidson,
Collins, Campbell & Haag,

1995, p. 8
). In this chapter three lines of research, a
n-
chore
d in this changing view on learning and i
n
struction, will be presented:

a)

situated learning in ‘realistic’ situations;

b)

discovery learning with computer simulations;

c)

collaborative learning with computers.

In each line of research we will discuss the relation

between underlying learning
principles and the use of new technologies.

4.

SITUATED LEARNING IN

REALISTIC SITUATIONS

Most of the research in situated cognition and the use of ‘realistic’ situations has
been done in science and mathematics. One of the best kn
own examples in educ
a-
tion is the Jasper series of the Cognition and Technology Group at Vanderbilt (1991,
1992, 1993)
. Constructivist principles of the series, in which Jasper Woodburry stars
in adventurous stories, led to a video
-
based presentation format to anchor instru
c-
6


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

tion, narrative format, embedded data design and complex authentic mathematical
problem s
olving. The assessment outcomes of the Jasper series were positive (1992,
p. 19), although alternative assessment procedures are necessary to demonstrate the
progress students made in other areas than traditional math problem solving. So,
anchoring instruc
tion in ‘real’ life situations by using digital, interactive video was
successful.

4.1

Situated cognition and mathematics

Offering rich, situational settings and the use of multiple representations has been a
prominent research topic in mathematics learning. R
esearching the use of situated
simulation models has become a central topic of attention in mathematics education.
With early, well
-
designed concrete models, students were expected to discover the
mathematics principles that were embedded therein. The inhe
rent problem of this
approach is that the knowledge represented by concrete models is clear only to those
who already possess this knowledge, having the long
-
term memory structures to “fill
in the gaps”. To paraphrase Tabachneck
-
Schijf and Simon (1996)
:

“Simple perception can yield the answer, but only to a student who has learned to n
o-
tice the relevant features of the presentations, who has acquired the appropriate infe
r-
ence operators and can thus make the needed inferences, and who can translate back
and forth between these operators and their inte
r
pretations” (p. 37).

Students thus merely see the concrete manifestations of these models, because for
them there is no ma
thematical knowledge to see (Tabachneck
-
Schijf & Simon,
1998
). Students’ progress in pure discovery models proved therefore too slow


there is too much knowledge missing.


Socio
-
constructivist theory stimulated research that tried to look at models from
an actor’s point of view instead of the observer’s point of view. The focus of this
research is on attempting to get students to link their experiential knowledge, often
ba
sed on naïve models, with formal knowledge. Past research has described such
naïve models and has shown that students encounter much difficulty integrating
them with formal knowledge. They often keep the two models separated (Gentner &
Stevens, 1983
) and thus lack the links to make formal knowledge useful in everyday
life.

One current type of research concerns making connections between motion as
perceived (experiental representation) an
d motion as represented mathematically in
graphs (formal representation) (Boyd & Rubin, 1996
; Scanlon, 1998)
. The research
of Kanselaar, van Galen,

Beemer, Erkens and Gravemeijer (1999)

examines how
students explore and understand the relations between graphs in situations where
digitised videos help them to revisit and refl
ect on an object’s motion. The a
d-
va
n
tage of digital video is that it is structurally segmented (i.e. in frames) and ra
n-
domly accessible, making it possible to measure certain aspects such as height and
fr
a
menumber, but when “played” can be experienced as c
ontinuous. When the e
x-
per
i
menter prepares the video, the students can measure distances between several
points of moving objects in the video. Video serves here as a medium to connect
experiential everyday and repr
e
sented worlds.

NEW TECHNOLOGIES

7

One of the research quest
ions in the research project of Kanselaar et al. (1999)
was how much guidance students needed in order to draw and understand graphs of
moving objects on a video. In one condition of our experiment the students were
guided by more precise questions and hin
ts (guided condition), in the other cond
i-
tion they were unguided, left on their own to find out how to answer more general
questions on drawing graphs. The students manipulated a video that showed a Ferris
wheel on the Dam (central square) in Amsterdam (se
e Figure 2). The students could
measure time, distance traveled, and number of seats on the video
-
images of the
Ferris wheel. Graphs could be presented on the screen in para
l
lel with the video.


Figure
2

Ferris Wheel on the Dam

Exa
mples of the more general questions the students had to answer were: When you
are seated in the Ferris wheel, how many seconds do you spend above the level of
the Palace on the Dam? Can you construct a graph that represents the number of
people getting in
and out the wheel during the time that you spend seated in the
wheel? Compared to students in the less guided condition, those in the guided co
n
d
i-
tion showed a lot of local behaviour, while global understanding was left wanting.
They did not understand wel
l enough what they were doing and the overall goal of
their activities. On the other hand, the students in the less guided condition were
more often satisfied with incomplete answers. Both conditions turned out to have
different advantages and disadvantage
s: the right amount of structure and support in
such realistic presentations is proving hard to define.

Not only the realistic aspects of the representations are important. We also pr
e-
sented the students with an animation in which four girls had to run a
race in the
playground. Students immediately posed the question: “Who wins?”. To answer this
question, they looked for the relation between the animation and the graph. In this
case they formulated relevant questions by their own. In another problem they h
ad to
8


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

analyze the relation between a car in a video that starts, accelerates, slows down and
stops, and the representation of distance
-
time and speed
-
time graphs. In this case the
students had problems in formulating the questions. So, realistic (video) re
present
a-
tion is not always easier to interpret than simple animations. Knowledge about the
real life situation, to which the artificial representation refers to, is also a very i
m-
portant determinant for the learning activities of the students.

4.2

Situated co
gnition and foreign language learning

The principles of situated cognition can also be applied to learning foreign la
n-
guages. One learns a foreign language in order to be able to communicate. Situated
cognition principles prescribe that the structuring of
learning materials should first
and foremost be determined by communicative needs. The traditional grammar
-
translation method, in which vocabulary was learned mostly by memorising L2/L1
word pairs did not serve this main purpose and did not directly lead t
o fluent oral
proficiency, though it largely enabled learners to read material in the foreign la
n-
guage. In the communicative approach, however, the linguistic context, the situ
a-
tion, speakers' roles and types of texts are all taken into careful considerati
on.

One aspect of the communicative concept in foreign language learning is a di
f-
ferent view of how vocabulary should be acquired. Nagy, Herman and Anderson
(1985, 1987
) showed that, usi
ng traditional approaches, learners between six and
sixteen years of age extended their vocabulary mainly by deriving word meanings
from reading words in meaningful contexts. Building one’s vocabulary this way has
several disadvantages. First, it has shown

to be slow. Situated cognition would pr
e-
dict the cause of this problem to lie in the fact that in order to effectively use a word,
the learner has to encounter the word in different contexts. As a result, if several
contexts are not offered within a short

period of time, the effectiveness of learning
words from contexts becomes visible only in the long term. To speed up vocabulary
building, then, various clear and informative contexts should be offered in a brief
time p
e
riod.

Kanselaar (1994)
designed an

experiment offering students a rich, situated env
i-
ronment containing such contexts. A computer program called “IT’S
-
English” on
CD
-
ROM (Kanselaar, 1994
; Jaspers, Kanselaar & Kok, 1993
) was used that co
n-
tained a total of over 5000 words and 2000 ‘context sentences’, synonyms, etc., u
s-
ing the
Collins Cobuild English Language Dictionary

(Sinclair 1987) as a source
.
This dictionary contains 70,000 keywords, which are defined and cited in "real En
g-
lish" context sentences. In
IT’S
-
English
, each individual word, its definition in En
g-
lish (the foreign language), one or more context sentences, and it
s pronunciation are
available, thus offering multiple layers of meaning and representation. It is also po
s-
sible to link short digital videos to certain words. While reading and listening to a
text, the user can call up various types of information as optio
ns.
IT’S
-
English

co
n-
tained different types of exercises, one of which was completing
Cloze texts

(a r
e-
productive exercise). In cloze texts words that are deleted from a text have to be
inserted by the learner.

