support for e-learning

almondpitterpatterIA et Robotique

23 févr. 2014 (il y a 3 années et 6 mois)

89 vue(s)

1

the future of computer
support for e
-
learning


Gerry Stahl

Drexel University, Philadelphia, USA

Gerry.Stahl@Drexel.edu


2


A research agenda for the future based on
past experiences: 4 example attempts to
support e
-
learning led to the need to:





Focus on supporting
group interaction





Develop a theory of collaboration





Extend & revise
methodologies

of
software design (software as artifact &
world
-
opening)





Envision the potential of e
-
learning in the
future: supporting both group meaning
-
making & individual interpretation

3

Overview


I.

Attempts at computer support for e
-
learning



II.

A shift of scientific paradigms



III.

Implications for methodologies of designing
computer support for e
-
learning



IV.

A potential future for e
-
learning



V.

Open questions and challenges

4

Returning what I learned from Germany


My PhD in philosophy:


1967/68 Heidelberg


Heidegger, Gadamer


1971/73 Frankfurt


Hegel, Marx, Adorno, Habermas


Importance of theory & its historical development


My PhD & research in computer science:


1989
-
01 Colorado


Gerhard Fischer L
3
D, visitors, AI


Situated cognition & support for communities of practice



Visiting scientist:


2001/02 GMD/Fraunhofer FIT


W. Prinz, W. Appelt, EU
ITCOLE (BSCL software)


Collaborative design to support CSCW in practice

5

Part I. Attempts at computer
support for e
-
learning


WebGuide
: interpretive perspectives on meaning


collaboration with Th. Herrmann (Dortmund), NOAA, teacher


SimRocket
: making learning visible in discourse


collaboration with communication, psychology & anthropology
profs, teachers & students


BSCL
: negotiating system design


EU project with designers
& pedagogy profs in Germany, Spain, Finland, Greece, NL,
Italy


Virtual Math Teams
: creating collaborative worlds


collaboration with online Math Forum, international CSCL

6

WebGuide
: interpretive
perspectives on meaning


Threaded discussion system that displayed
discussion items in different selections


Personal, team and class perspectives:


The class only included shared, agreed upon
ideas


The team perspective included ideas that the
team was working on as well as class ideas


The personal inherited class & team ideas, but
could edit, add to and re
-
organize them:
personal interpretation of shared meaning

7

Supporting collaborative interactions:
divergent & convergent group thinking


Gulch class perspective

Mine owner

P8

Mines
comp

Gulch class comparison

Environmental

Landown
er

Government

P9

P7

P6

P5

Blake

P3

P1

P2

P10

P11

P12

Government
comparison

Landowner
comp

Environ

comp

Mentor

Teacher

Student

Student
comp

Mentor
comp

Class
perspective

Team
perspective

Personal
perspective

Team
comparison

Class
comparison

8

SimRocket
: making learning visible
in discourse


Simulation of rocket launches for 8 different rockets
with different characteristics


Student task was to determine best characteristics
through experimental analysis


Study of digital video of student interactions
revealed how the group solved its task and
collaboratively learned practices that none of the
students knew


9

Discourse about
SimRocket


1:22:05 Brent This one’s different


1:22:06 Jamie Yeah, but it has same no…


1:22:08 Chuck Pointy nose cone


1:22:09 Steven Oh, yeah


1:22:10 Chuck But it’s not the same engine


1:22:11 Jamie Yeah, it is,


1:22:12 Brent Yes it is,


1:22:13 Jamie Compare two n one


1:22:13 Brent Number two


1:22:14 Chuck I know.


1:22:15 Jamie Are the same


1:22:16 Chuck
Oh


1:22:17 Brent It’s the same
en
gine.


