Clark Quinn (cnquinn@knowledgeu.com)
Wed, 17 Feb 1999 17:49:04 -0800
Date: Wed, 17 Feb 1999 17:49:04 -0800 From: "Clark Quinn" <cnquinn@knowledgeu.com> Subject: Re: Some questions for your to ponder
List address to send message to everyone: ifets-discuss@LISTSERV.READADP.COM
Details of current discussion: http://zeus.gmd.de/ifets/discuss.html
---------------------------------------------------------------------------
Dear David, your query triggered some thoughts:
>1. I am unaware of a general purpose adaptive authoring system that
>enables us to author materials that will automatically adapt to individual
>differences (learning styles, aptitudes, etc.). Are you aware of such a
>system? Who is working on the development of such a system? Where can we
>find additional information?
I don't know of any such system that is developed. I, too, would like to
hear.
>2. I suspect that there are some individual instructional programs
>available that do adapt to individual differences (learner styles,
>aptitudes, etc.). What are they? Where can we find additional information
>on these systems? How successful are they?
I have seen one that adapted on the basis of an initial assessment of
whether the learner was abstract or concrete, and sequential or random. A
bit coarse, but a first start. Unfortunately, it was a privileged
presentation, and I cannot share it.
Adaptivity on other criteria can be seen here:
http://www.psychologie.uni-trier.de:8000/projects/ELM/elmart.html
It's adapting on what the learner demonstrates as knowledge, not on learner
styles, however. This is typical of intelligent tutoring systems.
Another approach, by the way, is that of Magda Ritzen, who had students do
a learning style inventory, and then merely point out to them what the
consequences of those learning styles are, rather than be directive about
appropriate learning styles
>Some general questions to which I would appreciate your thoughts:
>
>3. Should we, can we build an adaptive authoring system? What would it
>look like? What individual difference variables should be included? Which
>should not be included? Should the individual difference variables be
>built-into the system? Or should the system be general and allow the
>author to include those individual difference variables that he or she
>thinks are important? What are the advantages and disadvantages of having
>the individual difference variables built-in?
Well, the advantage, done right, would be a system individualized,
producing a personally optimal (!) learning interaction (or transaction
:-). What would it look like? My proposal is that it would be suggestive
of activities, but the full suite would be available. Thus, there might be
a top down content presentation, practice activities, stories, case
studies, and more.
My question is, what variables are there, and what would we do differently
on the basis of them? For example, in abstract versus concrete, you might
allow the former to go to a conceptual presentation, while the concretes
might be given some problems. Similarly, the random/global might get a
menu of problems to look at, while the sequential might get a fixed path
(shudder).
However, as I indicated in my last message, the picture gets complicated.
There is some evidence that field independents prefer less structure than
field dependents. Yet this dimension seems to correlate alot with
abstract/concrete. However, the prescription for the latter is whether to
give content first or problems. These seem incommensurate. And, as I
asked before, do any of these styles stay fairly reliable? My
introspection (OK, not *that* scientific, but all I have) is that I change
based upon the media (text versus graphic) and domain, and....
It might also change in finer grain educationally, someone might be global
when reading content, but sequential in problems.
Consequently, I would reject being prescriptive for the reasons I indicated
in my last message: people change learning styles too frequently. I'd
rather a site be suggestive.
In general, we adapt based upon some knowledge. What knowledge CAN we
have? We can have knowledge a priori, and we can build knowledge during
the interaction. A priori, we might learn about: the individual's learning
style, preferences, experience, job, etc. On the basis of their
exploration/interaction (and the design of the system), we can learn about
their: knowledge of the domain of instruction, preferences in resources,
systematicity. Peter Brusilovsky has an interesting summary of adaptive
systems on the Web, that's relevant beyond the web at least in the
categorization.
http://www-aml.cs.umass.edu/~stern/webits/itsworkshop/brusilovsky.html
>4. Learner control of many of the instructional strategy variables is one
>way to enable a system to adapt to individual differences. In this type of
>a system the learner adapts the system to themselves.
What has been called adaptability, as opposed to adaptivity. Adaptability
means the system can be configured for user preferences. Adaptivity means
the system adapts to what is done.
>We see this to some
>extent in many application programs where the learner can customize menus
>and functions to their own needs. Is this a preferred approach to
>instructional systems? If so what strategy variable should the student be
>able to adapt? What are the advantages and disadvantages of such a system?
> Has any one built such a system? Where can we get additional information?
> Are they successful?
In general, my preference is for the system to be adaptable, not adaptive.
If adaptive, it should be suggestive, not prescriptive. There should be
default options for the timid/unexperienced (with the system) learner, and
flexibility for the learner.
I don't know of systematic studies. So, on what basis do I make these
claims? Some on principles like the Scandinavian ones for system design,
and empowering the learner. Others on experience with other Information
Technology systems. We want the challenge to be in the learning, not in
the interface! I'd sure like to have a better basis. I see some
interesting research PhD's in these questions and others.
I hope others will take up David's challenge, -- Clark
-- Clark Quinn Knowledge Universe Interactive Studio (510) 768-2408 cnquinn@knowledgeu.com--------------------------------------------------------- Forum website: http://zeus.gmd.de/ifets/ Forum's contact person: kinshuk@ieee.org Info on Join/Leave List: http://zeus.gmd.de/ifets/maillist.html ---------------------------------------------------------
This archive was generated by hypermail 2.0b3 on Thu 18 Feb 1999 - 02:52:22 MET