NEW TECHNOLOGIES

9

In the experiment, first the frequency with w
hich learners used the available o
p-
tions was examined in order to determine the usefulness of the different represent
a-
tions of the words. In the cloze texts, more than half of the information called up
consisted of context sentences (also missing the delet
ed word). The word to be filled
in is thus placed in several contexts, in addition to the context given in the exercise
itself, offering various ways of storing (and later finding) the information in one’s
mind. Guessing the word from its meaning definitio
n was also a frequently chosen
option: one
-
third of all the information called
-
up in cloze texts consisted of meaning
definitions.

Secondly, the researchers examined the usefulness of
IT’S
-
English

as a teaching
method by comparing the results of pupils be
ing taught English with traditional
methods to those using the

IT’S
-
English

environment. The pupils who had been
taught English with the communicative program proved to have gained more from
their experiences. Their knowledge of newly acquired English voca
bulary was
greater than that of the pupils who had been taught English with the more traditional
method; the researchers concluded that the representation of multiple and authentic
contexts had a positive effect on learning vocabulary of a foreign language
.

5.

DISCOVERY LEARNING W
ITH COMPUTER SIM
U
LATIONS

One of the new themes in learning and instruction clearly is the emphasis on lear
n-
ing as the
personal

construction

of knowledge. Technology can play an important
role in this approach by offering environments

that encourage learners to engage in
self
-
directed constructive learning processes. In the literature (see for example,
Duffy, Lowyck, & Jonassen, 1992
; Schank & Cleary,
1995
; Towne, De Jong &
Spada, 1993
) we find examples of ‘constructivistic’ computer based learning env
i-
ronment
s such as hypertext environments, concept mapping environments and mo
d-
elling environments. Simulation environments take a specific place.
Computer sim
u-
lations

are programs that contain a model of a real system. Basic actions the learner
can perform using a
n interface to the model are changing values of input variables
and observing the resulting changes in values of output variables (De Jong, 1991
; De
Jong & Van Joolingen, 1998a
; Reigeluth & Schwartz, 1989
).

In a survey of Dutch higher education, simulation was found to be the most po
p-
ular form of computer based learning environments (De Jong,
Van Andel, Leiblum,
& Mirande, 1992
). We can think of several reasons for this. The first is that compu
t-
er simulation is very well suited for
dis
covery learning
, in which lear
n
ers exper
i-
ment and construct knowledge like ‘scientists’. That is, they provide the simulation
with input, observe the output, draw their conclusions, and go to their next exper
i-
ment. It is also believed that knowledge that i
s gained by a process of self
-
directed
discovery learning has a deeper, better anchored, more intuitive cha
r
acter than
knowledge that is gained in the traditional lecture
(Grimes & Willey, 1990
;
Faryniarz & Lockwood, 1992
; Carlsen & Andre, 1992
; Rivers & Vockell, 1987
;
Swaak & De Jong, 1996
)
.
In addition to the expected
learning advantages
, simul
a-
tions are introduced in instruction for a number of other reasons. First, pract
i
ca
l re
a-
sons can make a simulation preferable to the real training situation, for exa
m
ple
10


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

when training on the job is dangerous to man, environment and/or material. Se
c
ond,
simulations offer the opportunity to change reality in such a way that learning is
fac
ilitated, for example when the time scale is changed so that real processes can be
slowed down or speeded up (see De Jong, 1991
). Reality can also be ‘augmented’ in
simulations. For example, in a simulation on optics in which defle
ction of light
through lenses is being studied, Hulshof, De Jong and Van Joolingen (1999)

intr
o-
duced an artificial ‘eye’ that can be used by learners t
o ‘see’ virtual points. In a
n
ot
h-
er simulation on learning to drive a car V
an Emmerik, Van Rooij and De Jong (in
preparation
)
introduced a visual cue to indicate the distance to an approaching car
for a driver who wants to overtake.

Another advantage of simulation is its motiv
a-
tional appeal (Ajewole, 1991
) and their relative efficiency over expository modes of
teaching (Choi & Gennaro, 1987
; Rivers & Vockell, 1987
; Shute & Glaser, 1990
).
The
disadvantage of simulations
, however, is that discovery learning proves to be a
difficult process. Here we give a short summary of De Jong and Van Joolingen
(1998a)

who present an extensive overview of many kinds of problems that have
been found in studies on discovery learning:



Finding new hypotheses
. A difficult process that distinguishes successful from
unsuccessful learne
rs. Important problems here are: (1) learners (even university
students) simply may not know what a hypothesis should look like; (2) learners
may not be able to state or adapt hypotheses on the basis of data gathered; (3)
learners can be led by considerati
ons that don’t necessarily help them to find the
correct (or best) theoretical principles. For example, they do not like to state h
y-
potheses that run a high risk of being refuted.



Design of experiments.

A crucial aspect that provides information for decidi
ng
upon the soundness of a hypothesis. When a learner does not yet have a hypoth
e-
sis, well
-
designed experiments can be used to generate ideas about the model in
the simulation. In the literature we find a number of phenomena indicating poo
r-
ly designed expe
riments: (1)
confirmation bias
, the tendency to seek for i
n
fo
r-
mation that confirms the hypothesis they have, instead of trying to disconfirm the
hypothesis; (2)
designing inconclusive experiments,

for example, in the co
n
text
of discovery learning with simu
lations, Glaser, Schaubel, Raghavan and Zeitz
(1992)
, point to a frequently observed phenomenon that learners tend to vary too
many variables in one experiment, with the result that they cannot

draw any co
n-
clusions from these experiments; (3)

inefficient experimentation beha
v
iour
. For
example, it is often found that subjects do not use the whole range of potential
informative experiments that are available, but only a limited set, and moreover
d
esign the same experiment several times; (4)
constructing exper
i
ments that are
not intended to test a hypothesis
. Schauble, Klopfer and Raghavan (1991)

ident
i-
fied what they have called the “engineering approach”, which d
e
notes the att
i-
tude to create some desirable outcome instead of trying to unde
r
stand the model.



Interpreting data.
Once having performed correct experiments,
data

that come
from these experime
nts
need to be interpreted

before the results from the e
x
pe
r-
iments can be translated into hypotheses on the domain. Here learners quite often
make misencodings. Klahr, Fay and Dunbar (1993)

found that subjects made m
i-
sencodings of experimental data ranging from a mean of 35% of at least one m
i-
NEW TECHNOLOGIES

11

sencoding, to a high 63%. Also the interpretation of graphs, a frequently needed
skill when interacting with simulations, is clearly a di
fficult process.



Regulative processes:

it is frequently reported that successful learners use sy
s-
tematic planning and monitoring, whereas unsuccessful learners work in an u
n-
systematic way (e.g., Lavoie & Good, 1988
; Simmons & Lunetta, 1993
). Shute
& Glaser (1990)

claim that successful learners plan their experim
ents and m
a-
nipulations to a greater extent, and pay more attention to data management i
s-
sues. Glaser et al. (1992)

report that successful discoverers followed a plan over
experiments, whereas unsuccessful ones used a more ra
ndom strategy, conce
n-
trating on local decisions, which also gave them monitoring problems. Though
Glaser et al. (1992)

mention persistence in following a goal as a characteristic of
good learners, these successful subjects a
lso were ready to leave a route when it
apparently would not lead to success. Goal setting is also reported as a problem
(for subjects with low prior knowledge). For the process of
monitoring
diffe
r-
ences, Lavoie and Good (1988)

report that good learners make more notes during
learning, and Schauble, Glaser, Raghavan and Reiner (1991)

report more sy
s-
tematic data recording amoung succes
s
ful learners.