1:22:18 Jamie So if you compare two n
one
,


1:22:19 Chuck Oh yeah, I see, I see, I see



From SimRocket transcript

10

BSCL
: negotiating system design


“Basic System for Collaborative Learning”
is based on BSCW


Includes threaded discussion, chat,
whiteboard in areas for personal, team and
class knowledge
-
building


Supports team negotiation of ideas;
promotes agreed to ideas from team to class
perspective


11

BSCL


Teams of students at Drexel used BSCL, analyzed
logs of their usage, designed functionality to
support problematic interactions, negotiated group
design proposals


12

Virtual Math Teams
: creating
collaborative worlds


New research project at Drexel University


Extend Math Forum (
www.mathforum.org
) online
“problem of the week”


Students who visit the site are invited into small teams
to solve math problems
collaboratively


Structured, mentored problems are done in e
-
learning
environment


Logs are analyzed to study group learning


Creates virtual teams from a 1,000,000 person global
community to collaborate in a new environment doing
a new activity

13

Results of attempts to support e
-
learning


WebGuide
: knowledge building goes through various
processes, like divergent brainstorming & convergent
negotiation;
need to support collaborative interactions


SimRocket
: knowledge is constructed by the group


not just transferred from individual to individual;
users
must learn the social practices embedded in their
software


BSCL
: software design requires many iterations of
trial in group usage;
users must learn the meanings &
affordances embedded in their software’s design


Virtual Math Teams
: new opportunities for
educational group interaction can be designed with
software that opens new shared worlds

14

Part II. A shift of
scientific paradigms


From


AI in Ed to


Augmented cognition, communication (ICT)


From


Human <
-
> computer (HCI) to


Human <
-
> human via computer (CSCW)


From


Individual psychology (pre/post tests; stats)


Group interaction (qualitative communication patterns)


From


Computer science as engineering


Interaction design, situated cognition, activity theory

15

Problems with the
focus on individuals


Computer science problem: software is designed
for individuals, not group interactions


Education problem: learning is conceived as in
individual minds, not social interaction


Philosophy problem: focus on individual thinker,
not social actor


Political problem: George W. Bush as cowboy
entrepreneur: an “ideology of individualism,” not
the EU “spirit of collaboration”

16

Need for a theory of collaboration


Software development experiences


Shifts in the academic field


Other problems with individual focus




Research on collaboration theory




Back to roots in German philosophy

17

Philosophy of the individual


Greeks stressed role of individual
(e.g., Odysseus


Adorno & Horkheimer)


Descartes stressed individual mind


German philosophy critiqued Descartes:


Kant: “Copernican revolution”


Hegel: social history of mind


Marx: social basis of ideologies


Wittgenstein: meaning in use, not in mental
propositions


Heidegger: tacit understanding, artifacts, situated
-
ness,
Mit
-
da
-
sein, Ereignis


Habermas: communicative action

18

Philosophies of the group


Vygotsky: physical, symbolic, cognitive
artifacts from shared culture


Sachs: conversation analysis


Garfinkel: ethnomethodology


Lave & Wenger: situated learning & social
practice & communities of practice


focus on: tacit practices, discourse, shared
culture, non
-
mentalist

19

Individualist & social philosophies
& theories of learning

Activity Theory

social theories

Husserl

Hegel

(1807)

Kant

(1787)

Marx

(1867)

Heidegger

(1927)

Schutz

Piaget

Vygotsky

(1934)

Habermas

Situated
Cognition

Constructivism

Evidence
-
Based
Instructionism

Ethno
-
methodology

Communicative
Action

Conversation
Analysis

Descartes (1633)

Rationalism

Empiricism

Cognitivist

Instructionism

Wittgenstein








Wittgenstein

Social Practice

anthropology

individual

theories

20

The philosophy of Martin Heidegger


Systematic re
-
thinking of human
understanding & activity as a
critique of Descartes & the
Western tradition


Has its own political & theoretical
problems


But, provides the root of many
new theories


And, has untapped fertile
approaches

21

Heidegger on artifacts


As applied to HCI by Winograd/Flores, Dreyfus,
Suchman, Ehn & Budde/Z
ü
llighoven,

Heidegger
distinguished things that are
explicitly present

to
us from artifacts that are
tacitly at hand
.


Explicit knowledge of things is only possible on
the basis of a wealth of tacit knowing, skills and
practices.


To know is to have a practical ability to do
something in certain situations, not primarily to
have stored propositions or mental representations
that correspond to things present in the world.

