[ifets] A MODEL FOR THE FUTURE OF LEARNING

About this list Date view Thread view Subject view Author view

Alfred Bork (bork@binky.ics.uci.edu)
Wed, 28 Oct 1998 08:39:14 +0100


From: "Alfred Bork" <bork@binky.ics.uci.edu>
Subject: [ifets] A MODEL FOR THE FUTURE OF LEARNING 
Date: Wed, 28 Oct 1998 08:39:14 +0100

A MODEL FOR THE FUTURE OF LEARNING

Alfred Bork
Information and Computer Science
University of California
Irvine California 92697-3425
Bork@UCI.edu

August 26, 1998

What do we want education and learning to look like in the future? We
begin with a vision, essential for wise planning. Then we discuss a
model of learning adequate to reach this vision.

The Vision

Everyone Learns

We require in our vision that EVERYONE LEARNS. We mean everyone
on earth, not a single country or region. We want in the words of
Theodore Hesburgh global educational opportunities for every human
being on earth, including not only the traditional areas but also
problem-solving, critical thinking, and creativity .

Everyone learning means from birth to death. It is no longer
acceptable for learning to take place only at school or university.
Learning should be lifelong in the modern world.

Everyone learning means that everyone learns to the mastery
level. Partial learning, currently with most students, is not
acceptable. There is reason to believe that everyone can learn
everything. Interaction is the key issue; for students to master all
subject matter, it critical to determine student problems so we
can give explicit assistance.

Learning is Enjoyable

Next, learning should be ENJOYABLE. For people to learn they must find
pleasure in learning, encouraging lifelong learning.

Learning is Affordable

The last aspect of our vision is that learning must be AFFORDABLE,
for individuals and governments. Often learning is too
expensive for everyone to learn, even in wealthy countries.

This vision, important for our future, will not be easy to
attain. Our current situation is far from this. Modern technology makes
it possible to realize this vision.

Next, we discuss models.

The Classroom-Teacher Model

This model is 2500 years old, from the golden age of Greece.
It comes to mind when someone speaks of learning, since most of us have
learned this way. The classroom is a small box, or often in
universities a large box, with many people in it. A teacher stands in
front and mostly talks. Most schools and universities work in this
fashion, much of the time.

Other means of learning may be employed, such as video,
computers, and cooperative learning groups. These are usually add
on for learning and expense. Primarily the group
proceeds with the teacher talking and students listening or taking
notes. There is little individualization , and many do not enjoy
learning in this way.

Education has traditionally been defined in physical terms.
We go to school - a college implies a campus. This is no longer true. For some time educational programs have been delivered
outside the classroom

Francis Dummer Fisher
Higher Education Circa 2005
Change January / February 1987

It is gross superstition to suppose that knowledge can be obtained
only by going to schools and colleges. The world produced
brilliant students before schools and colleges came into being.
There is nothing so ennobling or lasting as self study.

Gandhi
Wit and Wisdom of Gandhi, Nehru, Tagore
New book society of India, 1970

A few exceptional teachers can make the model work very well but
there are few such teachers and the number is growing smaller. Schools
and universities do not treat these teachers kindly and they often leave.

The classrom-teacher model is not equal to the needs of our world. We
must look at other models.

the dream [universal education] is impossible if education
is visualized as it has always existed in recent centuries:
a schoolroom with a teacher and students

Theodore Hesburgh
Looking Forward, ed John Templeton

The Personal Learning Model

The second model is the personal learning or independent
study model. There is no school and no teacher. The primary
responsibility for learning rests with the student, a child or adult
of any age. It is through the personal learning model that
we can realize our vision. Far more study is needed to implement this
model.

We have done personal study much of our lives. Before schools much
learning occurs, including learning a language. At college and schools,
we have homework. After formal education personal study is common.

There are many variants. An interesting way of classifying
these is by the degree and quality of the student interaction with the
learning material and with other people. Here are some possibilities.

Private Study

The first variant of this model is private study from books or video.
The student works alone without interaction with the learning material
and other people.

This is a common mode for homework, and common for individuals throughout
life. A person who wishes to learn something about Japanese cooking might
go to the library, check-out books, read them, and try the recipes.

Private Study and Cooperative Learning Groups

The second version is a minor difference from the first. Several students
work together, talking with each other. Getting help in learning from a
friend is an example.

A problem of learning with these modes is that motivation must
be internal to the student. A student may quickly lose interest.
Even more important, the student may not receive help for the learning
problems encountered. The student may not understand the problems
encountered, and so cannot ask the right question to other students in
the group.

Traditional Distance Learning

Another type of private study is traditional distance learning.
Correspondence courses by print or television are examples.
Sizable amounts of material may be involved and credit usually desired
perhaps toward a degree.