The conclusion that learners have difficulties with discovery learning indicates that
it might be necessary to
support the learner in the discovery process

in

order to i
n-
crease the efficiency and effectiveness of discovery learning.
S
IM
Q
UEST

is an a
u-
thoring environment that helps authors to create simulation based learning env
i
ro
n-
ments that are embedded in instructional support (see e.g., De Jong,
van Jooli
n
gen
,
Swaak, Veermans, Limbach, King, & Gureghian
, 1998b
).
Currently, the
S
IM
Q
UEST

author
ing environment provides the opportunity to create four types of instructional
support for lear
n
ers:



Model progression.

A learning environment created with
S
IM
Q
UEST

may contain
a number of different simulation models, ordered for example along a dimension

of difficulty. Introducing gradually more complex models helps the learner in the
regulative processes, since at the start the number of variables is limited and i
n-
creases per model progression step. When the environment is less complex there
is less to m
onitor and plan. Model progression also helps to stay close to the
learner’s starting knowledge. Through model progression, concepts that are new
and unfamiliar are introduced only when the learner has successfully mastered
the earlier model progression le
vel(s). In this way model progression may also
help the learner in forming better hypotheses.



Assignments
. Assignments provide the learner with short
-
term goals, like finding
a specified relation, predicting the behaviour of the simulation or achieving a
s
pecified simulation state. In conjunction with model progression, assignments
decompose the overall learning goal of a simulation into a number of subgoals
and in this way help the regulation of the learning process. The set of assig
n-
ments can also be used

to make the learner explore the complete domain and, by
setting the simulation in a specific state, have learners experiment with specific
phenomena.



Explanations.

In the
S
IM
Q
UEST

system the author can define textual, graphical,
and multimedia explanation
s. These explanations can be used to provide extra
12


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

information on variables, relations, or events in the simulation. These explan
a-
tions are always directly accessible to learners. Explanations give learners direct
access to necessary prior and background k
nowledge, which may help them in
interpre
t
ing data and stating hypotheses.



Monitoring
. The monitoring tool helps learners save, compare, and replay the
experiments they have been doing, and can provide feedback on the relation b
e-
tween the experiments and a
nswers chosen (see De Jong et al., 1998b
).

ing, and stepping through the simulation).

Figure
3

gives an example of a part of a learner interface created with
S
I
M
Q
U
EST
. This example is part of a large
S
IM
Q
UEST

learning environment on se
w-
age plants. Part of this environment concerns the role that bacteria play in the purif
i-
c
a
tion of water. The elements shown in
ing, and stepping through the simulation).

Figure
3

come from the first model progression level where learners can only o
b-
serve (and not manipulate) a phenomenon (by starting, pausing, and stepping
through the simulation).

Figure
3

Example of parts of a simulati
on environment on growth of bacteria (D
e-
veloped by Jan Blankenburg, IPN,.Kiel)


NEW TECHNOLOGIES

13

The output is presented in a graph, numerically, and by an animation at the bottom
of the simulation window showing bacteria that divide. The assignment asks learners
for expla
nations of what they can observe (there are multiple correct answers). After
selecting answers feedback is presented. The figure also shows an example of an
explanation presenting background information. In later levels of this simulation,
learners can the
mselves manipulate the stocks and inflow of nutrients for the bact
e-
ria and observe the effects on the bio
-
mass.

Several empirical studies were conducted to evaluate the support the researchers
introduced (e.g., De Jong, Härtel, Swaak & Van Joolingen, 1996
; Swaak, Van
Joo
l
ingen & De Jong, 1998
; De Jong

et al., in press;

Swaak, Blokhuis, Gutierrez, &
López, 1997).

The overall concl
usions of these stu
d
ies are as follows:



Intuitive knowledge improves.

There is an improvement in learning if measured
with intuitive knowledge tests. Several studies were conducted under the a
s-
sumption that simulation would lead to a more qualitative, intu
itive, type of
knowledge (see Swaak & De Jong, 1996
, for a more detailed definition) than the
knowledge that is acquired in more expository ways of teaching. The overall r
e-
sults of these stud
ies are that in learning with simulations, intuitive knowledge
improves to a far larger extent than does definitional know
l
edge.

Assignments and explanations are very popular.
The learners consult instru
c-
tional measures such as assignments and explanations

very frequently. In the
S
IM
Q
UEST

environments, learners have access to a large number of assignments and
explanations (see
ing, and stepping through the simulation).



Figure
3

for illustrations of explanations and assignments). Th
e assignments are
of several types (e.g., explaining a phenomenon) and the explanations include
many kinds of multimedia. Logfile analysis shows that learners use most of the
assignments and most of the available explanations to such a degree that it may
b
e problematic to describe this as free discovery behaviour (see De Jong et al.,
1998b
).



Adding assignments improved results.

Adding specific assignments to a simul
a-
tion environment improved t
he learning results significantly as compared to the
same simulation without specific assig
n
ments.



Results of adding model progression are inconclusive.

The introduction of model
progression, this is starting with simple models before introducing more comp
lex
ones, did not raise student’s learning results as compared with environments
where the most complex situation was given from the start, but the studies could
not give a concl
u
sive picture here.



Adding instructional measures, overall, did not increase t
he experienced cogn
i-
tive load
. A concern when adding instructional support to a simulation is that the
complexity of the environment as a whole increases, which may raise the exper
i-
enced cognitive load. In several experiments the cognitive load experienced

was
measured by using an on
-
line self
-
rating questionnaire.

These findings are in line with other studies on discovery learning with simulations
as we find them in the literature (see e.g., De Jong & Van Joolingen, 1998a
).

14


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

6.

COLLABORATIVE LEARNI
NG WITH COMPUTERS

“The idea that authentic learning only occurs in collaboration with others has b
e-
come the central pillar of constructivist orthodoxy and is the one on which pract
i
ca
l-
ly every other principle is dependent to s
ome extent” (Petraglia, 1998
, p.77
). The
computer can support collaborative learning in several ways. For example, Erkens
(1997)

distinguishes four different types of use:

(1)
Computer
-
based collaborative tasks (CBCT):

the computer presents a task
environment to foster student collaboration. The extra advantages of the medium
(compared to collaborating without it) may

be the shared problem representation
that can function as a joint problem space, the ease of data
-
access, and, in some ca
s-
es, intelligent coaching. Example systems are Sherlock (Katz & Lesgold, 1993
) and
the Envisioning Machine (Roschelle & Teasley, 1995)
.

(2)
Co
-
operative tools (CT):

the computer is used as a co
-
operative tool, a par
t-
ner who may take over some of the bu
rden of lower
-
order tasks, while functioning
as a (non
-
intelligent) tool during higher
-
order activities. Examples are Writing Par
t-
ner (Salomon, 1993
), CSILE (Scardamalia, Bereiter & Lamon, 1994
), and Case
-
Based Reasoning tool (Kolodner, 1993
).

(3)
Computer mediated communication (CMC), or Computer
-
Supported Co
l
la
b-
orat
ive Learning (CSCL)
: supports collaborating over electronic networks. The
computer serves as the communication interface, which allows interaction and co
l-
laboration between several students at the same time or spread out asynchronously
over a specific peri
od. Email conferencing (Henri, 1995
) and GroupWare systems
(Mitchell & Posner, 1996
) fall into this category. In addition, representations and

interfaces that support problem solving and communication are sometimes provided,
such as in Chene (Baker & Bielaczyc, 1995
), Belvédère (Suthers, Weiner, Connelly,
& Paolucci, 1995
), or the Collaborative Text Production Tool (Andriessen, Erkens,
Overeem & Jaspers, 1996
).