22

Heidegger: “Situation”


“Zum Beispiel mit diesem Zuhandenen
[“artifact”], das wir deshalb Hammer nennen,
hat es die Bewandtnis beim H
ä
mmern, mit
diesem hat es seine Bewandtnis bei Befestigung,
mit dieser bei Schutz gegen Unwetter; dieser
>ist< um
-
willen des Unterkommens des
Daseins, das heisst, um einer M
ö
glichkeit seines
Seins willen. Welche Bewandtnis es mit einem
Zuhandenen hat, das ist je aus der
Bewandtnisganzheit [“Situation”]
vorgezeichnet.”
Heidegger,
Sein und Zeit
. S. 84

23

Heidegger: “Mitsein”


“Auf dem Grunde dieses mithaften In
-
der
-
Welt
-
seins
ist die Welt je schon immer die, die ich mit den
Anderen teile. Die Welt des Daseins ist Mitwelt. Das
In
-
Sein ist Mitsein mit Anderen.”
SuZ
§
26


At this point in his analysis, Heidegger overcomes
the tradition of individualistic philosophy and can
analyze situated meaning and language as based in
the community

24

Heidegger on Being
-
with


But in the next section of
Sein und Zeit
, Heidegger
rejects the social basis of human being in favor of
an “authentic” stance of the individual toward his
own finitude as the basis of meaning.


Adorno, in his
Jargon of Authenticity
, tied this
move to a politically conservative ideology.


One can see this as a source of Heidegger’s
political problems, as well as his philosophical
problem of not understanding the social basis of
phenomena like language and history.

25

Heidegger: “Welter
ö
ffnung”


“Van Goghs Gem
ä
lde ist die
Er
ö
ffnung dessen, was das
Zeug, das Paar bauernschuhe,
in Wahrheit
ist
. Dieses
Seiende tritt in die
Unverborgenheit seines Seins
heraus. . . . Im Werk ist, wenn
da eine Er
ö
ffnung des
Seienden geschieht, in dem
was und wie es its, ein
Geschehen der Wahrheit am
Werk.”
Martin Heidegger, Der
Unsprung des Kunstwerkes in
Holzwege

(Frankfurt: Klostermann, 1963), S. 25
.


26

Heidegger on opening shared worlds


“Sculpture: an embodying
bringing
-
into
-
work of places and
with this an opening up of realms
of possible living for men, of
possible persisting for the things
which surround and concern
men.”
Martin Heidegger,
Raum und Kunst

(St. Gallen: Erker Verlag), 1969), S. 13


27

Vygotsky applies German
philosophy to learning


We use tools (physical artifacts) to do things in the
world. They mediate & transform our tasks


We use words (and other symbolic artifacts) to
interact with others and to control ourselves


We learn to use these artifacts in interaction with
other people.


We often internalize these artifacts for use silently
to ourselves, developing sets of cognitive artifacts


which supplement, vastly extend and transform
our biological mental skills like memory, attention


Learning takes place primarily through interaction
and discourse

28

The focus on interaction
in social science


Harvey Sachs,
Lectures on Conversation
, interprets
interaction without speculating on participant’s
thoughts, feelings, motivation


Harold Garfinkel,
Studies in Ethnomethodology
,
analyzes unique events as community members
interactively accomplishing tasks by negotiating
methods


Lave & Wenger describe learning as increasing
one’s centrality in a community of practice


Such approaches focus on tacit skills, social
practices, discourse and shared culture rather than
explicit propositional knowledge or mental
representations

29

Unit of analysis

Individual

Collaborative group

Meaning
-
making

Meaning is in
individual minds

Meaning is shared,
social, cultural

Knowledge
-
building

Explicit propositions
are primary

Tacit practices are
foundational

Communication
model

Transmission from
mind to mind

Interactive
construction

Education model

Transmission of
propositional
knowledge from
expert teacher to blank
slate student

Co
-
construction in
discourse

30

Interpretation as synthesis of
individual and group


Artifacts (linguistic, physical, practices) embody, store,
transmit shared meanings


We are born into a meaningful world of language,
artifacts, social practices, habits, culture


which we must
learn to understand (Heidegger hermeneutics)


We interpret meanings based on our pre
-
understanding
(Vygotsky “Zone of Proximal Development”)


Our activity is situated in activity context, shared world,
personal perspectives


Meanings

are shared, in
-
the
-
world, socially created in
collaboration; individuals
interpret

the meanings from
their personal perspectives, tacit pre
-
understanding


G. Stahl, Meaning and Interpretation in Collaboration, CSCL ’03.

31

Part III. Implications for
methodologies of designing
computer support for e
-
learning