The hallmark such learning is the United Kingdom Open University. The key
to its success lies in three factors. The first is the the careful
development of learning material by skilled groups over several years.
The second is their understanding that learning can be inexpensive with
expensive curriculum development if delivery costs are low and classes are
large. Courses may have thousands of students. The third factor is that
materials are carefully evaluated and improved before used with students,
seldom done with educational materials. Universities in the United States
discussing distance learning do not seem to understand these factors.

The Open University offers individualized help in tutorial centers
throughout the country. The careful evaluation allows the material to be
flexible with student problems. But the degree of interaction is still low,
for many students.

Personal tutors

A single individual, or small group of individuals, working with a tutor,
is an old idea. An early example was Socrates. The tutorial system at
Oxford University and Cambridge University was similar. The education
of the wealthy has often used private tutors living in homes.

The dialogues of Plato show Socrates in action, we assume.
Socrates did not lecture. He did not deliver information. He asked
questions and student replies stimulated further questions
Socrates asked each student to create her or his own knowledge by
replying to the questions. Today we might this a discovery or
constructivist approach. It resembles to the scientific method, where
questions are asked of nature. One might compare this approach with a
midwife, bringing learning rather than a baby.

The skilled tutor allows for maximum interaction, responding to the
needs of each student. The difficulty is cost. We cannot afford enough
tutors, or even find enough, for everyone. So this attractive model
is not one that we can look to for the future. But it offers valuable
lessons.

Technology based low interaction learning

The final two models involve extensive use of computers. They differ
in the degree of interaction between the student and the learning
units.

Computers have been used in learning for over 40 years, mostly without
the full interactive capability of the computer. Units available now
often show weaker interaction than those of 10 years ago. One
problem is the rise of the mouse, and therefore the use of pointing
as the major mode of interaction.

Examples of low interaction with computers are most learning material on
the World Wide Web. The common interaction is pointing, not providing
the individualized attention needed to attain our vision. The web may
become more interactive, as greater bandwidth is available and developers
are more concerned with quality interaction.

Technology based highly interactive learning

This is the model that offers the best possibility to reach the visions.

We are using the term HIGHLY interactive. The word interactive is much
used with computers, so has lost meaning. Hence it is important
to state what we mean by highly interactive. This is is not difficult,
as we are concerned with learning.

Our model for highly interactive learning material comes from the
personal tutor model. Tutors may work with students in different ways.
These procedures, for the good tutor, stress interaction.
Critical to this student-tutor interaction is the student's native language.
The Socratic tutor asks questions using the student's language and replies
to these questions are in the student's language; students and tutors
have a common language, essential for communication in learning. No
tutor and student have ever communicated by pointing. Language is
critical to understanding student problems. We expect voice input to be
of increasing importance in highly interactive units, essential in some
situations.

Highly interactive learning is very different than lectures. It
resembles a conversation with both people talking frequently. Hence
highly interactive learning units are conversationally interactive.
Long range memory is an important factor; the tutor remembers
characteristics and problems of the learner from previous sessions,
and uses this information in new learning activities.

The importance of highly interactive learning cannot be exaggerated.
The central problem of learning at any moment is to determine what the
student knows and does not know, and to offer appropriate help based
on that knowledge. This allows us to be responsive to the needs of the
student. This knowledge can be gained only with careful interaction with
the student. So non interactive or weakly interactive forms of learning,
can never help all students to learn. Skilled teachers and tutors
recognize this need. The other models cannot fulfill our vision for
the future of learning.

Conversational interaction, based on computer learning units, may seem
a difficult goal to attain. Most learning material available from computers
is, as noted, weakly interactive. Highly interactive learning material
requires different software than previously, far more responsive to
individual student needs.

It may appear that highly interactive software demands
use of artificial intelligence. But this is not the case. We have been
developing such software at the University of California, Irvine, for
30 years, beginning with computers far weaker than today's computers.
As an example, consider the Scientific Reasoning Series, marketed by
IBM. Its aim is to help students understand how scientists
work. The programs are highly interactive, even Socratic.

One program in this series is Batteries and Bulbs. It is based
on an earlier unit developed by the Science Curriculum Improvement Study,
discovering simple laws of electrical circuit behavior. The student
engages in on-line experimentation. The program watches closely to see
if the student is looking for the right data, and is drawing reasonable
conclusions . Help is offered as necessary, but every student
will make the discovery . Another program in the Scientific Reasoning
Series allows students to discover the laws of Mendelian genetics, again
with a highly interactive approach.

These interactive programs, and eight similar programs, were developed
on a computer with only 64 K of memory and no hard disk . They have
been moved to modern personal computers . They include simulations,
but the simulation is only a small part of the program. Most of the code
is for interacting with the student, analyzing students' responses and
deciding what to do next. The interaction is in English, with very little
use of pointing or multiple choice. A graduate student, David Britton,
is working on a voice input version to compare with keyboard input.