(4)
Intelligent Co
-
operative Systems (ICS)
: to set it off from (2) in ICS the co
m-
puter functions as an intelligent co
-
operative partner (DSA: Erkens, 1997
), a co
-
learner (People Power: Dillenbourg & Self, 1992
), or learning companion (Integr
a-
tion
-
Kid: Chan & Baskin, 1990
).

In the curr
ent chapter, we focus on the third type of computer supported tool:
CMC and CSCL. These types of applications involve the use of distance learning
software, i.e. software by which people collaborate over a network. The role of the
computer is to facilitate

learning by providing means for communication and pro
b-
lem solving. Collaboration over a network may also incorporate aspects of the other
types of applications for collaborative learning, such as shared problem represent
a-
tions, and the use of tools or int
elligent agents. In addition, it evokes its own inte
r-
es
t
ing prospects for collaboration, for example in the form of electronic discussion
tools. For the purpose of this section, collaborative learning occurs between two or
more learners engaged in a task t
hat requires them to co
-
operate and interact. The
role of the instructor is to support and sometimes coach the interaction, but our main
focus of interest is on the activities carried out by collaborating learners.

NEW TECHNOLOGIES

15

Two closely related questions are centra
l here: (a) what are the characteristics of
electronic collaboration for the purpose of learning something?, (b) how to support
such electronic collaborative learning, for example in terms of interfaces, shared
representations, tools, moderating and coachi
ng agents?

The task domains, we focus on in this section are (1) writing and (2) electronic
argumentation. A comprehensive overview of the pros and cons of collaboratively
carrying out these tasks in electronic environments is as yet lacking. Therefore, th
e
presentations and discussions that follow remain rather exploratory. The domains we
studied can be considered as characteristic of 'new learning' domains, as they pertain
to processes and skills that are not meant to be acquired for their own sake (e.g.
what
is the goal of writing?), but are a crucial part of many other tasks, in school as well
as in daily life (e.g. I want to write a love letter to a girl). In addition, they require
extensive negotiation and the acquisition of the needed language of a co
mmunity of
practice or academic profession (Petraglia, 1998
; Andriessen & Sandberg, in press
).
In the discussion that
follows we assume the following:



The assignment supports the acquisition of something else, e.g. general abilities
and skills applicable in many other tasks and domains. Hence sp
e
cific task
-
related processes such as task acquisition or successful problem
-
solving are not
our pr
i
mary focus of attention.



Collaborative assignments should be truly collaborative, that is, both (all) par
t-
ners need to particpate fully in the collaboration in order to accomplish a task or
reach a solution.



Electronic environments f
oster interaction and problem solving in specific and
not always obvious manners. The main goal for the study of collaboration in
such environments is to analyse the complex interactions between collaborators
and the relation of interactions to information

exchange, in order to find the sp
e-
cific constraints under which different electronic environments foster (or hinder)
specific kinds of collaborative learning. Social, cognitive, and technological co
n-
straints seem to be interdependent in a variety of ways,

depending on the task,
the instruction, the roles of instructors and students, student characteristics, the
medium and the interface (Veerman, Andriessen & Kanselaar, submitted
).

6.1

Collaborative writing

Writing clearly is a domain of interest for new learning. When writers are purpos
e-
fully engaged in knowledge transforming activities (Bereiter & Scardamalia, 1987
),
writing can even be one of the main vehicles to foster learning. The problem of ge
t-
ting writers engaged in knowledge transforming activities, however, is quite serious.
Several studies have shown that writers of var
ious ages, either novices or experts, do
not engage in knowledge transformation by default (e.g. Torrance, 1996
). A cause
can be that writing texts of any length has been shown to be a complex process in
w
hich several interrelated subprocesses can be distinguished, each with its own d
y-
namics and constraints (for a review: Alamargot & Chanquoy, in preparation
).

The main advantage of
collaborative writing, compared to individual writing, is
to offer a workspace where the writers can receive immediate feedback from each
16


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

other on their writing actions. Furthermore, the discussions generated by the activity
make the collaborators verbalis
e and negotiate many things: representations, pu
r-
pose, plans, doubts, etc. Collaborating writers have to test their hypotheses, justify
their propositions, and make their goals explicit. This may lead to progressively
more conscious control and increased a
wareness of the processes (Giroud, 1999
).
According to the literature (e.g. Gere & Stevens, 1989
; Piolat, 1990
; Roussey &
Gombert, 1992
), the dialogues created by all these activities lead to more revision,
more critical control, and more consideration of audience.

Electronic collaborative w
riting tools could provide support in several ways (D
i-
ermanse, 1997
):



Brainstorming:

although there is evidence showing that individual brainstorming
tends to be more effective than collective brainstorm
ing (Kanselaar & van der
Linden, 1983
), it may be that after an individual phase some negotiation about
the feasibility of generated ideas could be supported by a tool t
hat supplies a
common window and prompts such as "are there conflicts between ideas?" or
"are all ideas realistic?" (Sharples, 1993
).



Concept mapping:

concept mapping may be quite effectively carried out i
n co
l-
laboration (Roth & Roychoudhury, 1993
; Van Boxtel, Kanselaar & van der Li
n-
den, 1998,

and a graphical tool allowing collaborative concept mapping could be
quite useful.



Planning:

in contrast to individual planning, collaborative planning is not co
n-
sidered very important or fruitful (Beck, 1993
) and it is unclear how it should be
supported. Maybe providing schemas or outliners could be useful, but for lear
n-
ing purposes, these should not constrain reflection too much.



Information retrieval:

many tools are available for searching in
formation, but
collaborative search and selection (in any other sense than CBCT, see above) is
not supported.



Note
-
taking:

collective and individual notebooks should be provided.



Writing text:

collaborative writing tools currently provide only rudimentary
word
-
processing capabilities (apart from situations of sharing the same word
-
processor). Sharples and O'Malley (1992)

describe the design for a writer's a
s
si
s-
tant that allows suppor
t at the level of management of writing activities. At the
interface, turn taking should be managed and all interactions should be logged
and open for i
n
spection in a user
-
friendly way.



Revising:

users should be able to evaluate and revise their text easi
ly. Procedural
facilitation, which has been shown to be useful for individual writing (Bereiter &
Scardamalia, 1987
; Salomon, 1993
) could foster collabo
rative writing as well.
How collaborative revision should be fostered electronically has not been stu
d-
ied.

At our laboratory in Utrecht, we study electronic collaborative text production with
respect to the relationship between characteristics of collabora
tion on the one hand
and learning and problem solving on the other hand. A network
-
based (Collabor
a-
tive Text Production: CTP
-
) tool was developed which combines a shared word
-
processor, chat
-
boxes and private information sources to foster the collaborative

distance writing of texts. Collaboration and the sharing of windows is restricted to
NEW TECHNOLOGIES

17

dyads of students. The working screen of the program displays several private and
shared windows (see
Figure
4
). The two private information win
dows at the top
(“Task window” and “Arguments” window) both contain task information. The
Task Window displays the task assignment and the Arguments window displays
additional information an individual participant is provided with. A
turn pages bu
t-
ton

may
be used to turn the pages when the reasons are represented in pictorial fo
r-
mat. To communicate with his partner the student has a Chat Box. It can be used to
write messages that simultaneously can be seen by the writing
-
partner in the Other's
Chat Window.
This arrangement allows partners to send messages simultaneously.
When a message is ready, pressing the return button will enter it into the shared Chat
History
,

where the previous dialogue is available for review by both partic
i
pants.