Need new methods corresponding to a focus
on group interactions rather than individual
mental representations:


Theoretical framework, theory of collaboration


Analytic method, to analyze support needs


Design approach, to develop software


Evaluation technique, to assess software &
supported interactions

32

Theoretical framework


Need a
theory

of collaboration


Group as primary
unit of analysis


Avoid
mentalist assumptions
, hypotheses,
speculation about motivations & plans


Central role of
artifacts

(tools, language, software,
cognitive skills)


Mind as social construct



developed in interaction
by internalizing tools and practices


WebGuide

supports group interactions with
individual and group perspectives, allowing for a
hierarchy of groups & sub
-
groups. Also provides
comparison & negotiation perspectives

33

Analytic method


Need a methodology for analyzing the processes
of e
-
learning that need to be supported.


Conversation analysis of logs


Analysis of interaction structures and common
patterns


Break
-
down & problems reveal interaction needs


Micro
-
analysis of content of logs and digital video
can reveal learning as it takes place in a group


SimRocket

usage was analyzed with conversation
analysis of digital video, showing how a group
repaired divergent interpretations and achieved
basic scientific practices as a group

34

Design approach


Need an approach to designing e
-
learning software.


Instead of viewing e
-
learning software as a neutral transmission
service for explicit knowledge:


Software as artifact


Embodies meaning & designed practices


Requires learning to use and interpret


Transforms practices


Software as medium


Limits types of content


Enables certain types of interactions


Software as opening worlds


Creates new opportunities


Brings people together in new communities


Defines new social practices


Virtual Math Teams

is being designed by observing student
teams interacting F2F and online, under different conditions of
group composition and software support. Break
-
downs of
interaction will indicate what functionality is needed. Question
is: how to open & structure new worlds for interaction; what
kinds of breakdowns are likely

35

Evaluation technique


Need methodology for evaluating the success of e
-
learning
software in supporting effective group knowledge
-
building.


Lab studies are very limited


focus on isolated individual out
of learning situation


Quantitative studies miss nuances & content of learning
interactions


Need to design software, tasks, content, groups, settings to
produce learning interactions (design studies)


Unique case studies are not replicable, but also not anecdotal;
they reveal generalizable patterns and learning possibilities


BSCL

has been evaluated in European schools using
interviews without significant findings. Evaluation by teams of
HCI students using it in design studies at Drexel found many
weaknesses and proposed many improvements

36

Individualist
approach

Collaborative
approach

Software
example

Theoretical
framework

Psychological
model

Collaborative model of group
cognition; situation
-
, artifact
-

& group
-
centric

WebGuide

Analytic
method

Algorithmic,
analytic

Discourse analysis & analyze
practices (not mental reps);
make knowledge
-
building
visible; unique cases;
structures of interaction;
patterns

SimRocket

Design
approach

Specifications

Shared
-
world creating; media
& environments for human
social interaction; software as
artifact

Virtual
Math
Teams

Evaluation
technique

Quantitative
laboratory
studies

Reflective discourse; design
studies

BSCL

37

Part IV. A potential future of e
-
learning


Find & form collaborations with special
people anywhere, anytime


Form non
-
hierarchical, non
-
moderated
collaborations


Form global networks of collaborations


Build communities with micro
-
structure


Invent new ways of learning & new tools to
support collaboration & learning

38

e
-
learning & cognitive development


Another quantum jump in artifact
-
cognition:


Physical artifacts: tools


Symbolic artifacts: language


Cognitive artifacts: internalization


Persistent artifacts: inscription (writing,
drawing)


Ubiquitous artifacts: mechanical reproduction


Digital artifacts: computers


Collaborative artifacts: networked computers &
e
-
learning

39

Theory


We can build on existing
socio
-
cultural theory


There is a rich socio
-
cultural theory tradition with
a continuous history (in Germany) of over 200
years (at least)


This tradition has strong roots in the mainstreams
of 20
th

century philosophy & theory: Marx,
Heidegger, Wittgenstein


But, we need to
apply this theory to collaboration,
e
-
learning, software design


40

Analysis


We can analyze the nature of collaboration using a
strong
definition of “
collaboration:



group meaning
-
making

(not just incidental interaction of
individual meaning makers)