The programs in The Scientific Reasoning Series are mentioned to show it
is possible to develop highly interactive materials, even for computers
much weaker than today's personal computers.

Designing Highly Interactive Learning Modules

Our development of highly interactive programs is not dependent on
fancy technology, as commented. It depends on a production process
developed for over 30 years, the Irvine-Geneva system; the developers
were at the University of California, Irvine, and the University of Geneva.
Other systems are possible. Since the materials are to be used worldwide,
the production system eases the process of converting to other languages.

Management

The first stage in the Irvine-Geneva approach is developing a management
system, since the project is likely to be large. Knowledge about such
systems is typically derived from corporations, where management is a
critical problem.

Design

The next stage in development of highly interactive material is pedagogical
design. This is done by skilled teachers from the area developed. There
are two stages, overall design and detail design. For overall design the
general outline is developed, and written descriptions of each module
are prepared.

Detailed design is the most important part of development.
The quality of the materials is highly dependent on what
happens in this process . Interaction is described in detail, as are
details of what is to be stored and how this is to be used.
Skilled teachers work in groups of about four designing a module. The
nature of interactive material is described carefully during the
first morning of group activity. Results are in a 'script,' either on
paper or in a computer.

The responsibilities of the developing group include design of
all media for the final product, in descriptive form. Skilled professionals
fill in the details.

Implementation

The third stage in the Irvine-Geneva process is implementation. This
must include computer code and necessary media. With the online script
editor, much of the programming is done by the computer.

Evaluation

The last stage is formative evaluation, looking for problems associated
with learning. We use the unit with many students, and make improvements
based on information gathered by the computer and professional evaluators.
At least two stages of formative evaluation and improvement are recommended.
Summative evaluation should follow.

Highly Interactive Learning And Our Vision

Finally we consider how highly interactive learning allows us to meet our
vision.

Everyone Learns

The first goal is that everyone learns, in all locations, at all ages,
to the mastery level. This demands that we develop, evaluate, and
deliver a very large quantity of excellent highly interactive computer-based
learning material, at all levels, and eventually in many languages. Very
little such material exists.

It is not necessary to develop a full range immediately. The early material
should be trial development, refining our tactics before larger efforts.
Initial development may be only in one or two languages. This material can
be tested with many students. We can begin with material most needed and
therefore with the largest market.

Because of mastery and individualization, learning may require less time.
This allows more be learned by each individual. One way to insure mastery
is to make frequent testing an intrinsic part of learning and guide
future student study on the results. Testing and learning would no longer
be separated, but would be in intimate combination. Cheating as a
problem vanishes.

Learning is Enjoyable

The importance of motivational aspects of learning cannot be too
highly stressed. We want students to enjoy learning, to want to
continue learning all their lives. This has implications for both
development and evaluation. Pedagogical designers should be frequently
concerned with motivational issues of learning. Highly interactive
units are intrinsically motivating; interaction maintains interest.

Evaluation of motivation is very important. This can
best be done in public environments such as of shopping centers and
libraries. The students are under no pressure from teachers or future exams.
We watch for locations where many students leave. People stay working
only if their interest is maintained. We rework those areas where
motivation is low, and repeat cycles of evaluation and improvement.

Learning is Affordable

The affordability of learning materials is complex. As with the Open
University, two factors must be considered, cost of development and
cost of delivery to the students. The cost of development of very
large amounts of highly interactive learning units will be great.
In the United States costs of development is often underestimated. In
recent years there has been little serious development of learning
material in any mode.

If we deliver material to large numbers of learners, the cost for each
student is decreased. Further analysis is available from the author.
After the process is started, early sales can support future development,
but major initial funding is essential.

Conclusion

The process of developing, testing, and delivering the large amount of
highly interactive units will not be a simple one, but it is possible.
The process should begin in the wealthy countries, and then spread
elsewhere I see no other
possibility for attaining a twenty-first century educational system that
matches our vision: everyone learning, learning is enjoyable, and
learning is affordable. We should proceed with careful initial testing.

those of you who are unfortunately caught up in expedient
movements in education need to take a close look at the nature
of the hope on Pandora's chip. You're dealing with as powerful
a tool as the gods have ever given us.

Hugh Burns
Pandora's Chip: Concerns About Quality CAI
Pipeline, Fall 1981

Further details are available. The longer version of this paper was
spoken to the computer with IBM ViaVoice Gold.

---------------------------------------------------------
Forum website: http://ifets.gmd.de/
Email address for sending message to everyone on the list ifets@gmd.de
Forum's contact person ifets-info@gmd.de
Join/Leave Digest or Normal List http://ifets.gmd.de/maillist.html
---------------------------------------------------------


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b3 on Wed 28 Oct 1998 - 20:32:51 MET