Figure
4

Collaborative Text Production Tool (CTP)

The CTP Text Box is the shared space in which the participants may enter, edit and
revise the text they are currently writing. They both can write in the same text but
not simultaneously. Two button
s and a traffic light under the CTP Text Box (here
barely visible) are used to signal turn
-
taking intentions and turn giving. The CTP
collaborative writing
-
tool may be used in different contexts for practical and r
e-
search purposes.

One study of the CTP sy
stem (Andriessen et al., 1996
) showed that, in the wri
t-
ten product, students explored multiple viewpoints and elaborated upon their arg
u-
ments, despite the fact that no dialogue moves or t
urn
-
taking controls were available
in the communication window. In this study the system was used to gather data to
18


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

study the effects of external information representations on argumentative text. The
discussion by the participants, the chat
-
messages, the
button
-
actions and all changes
in the text were logged in a time
-
based protocol to be used for further analysis. In
this experiment, 74 university students in social sciences, working in pairs, were
instructed to write two texts (1) considering the problem

of the overpopulation of
rabbits and (2) considering labour policy on employability. Pairs were randomly
assigned to two different conditions. Students were provided with some predefined
arguments in (a) te
x
tual format or in (b) graphical representations.

Except for turn
-
taking facilities in the CTP Text Box, interaction was not structured. The two cond
i-
tions affected the frequencies of reasons of different types. The pictorial information
gave rise to a greater number and variety of elaborations in the wr
itten products than
did the textual information. This, however, did not relate to more coherent texts or
advanced text production strategies. Content elaboration and coherent collaborative
writing seemed to rely on different processes. We are currently ana
lysing the co
l
la
b-
oration protocols from the Chat Histories and their relationship to text characte
r
i
s-
tics.

6.2

Collaborative argumentation

A related domain is learning by collaborative argumentation. Collaborative arg
u-
mentation can be understood as two or more

persons engaging in a discussion about
problems and issues triggered by an assignment with some specific learning goal. In
one sense of the term, learning can be understood as the process that transpires
through argumentation among teachers, students and
their real worlds (Petraglia,
1998
). In what way learning results through such a process is quite unclear. Pr
o-
po
s
als have been put forward concerning mechanisms such as belief revision, co
n-
ce
p
tual change, externalising know
l
edge
and opinions, self
-
explanations, co
-
construction of knowledge, reflection and the reconstruction of knowledge through
critical discussion (e.g. Piaget, 1952
; Doise & Mugny, 1984
; Savery & Duffy, 1994
;
Baker, 1996
; Erkens, 1997
; Petraglia, 1998
; Veerman & Treasure
-
Jones, in press
).

Specific computer tools have been designed to support collaborative argument
a-
tio
n. Belvédère is a synchronous network
-
tool developed by the LRDC at the Un
i-
versity of Pittsburgh (Learning Research and Development Centre, 1996
). Individ
u-
als or groups of students of any size can use Belvédère for constructing argument
a-
tive diagrams online. The working screen of the program (see
Figure
5
) displays
private and shared windows. To communicate with a part
ner the student has a pr
i-
vate text
-
based chat
-
box, as in the CTP system, in which multi
-
line messages can be
created and sent. Messages will then be displayed, coupled with the writer's name, in
the shared chat
-
history. Adding data into the shared diagram
is constrained: students
must use the predefined set of boxes ('
hypothesis'
, '
data'
, '
unspecified')

and links
('
for'
, '
against'
, '
and'
). These are shown in the menu bar in
Figure
5
. Thickness of
links can be modified to reflect a
participant’s confidence in the information. The
thicker the line, the more certain the student is. Participants’ names can be displayed
alongside their contributions. So, students can keep track of who is responsible for
each component in the shared argum
ent diagram. An electronic coach (the 'light
NEW TECHNOLOGIES

19

bulb') is available to give help on demand. The coach gives advice on how to i
m-
prove the argument structure, based on the representation of the argumentative
stru
c-
ture in the diagram.


Figure
5

Belvédère Planning Tool

In one study with Belvédère (Veerman & Andriessen, 1997
) students working in
pairs produced conflicting stances about three different aspects of a c
onceptual d
e-
sign task. Each aspect had to be discussed in separate sessions using Belvédère.
Three argumentative diagrams had to be submitted as the final product. The average
proportion of argumentative utterances was about .90 in the dialogues and circa
.96
in the diagrams. When comparing the frequency of arguments produced in the Be
l-
védère dialogues and in the CTP chats (in a different task, see above) we found sim
i-
lar proportions, despite the provision of argumentative moves such as
'Hypoth
e
ses'
,
'Data'
,
'For'
, and
'Against'

in the task window of the Belvédère system. In co
m
pa
r-
ing different systems, structuring the interaction does not necessarily seem to
pr
o-
voke

argumentation. This may depend on task characteristics such as the instru
c
tion
to compete ve
rsus instruction to compromise (Veerman & Treasure
-
Jones, in press
).

In addition to the study of the ways computer support may trigger the frequency
of argumentation, one must study the relationship between argumen
tation and lear
n-
ing. One conclusion derived from research on collaborative learning is that during
the exploration of multiple points of view, regular checking of mutual unde
r
standing
and focus maintenance are essential for engagement in fruitful discussio
ns (Erkens,
1997
; Baker, 1994, 1996
). To support and optimise students' engagement in arg
u-
20


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

mentative dialogues for learning purposes, Computer Mediated Communication
(CMC) prov
ides interesting research opportunities. The permanence and explicitness
of text together with the possibility for time
-
delays in asynchronous text
-
based
communication provide opportunities for reflection and scrutinising information for
participants in th
e discussion as well as for educational researchers. In addition,
CMC may offer educators ways to intervene.

However, in CMC situations, the effects of both online and asynchronous tuto
r-
ing are not known very well. For example, whereas Hightower and Sayee
d’s (1995)

study on synchronous electronic discussion showed a higher degree of biased b
e
ha
v-
iour than in face
-
to
-
face discussions, indicating a strong tendency towards co
m
pr
o-
mising with the tutor, we experienced the opposite result in our own educational
practice (Veerman & Andriessen, 1996
). In an asynchronous discussion task, we
experienced a tutor

being completely ignored. However, we also found that a tutor's
intervention could disrupt a discussion at once, causing unrecoverable breakdowns
in student's negotiation processes. It seems that the relationship between mode of
communication, task enviro
nment, knowledge domain and the roles of teachers and
students are quite complex and far from understood.

In a recent experiment, Veerman, Andriessen and Kanselaar (submitted
)

an
a-
lysed synchronous discussions in terms of constructive activities.
Constructive a
c
ti
v-
ities
embody 4 types of goal
-
oriented activities in which (1) re
levant information is
added

or
explained
, and/or (2) information is reformulated and specified through
summarising
, (3) old and new knowledge is integrated and
transformed

and/or (4)
information is
judged
on strength or relevance
.
Less constructive or even

destructive
activities are information exchanges such as repetitions, off
-
task information, or
information from unrel
i
able sources.

This research example is focused on student pairs, supported by a peer
-
tutor. The
purpose of the task was to collaborative
ly solve a short discussion task in a synchr
o-
nous CMC environment. After some technical instruction a group of students (the
'tutors') were randomly assigned to two training groups aimed at the use of (1)
"structured" versus (2) "reflective" peer
-
tutoring
strategies. During these half
-
an
-
hour training sessions of the tutors, another group of students (the 'students') an
a-
lysed a written dialogue of a tutoring session, using Laurillard's (1993)

'con
vers
a-
tional framework'. In a critical review of this framework by Bostock (1996)
, this
framework is tested on consistency and is judged to be only partly accurate, thus
students could be expected to genera
te differing analyses. The first part of the a
s-
signment for the 'students', an individual preparation for later discussion, took about
10 minutes and involved labelling sentences according to the most appropriate cat
e-
gories of the conversational framework.