Meaning is constructed in discourse by groups; analyze
meaning
-
making at group unit of analysis as in SimRocket
transcript


Meaning must be interpreted by involved individuals from
their situated perspectives


Software should support these meaning
-
making
interactions


Collaborative learning is foundational, not secondary

41

Design


In order to design for collaborative
interactions:


Design software as a
mediating artifact



not an
explicit, visible interface


Or as a
medium of communication

&
interaction


Or as the opening of a new
shared world

42

Evaluation


Design so new learners will
learn the
meanings and practices embodied

in the
software


This may mean designing more than
software: content, use scenarios, mini
-
curricula, social practices, guidance for
usage, etc.

43

Part V. Open questions & challenges


e
-
Learning


in order to get started


had to use the
ideas, tools & approaches inherited from an era
focused on the individual


Now it is time to rethink our theory, pedagogy,
analytic methods & technologies with a focus on
groups & collaboration


A conference in Germany is a fitting starting point
to move away from the philosophy of individualism
(something harder to do in America these days)


I have just tried to pose questions and a research
agenda, not to provide complete arguments, theories
or methodologies


that is something I hope we will
work on collaboratively in the future

44


Should we
define “collaboration
” as joint
meaning
-
making by a group or community?


A
model of collaboration
; typical processes?


Relation of
individual thought to social
meaning
-
making
:


Internalization (Vygotsky)?


Personal interpretation?


How can we reach a dialectical synthesis
(Aufhebung) of the philosophies of individual
cognition & group interaction?


We need a
theory of collaboration

& pedagogy
of collaborative learning

Theory questions

45

Analysis questions


Can
conversation analysis

be made into a
technique for practical e
-
learning design?


How to gather
requirements

when users are
novices, virtual, groups, not
-
yet
-
identified?


Are requirements
unique

to situations,
activities, communities?

46

Design questions


Use of lab tests, measurements versus
situated usage

by groups?


Software as
artifact

with affordances?


Software as
medium

for communication?


Software as
opening worlds
?


Participatory

design?


Design lifecycle, design by
evolution
?

47

Evaluation questions


Combining
quantitative & qualitative

results? Do
quantitative point to issues for investigation or
qualitative descriptions point to hypotheses?


Generalizability of
case studies
? Analysis of
interaction structures in unique cases?


Assessment of learning



by group or by
individuals? Over time periods?


What forms of learning are
appropriate for e
-
learning
? For collaborative learning? For
individual instruction?

48

Concluding questions for
future e
-
learning software


How can we support collaborative
interactions within groups learning
together?



How can software open up new worlds of
collaboration and learning?



How will new users discover the meanings
and practices embodied in software?

49

E
-
learning
about

e
-
learning


How can we use e
-
learning technology to
answer these questions?



This conference may be an important step!



What can be done to foster collaboration
between conferences?

50



Gerry Stahl

Drexel University, Philadelphia, USA

Gerry.Stahl@Drexel.edu



my papers on e
-
learning:

http://www.cis.drexel.edu/faculty/
gerry/cscl/papers


51

Heidegger on “situation”


“For example, the artifact at hand which
we call a hammer has to do with
hammering, the hammering has to do with
fastening something, fastening has to do
with protection against bad weather. What
significance artifacts have is prefigured in
terms of the
situation

as a totality of
relationships of significance.”
Heidegger,
Sein
und Zeit
.

52

Heidegger on Being
-
with


“On the basis of this being
-
in
-
the
-
world with others, the
world is always the world which I share with others. The
world is always the shared world. Being in the world is
being there with others.”
SuZ
§
26
.


At this point in his analysis, Heidegger overcomes the
tradition of individualistic philosophy and can analyze
situated meaning and language as based in the community.

53

Heidegger on opening shared worlds


“Van Gogh’s painting is an
opening
-
up of that which the
artifact, the pair of farmer’s
boots, in truth is. This being
moves into the unconcealment
of its being. . . . There is a
happening of truth at work in
the work, if an opening
-
up of
the being takes place there
into that which and that how it
is.”
Martin Heidegger, Der Unsprung
des Kunstwerkes in
Holzwege

(Frankfurt: Klostermann, 1963), S. 25
.