This manipulation was inspired by a report
by Bull and Broady (1997)

in which students comparing individually prepared sol
u-
tions to exercises became naturally engaged in extensive discussion.
In the exper
i-
mental session the 'tutors' guided pairs of 'students' during the discussion of their
individual results. These interactions were carried out electronically with each pa
r-
ticipant seated behind his or her individual computer screen. Contrary to

expect
a-
tions the frequency of constructive activities was rather low, less than 35%, and i
n-
volved only adding, explaining and evaluating. Many exchanges involved problems
NEW TECHNOLOGIES

21

with the interface (a shared word processor and a simple on
-
line chat), also the cau
se
of many focusing problems. A post
-
hoc cluster analysis revealed three groups of
subjects. The largest group of subjects, called the achievers (1), could be characte
r-
ised by a strong focus on the application of information, that is on solving the exe
r-
cis
e. These students did not produce as many constructive activities as the second
group, called the conceptual achievers (2). This second group produced more than
twice as many constructive activities as the first group. They shift back and forward
between d
iscussing the meaning of concepts and the use of concepts. The third
group (3) differs from the other two in that they discuss the meaning of the concepts
and the task strategy, while discussing the use of concepts is quite a bit lower.

Such results seem i
mportant, because they are drawn on the basis of the analysis
of quite complex interactions in which a number of essential factors have been taken
into a
c
count.

Research on electronic collaborative learning needs many more such detailed r
e-
sults in order t
o find reliable answers to questions concerning its educational vali
d
i-
ty. It seems that fruitful application of electronic tools for collaborative learning is
not only a matter of instructional arrangement, but requires better understanding of
the precise
interactions between the affordances of tools and environments, learner
characteristics and specific learning r
e
quirements.

7.

REFERENCES

Ajewole, G.A. (1991). Effects of discovery and expository instructional methods on the attitude of st
u-
dents to biology.
J
ournal of Research in Science Teac
h
ing
,
28
, 401
-
409.

Alamargot, D., & Chanquoy, L. (in preparation).
Through the models of text production.
Amsterdam:
Amsterdam University Press.

Andriessen, J., Erkens. G., Overeem, E., & Jaspers, J. (1996, September).
Usi
ng complex information in
argumentation for collaborative text production
. Paper presented at the UCIS '96 conference. Po
i-
tiers, France.

Andriessen, J.E.B., & Sandberg, J.A.C. (in press). Where is education and how about AI?
Journal of
Artificial Intellige
nce in Education.

Baker, M. (1994). A model for negotiation in teaching
-
learning dialogues.

Journal of Artificial Intell
i-
gence in Education, 5,
199
-
254.

Baker, M. (1996). Argumentation and Cognitive Change in Collaborative Problem
-
Solving Dialogues.

COAST

Research Report Number CR
-
13/96
, France.

Baker, M., & Bielaczyc, K. (1995). Missed opportunities for learning in collaborative problem
-
solving
interactions. In J. Greer (Ed.),
Proceedings of AI
-
ED 95


7
th

World Conference on Artificial Intell
i-
gence in Ed
ucation

(pp. 210
-
218). Charlottesville: Association for the Advancement of Computing in
Education (AACE).

Bartlett, F.C. (1932).
Remembering: a study in experimental and social psychology.
Cambridge: Ca
m-
bridge University Press

Beck, E.E. (1993). A survey o
f experiences of collaborative writing. In M. Sharples (Ed.),
Computer
supported collaborative writing.

(pp. 87
-
112). London: Springer.

Bereiter, C., & Scardamalia, M. (1987).
The Psychology of Written Composition.

Hillsdale, NJ: La
u-
wrence Erlbaum Associat
es.

Bostock, S.J. (1996). A critical review of Laurillard’s classification of educational media.
Instructional
Science, 24,
71
-
88.

Boxtel, C. A. M. van, Linden, J. L. van der, & Kanselaar, G. (1998). Collaborative construction of co
n-
ceptual understanding:
interaction processes and learning outcomes emerging from a concept ma
p-
ping and a poster task.
Journal of Interactive Learning Research, 8
, 341
-
361.

22


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

Boyd, A., & Rubin, A. (1996). Interactive Video: a Bridge between Motion and Math.
International
Journal of

Computers for Mathematical Learning, 1,

57
-
93.

Bull, S., & Broady, E. (1997). Spontaneous peer tutoring from sharing student models. In B. du Boulay &
R. Mizoguchi (Eds.),
Proceedings of Artificial Intelligence in Education
(pp. 143
-
150). Kobe: IOS
Press.


Carlsen, D.D., & Andre, T. (1992). Use of a microcomputer simulation and conceptual change text to
overcome students’ preconceptions about electric circuits.

Journal of Computer
-
Based Instruction
,
19
, 105
-
109.

Chan, T.W. & Baskin, A.W. (1990). Learning c
ompanion systems. In C. Frasson & G. Gaulthier (Eds.),
Intelligent Tutoring Systems: At the crossroads of artificial intelligence and education
(pp. 6
-
34).
Norwood, NJ: Ablex.

Choi, B., & Gennaro, E. (1987). The effectiveness of using computer simulated e
xperiments on junior
high school students understanding of the volume displacement concept.
Journal of Research in Sc
i-
ence Teaching
,
24
, 539
-
552.

Cognition and Technology Group at Vanderbilt. (1991). Technology and the Design of Generative Lear
n-
ing Environ
ments.
Educational Technology, May 1991
, pp. 34
-
40.

Cognition and Technology Group at Vanderbilt. (1992).
The Adventures of Jasper Woodbury: Assessment
of Instructional Outcomes.

Internal publication, pp. 1
-
19

Cognition and Technology Group at Vanderbilt.

(1993). Anchored Instruction and Situated Cognition
Revisited.
Educational Technology, March 1993
, pp. 52
-
71.

Cunningham, D.J. (1992). Assessing constructions and constructing assessments: A dialogue. In T. M.
Duffy & D. H. Jonassen (Eds.),
Constructivis
m and the technology of instruction: A conversation.

Hillsdale, NJ: Lauwrence Erlbaum Associates.

de Jong, T. (1991). Learning and instruction with computer simulations.
Education & Compu
t
ing
,

6
, 217
-
229.

de Jong, T., & van Joolingen, W.R. (1998a). Scient
ific discovery learning with computer simulations of
conceptual domains.
Review of Educational Research
,
68
, 179
-
202.

de Jong, T., Härtel, H., Swaak. J., & van Joolingen, W. (1996). Support for simulation
-
based learning; the
effects of assignments in learn
ing about transmission lines. In A. Díaz de Ilarazza Sánchez & I. Fe
r-
nández de Castro (Eds.),
Computer aided learning and instruction in science and engineering

(pp. 9
-
27). Berlin: Springer Verlag.

de Jong, T., Martin, E., Zamarro J
-
M., Esquembre, F., Swaa
k, J., & van Joolingen, W.R. (in press). The
integration of computer simulation and learning support; an example from the physics domain of co
l-
lisions.
Journal of Research in Science Teaching
.

de Jong, T., van Andel, J., Leiblum, M., & Mirande, M. (1992).
Computer assisted learning in higher
education in the Netherlands, a review of fin
d
ings.
Computers & Education
,
19
, 381
-
386.

de Jong, T., van Joolingen, W.R., Swaak, J., Veermans, K., Limbach, R., King, S., & Gureghian, D.
(1998b). Self
-
directed learning i
n simulation
-
based discovery environments.
Journal of Computer
Assisted Learning
,
14
, 235
-
246.

DeCorte, E. (1990). Learning with new information technologies in sch
ools; perspectives from the ps
y-
chology of learning and in
struction.
Journal of Computer As
sisted Learning,

69
-
84 Oxford: Pe
r-
gamon Press.

Diermanse, E. (1997).
Samenwerkend schrijven en computerondersteunde hulpmiddelen.

[Collaborative
writing and computer supported tools]. Utrecht: Internal report, dept. of Educational Sciences,
Utrecht Univer
sity.

Dillenbourg, P., & Self, J.A. (1992). A computational approach to socially distributed cognition.
Eur
o
p
e-
an Journal of Psychology of Education, 3
, 353
-
372.

Doise, W., & Mugny, G. (1984).

The social development of the intellect.

Oxford: Pergamon.

Duffy
, T.M., & Jonassen, D.H. (1991). Constructivism: new implications for instructional tec
hnology.
Educational Technology, May 1991
, 7
-
12.

Duffy, T.M., Lowyck, J., & Jonassen, D.H. (Eds.). (1992).
Designing environments for constructive
learning.

Berlin: Sp
ringer
-
Verlag.

Erkens, G. (1997).
Cooperatief probleemoplossen met computers in het onderwijs: Het modelleren van
cooperatieve dialogen voor de ontwikkeling van intelligente onderwijssystemen.

[Cooperative pro
b-
lem solving with computers in education: Model
ling of cooperative dialogues for the the design of i
n-
telligent educational systems] Ph.D. thesis, Utrecht University, the Netherlands.

NEW TECHNOLOGIES

23

Faryniarz, J.V., & Lockwood, L.G. (1992). Effectiveness of microcomputer simulations in stimulating
environmental proble
m solving by community college students.
Journal of Research in Science
Teaching
,
29
, 453
-
470.

Freudenthal, H. (1978).
Weeding and sowing. Preface to a science of mathematical education.

Dordrecht:
Reidel Publishing Company.

Gentner, D. & Stevens, A. (Eds.
). (1983).
Mental Models.

Hillsdale, NJ: Lawrence Erlbaum Associates.

Gere, A., & Stevens, R.S. (1989). The language of writing groups: How oral response shapes revision. In
S.W. Freedman (Ed.),
The acquisition of written language: Response and revision

(p
p. 85
-
105).
Norwood, NJ: Ablex.

Giroud, A. (1999). Studying argumentative text processing through collaborative writing. In J. Andrie
s-
sen & P. Coirier (Eds.),
Foundations of argumentative text processing.

Amsterdam: Amsterdam Un
i-
versity press.

Glaser, R.,
Schauble, L., Raghavan, K., & Zeitz, C. (1992). Scientific reasoning across different domains.
In E. de Corte, M. Linn, H. Mandl & L. Verschaffel (Eds.),
Computer
-
based learning environments
and prob
lem solving

(pp. 345
-
373). Berlin: Springer
-
Verlag.

Glas
ersfeld, E.V. von. (1978). Radical constructivism and Piaget’s concept of knowledge. In L.P. Steffe
& T. Wood (Eds.),
Transforming children’s mathematics education,
(pp. 30
-
38). Hillsdale, NJ: La
w-
rence Erlbaum Associates.

Grimes, P.W., & Willey, T.E. (1990
). The effectiveness of microcomputer simulations in the principles of
economics course.
Computers & Education
,
14
, 81
-
86.

Henri, F. (1995). Distance learning and computer mediated communication: Interactive, quasi
-
interactive
or monologue? In C. O'Malley
(Ed.),
Computer supported collaborative learning

(pp. 145
-
165).
NATO ASI Series, Vol. 128. Berlin: Springer
-
Verlag.

Hightower, R., & Sayeed, L. (1995). The impact of computer
-
mediated communication systems on biased
group discussion.
Computers in Human B
e
haviour, 11,

33
-
44.

Hulshof, C., De Jong, T., & Van Joolingen, W. (1999). The influence of prior knowledge on scientific
discovery learning processes in simulation
-
based science domains. Paper presented at the 1999 A
E-
RA Annual Meeting. Montreal, Canada.

J
aspers, J.G.M., Kanselaar, G., & Kok, W.A.M. (1993) Productive Language Use with IT'S ENGLISH.
Computers & Education,

21
, 67
-
75.

Jonassen, D., Davidson, M., Collins, M., Campbell, J., & Haag, B. (1995). Constructivism and computer
-
mediated communication in

distance education.
American Journal of Distance Education 9
, 7
-
26.

Kanselaar, G., & van der Linden, J.L. (1983).
Sociaal Kognitieve Komplexiteit.

[Social cognitive co
m-
plexity]. Doctoral Dissertation, Utrecht: Utrecht university.

Kanselaar, G. (1994). F
oreign Language Teaching with Multimedia. In P. Broeder & J. Murre (Eds.),
Abstract Proceedings of the Workshop on Cognitive Models in Language Acquisition. 21
-
23 April.

(pp. 29
-
30). Tilburg: Tilburg Universiteit.

Kanselaar, G., van Galen, F., Beemer, H.,
Erkens, G., & Gravemeijer, K. (1999).
Grafieken leren met de
computer (A realistic approach of learning mathematical principles in an ICT
-
learning
-
environment).
Utrecht: ISOR.

Katz, S. & Lesgold, A. (1993). The role of the tutor in Computer
-
Based Collabora
tive Learning Situ
a-
tions. In S.P. Lajoie & S.J. Derry (Eds.),
Computers as Cognitive Tools

(pp. 289
-
317).

Hillsdale, NJ:
Lawrence Erlbaum Associates.

Klahr, D., Fay, A.L., & Dunbar, K. (1993). Heuristics for scientific experimentation: A developmental
stu
dy.
Cognitive Psychology
,
25
, 111
-
146.

Kolodner, J. (1993).
Case
-
Based Reasoning.

San Mateo, CA: Morgan Kaufmann.

Kozma, R.B. (1991). Learning With Media.
Review of Educational Re
search, 61
, 2, 179
-
211.

Laurillard, D. (1993).
Rethinking university teachin
g: a framework for the effective use of educational
technology.

London: Routledge

Lavoie, D.R., & Good, R. (1988). The nature and use of predictions skills in a biological computer sim
u-
lation.
Journal of Research in Science Teach
ing
,

25
, 335
-
360.

Learning

Research and Development Centre (1996).
Advanced Cognitive Tools for Learning.
http://advlearn.lrdc.pitt.edu/belvedere/index.html
.

Pittsburgh: University of Pittsburgh.

Mitchell, A., & Po
sner, I. (1996).
Field study of collaborative writing by grade 6 students.

http://www.dgp.toronto.edu/people/alex/unpublished/Huron.html#RTFToC41
.

Nagy W.E., Herman, P.
A., & Anderson, R.C. (1985). Learning words from context.
Reading Research
Quarterly, 20
, 233
-
253.

24


KANSELAAR, DE JONG,
ANDRIESSEN & GOODYEA
R

Nagy W.E., Herman, P.A., & Anderson, R.C. (1987). Learning word meanings from context during no
r-
mal reading,
American Educational Reserach Journal, 24,

237
-
2
70.

Petraglia, J. (1998).
Reality by Design: the rhetoric and technology of authenticity in education.

Mahwah,
NJ:Lauwrence Erlbaum Associates.

Piaget, J. (1952).
The origins of intelligence
. New York: International University Press.

Piolat, A. (1990).
Ve
rs l'amélioration de la rédaction de texte.

[Towards improved text composition].
Dossier d'habilitation à diriger des recherches. Aix
-
en
-
Provence: Université de Pr
o
vence.

Reigeluth, C.M., & Schwartz, E. (1989). An in
structional theory for the design of co
mputer
-
based sim
u
l
a-
tions.
Journal of Computer
-
Based Instruction
,

16
, 1
-
10.

Rivers, R.H., & Vockell, E. (1987). Computer simulations to stimulate scientific problem solving.
Jou
r-
nal of Research in Science Teaching
,

24
, 403
-
415.

Roschelle, J., & Teasley, S.D
. (1995). Construction of shared knowledge in collaborative problem sol
v-
ing. In C. O'Malley (Ed.),
Computer
-
supported collaborative learning
. New York: Springer
-
Verlag.

Roth, W. & Roychoudhury, A. (1993b). The concept map as a tool for the collaborative c
onstruction of
knowledge: A microanalysis of high school physics students.
Journal of Research in Science Teac
h-
ing, 30,

503
-
534.

Roussey, J.Y., & Gombert, A. (1992). Ecriture en dyade d'un texte argumentatif par des enfants de huit
ans. [Writing in dyads o
f an argumentative text by 8 year old children].
Archives de Psychologie
,
297
-
315.

Salomon, G. (1993). On the nature of pedagogic computer tools: The case of the writing partner. In L
a-
joie, S.P. & Derry, S.J. (Eds.),
Computers as Cognitive Tools,

(pp. 289
-
317).

Hillsdale, N.J.: La
w-
rence Erlbaum Associates.

Salomon, G. (1997).
Novel Constructivist Learning Environments and Novel Technologies: Some Issues
to Be Concerned With
. Invited Key note Address presented at the EARLI meeting, Athens, August
1997

Savery
, J.R., & Duffy, T.M. (1994). Problem based learning: An instructional model and its constru
c-
ti
v
istic framework
. Educational Technology (
august

1994).

Scanlon, E. (1998). How Beginning students Use Graphs of Motion. In M.W. van Someren, P. Reimann,
H.P. B
oshuizen & T. de Jong (Eds.),
Learning with Multiple Representations
(pp. 67
-
86). Amste
r-
dam: Pergamon, Elsevier Science Ltd.

Scardamalia, M., Bereiter, C., & Lamon, M. (1994). The CSILE
-
project: Trying to bring the classroom
into World 3. In K. McGilly (Ed
.),
Classroom lessons:Integrating cognitive theory and classroom
practice
(pp. 202
-
229). Cambridge: MIT Press.

Scardamalia, M., & Bereiter, C. (1996). Adaptation and Understanding. In S. Vosniadou, E. DeCorte, R.
Glaser & H. Mandl (Eds.),
International Per
spectives on the design of Technology
-
Supported Lear
n-
ing Environments

(pp. 149
-
183). Mahwah, NJ: Lawrence Earlbaum Associates.

Schank, R.C., & Cleary, C. (1995).
Engines for education.

Hillsdale, NJ: Lauwrence Erlbaum Associates.

Schauble, L., Glaser, R.,
Raghavan, K., & Reiner, M. (1991a). Causal models and experimentation strat
e-
gies in scientific reasoning.
The Journal of the Learning Sciences, 1,

201
-
239.

Schauble, L., Klopfer, L., & Raghavan, K. (1991b). Students’ transitions from an engineering to a sc
ience
model of experimentation.
Journal of Research in Sc
i
ence Teaching
,
28
, 859
-
882.

Sharples, M., & O'Malley, C. (1992). A framework for the design of a writer’s assistant. In P. Holt & P.
Williams (Eds.),
Computers and writing: State of the art
(pp. 319
-
336). Wo
r
chester: Billing & Sons.

Sharples, M. (Ed.), (1993).
Computer supported collaborative writing.

London: Springer
-
Verlag.

Shute, V.J., & Glaser, R. (1990). A large
-
scale evaluation of an intelligent discovery world: Smithto
wn.
Interactive Learnin
g Environments
,

1
, 51
-
77.

Simmons, P.E., & Lunetta, V.N. (1993). Problem
-
solving behaviors during a genetics computer simul
a-
tion: beyond the expert/novice dichotomy.
Journal of Research in Science Teaching
,
30
, 153
-
173.

Sinclair, J. (Ed.). (1987).
Collins
COBUILD English Language Dictionary
. London and Glasgow: Collins
Publishers.

Suthers, D., Weiner, A., Connelly, J., & Paolucci, M. (1995).
GroupWare for developing critical discu
s-
sion skills.

University of Pittsburgh, Pittsburgh <
http://www.pitt.edu/~suthers/belvedere
>.

Swaak, J., & de Jong, T. de (1996). Measuring intuitive knowledge in science: the development of the
what
-
if test.
Studies in Educational Evaluation
,
22
, 341
-
362.

Swaak, J., Blokhuis,
V., Gutierrez, M., & López, M. (1997).
Evaluating simulation discovery enviro
n-
ments with learners.

SERVIVE

deliverable D32. Enschede: University of Twente, Faculty of Educ
a-
tional Science and Technology, OCTO.

NEW TECHNOLOGIES

25

Swaak, J., van Joolingen, W.R., & de Jong, T. (
1998). Supporting simulation
-
based learning; the effects
of model progression and assignments on definitional and intuitive knowledge.
Learning and I
n
stru
c-
tion
,
8
, 235
-
253
.

Tabachneck
-
Schijf, H.J.M., & Simon, H.A. (1996). Alternative representations of ins
tructional material.
In D. Peterson (Ed.),
Forms of Representation
, pp 28
-
46. Wiltshire: Cromwell Press.

Tabachnek
-
Schijf, H.J.M., & Simon, H.A. (1998). One Person, Multpile Representations: An Analyses of
a Simple, Realistic Multiple Representation Learn
ing Task. In M.W. van Someren, P. Reimann, H.P.
Boshuizen, & T. de Jong (Eds.),
Learning with Multiple Representations,
(pp. 197
-
236). Amsterdam:
Pergamon, Elsevier Science Ltd.

Torrance, M. (1996). Is writing expertise like other kinds of expertise? In G.

Rijlaarsdam, H. Van den
Bergh, & M. Couzijn (Eds.),
Theories, Models and Methodology in writing Research.

Amsterdam:
Amsterdam University Press.

Towne, D.M., de Jong, T., & Spada, H. (Eds.). (1993).
Simulation
-
based experiential learning
. Berlin:
Springer
-
Verlag.

Van Boxtel, C. A. M., Linden, J. L. van der, & Kanselaar, G. (1998). Collaborative construction of co
n-
ceptual understanding: interaction processes and learning outcomes emerging from a concept ma
p-
ping and a poster task.
Journal of Interactive Lear
ning Research, 8
, 341
-
361.

Van Emmerik, M.L., van Rooij, J.C.G.M., & de Jong, T. (in preparation).
Differences in instruction for a
simulated driving task.

TNO report

Veerman, A.L., & Andriessen, J.E.B. (1996, December).
Argumentatie tijdens zwak gestructu
reerd pro
b-
leemoplossen

[Argumentation during solving ill
-
structured problems]. Paper presented at the
VIOT

congress, Utrecht.

Veerman, A.L., & Andriessen, J.E.B. (1997, September).
Academic learning & writing through the use of
educatianal technology.
Pap
er presented at the conference on Learning & Teaching Argumentation.
Middlesex University, London.

Veerman, A.L., & Andriessen, J.E.B. (1999),
Enhancing learning through electronic discussion.
Paper
presented at Call99, London.

Veerman, A.L., & Andriessen,

J.E.B., & Kanselaar, G. (submitted). Enhancing learning through collab
o-
rative argumentation.

Veerman, A.L., & Treasure
-
Jones, T. (in press). Software for problem solving through collaborative
argumentation. In J.E.B. Andriessen & P Coirier (Eds.),
Foundat
ions of argumentatieve text pr
o-
c
essing.

Amsterdam: Amsterdam University Press,


Vygotsky, L. (1978).
Mind and society.

Cambridge, MA:Harvard University.


Draft, to appear