Can technology make the recommendations of learning science practical and affordable?
First Monday

Can technology make the recommendations of learning science practical and affordable?

Can technology make the recommendations of learning science practical and affordable? by Henry Kelly

Computers, communication devices, and software are transforming the way our world works — with the painful exception of education. Systems capable of transforming the way people engage ideas and works of creative expression are only beginning to be used in classrooms. Museums and libraries are free of many of the constraints faced by traditional school systems and have a unique responsibility to experiment and point the way. Many opportunities present themselves and a national effort (the Digital Opportunity Investment Trust) can provide a much needed increase in funding.


Redefining the possible
Why has progress been so slow?
State of the art
Why be optimistic?
The role of museums and libraries
But will there be any money?




Redefining the possible

Teachers and the scientists that make a formal study of learning have known for years that there are many ways to improve on the standard methods of classroom instruction. Students learn faster and retain more in environments where they can immediately apply what they have learned to practical and interesting problems. They learn when they fail and can receive help from tutors who can answer questions quickly and in a way that reflects a deep understanding of the individual student’s background, strengths, weaknesses, and interests. And they certainly learn more when they’re given a challenge that motivates them to spend the time and intellectual resources needed to succeed. The problem, of course, is that it has always seemed unaffordable to actually implement such concepts. They would seem to require a huge increase in the number of teachers and infinite resources for equipment, field trips, and expert advice.

New information tools can change the rules. Take the problem of finding a way for students to put their formal knowledge to work in addressing interesting challenges. Computer games prove that students will learn all manner of obscure things to get to new game levels in highly sophisticated virtual environments and there is no reason the same basic tools can’t work in learning. Compelling simulations can recreate the experience of walking through a foreign city — viewing its monuments, watching people at work and play, and experiencing works of art in their original settings. Students can be given access to simulated equipment such as powerful telescopes with complete sky maps, gene sequencing equipment, machine tools, airplanes, and rockets so true to the original that users often have trouble telling the difference. New technologies make it possible for students to direct synthetic players onstage or coach them through a ballet. They can build a virtual bridge and see if it collapses under simulated traffic, or try to find out why virtual frogs are dying in a virtual environment.

The challenge of providing personalized guidance from tutors who know the students, the subject, and the context of questions can be met using technology already being used to provide "help desk" assistance to customers and employees of major firms, and "just in time" advice to everyone from intelligence analysts swamped in data to people trying to get new software operating or to un–jam the office copier. Many modern systems put "cookies" on your computer that can help contextualize answers to reflect a deeper knowledge of the person asking and a record of what has been asked in the past. Anyone who has fought to get the Microsoft paperclip off of the screen knows that these systems are far from perfect. But they are improving. Some of the best skilfully blend automated answers to comparatively simple questions with a link to live experts for questions that require more sensitive and intelligent responses.

Systems that constantly observe what a student is trying to achieve and how they’re attacking problems can also provide a rich and sensitive measure of what the student knows and doesn’t know. Given adequate research, systems can be designed to measure both formal knowledge and provide a vivid measure of how well the formal knowledge transfers into practical skills — whether the skill is composing music, flying an airplane, or unravelling a complex problem in environmental chemistry. They can, in principal, provide tests of competence that make sense to the students, their current and future instructors, and prospective employers.



Why has progress been so slow?

The spectacular tools that have transformed service industries such as banking and led to the creation of entire entertainment industries such as computer games have not had a major impact on the nation’s educational systems. The nation benefits from a diverse system, but the highly fragmented management accompanying it makes for a nightmarish market for private developers. Even the most sophisticated educational organizations — say Princeton University — have no management infrastructure for evaluating how to manage technical advances in instruction, no facility for procuring innovative content, and no significant research program focused on designing and testing new approaches to learning. Government programs have failed to fill the gap. The U.S. spends about a trillion dollars a year on education and training but probably less than $200 million on research on how to use technology to improve learning — and most of this is spent by the Department of Defense.

No firm that has survived a transformation to new strategies based on information technology would argue that the process has been painless. It is often necessary to completely re–evaluate what clients want and what firms can provide. It is necessary to dramatically change management strategies and redefine jobs. These are not easy choices for political leaders to make.

The situation has not been helped by the fact that advocates of technology–based learning have often made wildly optimistic claims for what the technology could achieve. Their biggest mistakes have not been in their claims for what is possible — indeed spectacular progress is possible. They vastly underestimated how hard it was going to be, how long it would take, and how much it would cost. The fact that a lot of investors lost a lot of money believing these people during the dot com bubble hasn’t helped build enthusiasm for new efforts.



State of the art

One painful artifact of the confused market for innovation in traditional educational systems is the enormous underinvestment in research. Not only is the total investment small, but there is also no tradition for the kinds of systematic research planning that have been essential for progress in all other fields. A group of some of the nation’s leading scholars in cognitive science, pedagogy, computer science, and other fields got together a few years ago to assess the state of the art. There was universal agreement both that a dramatic increase in funding was needed, and that the community was completely unprepared to spend new money. There was no organized program for setting research priorities, balancing basic and applied research, or taking the problem apart into manageable components. Traditional educational research involved giving a few tens of thousands of dollars to an individual and asking him or her to invent a new approach to instruction. This "cottage industry" research approach is completely inappropriate for the enormously complex and expensive research problems faced in trying to design and use elaborate tools for simulations, question management, evaluation, and for providing other services needed to achieve the learning gains that cognitive experts suggest are possible.

The original discussions led to the formation of a group called the Learning Federation, which was organized to address the problem of research design. Hundreds of the nation’s leading experts drawn from universities, companies, and government research organizations who worked in diverse fields ranging from software development to learning science worked for over two years to develop a detailed set of research priorities and a detailed plan that could be put into action quickly. The plan, available online at, began by dividing the research problem into five distinct (but highly interrelated) research topics each with a separate set of goals and timetables. They are:

1. Instructional Design for New Technology–Enabled Approaches to Learning

This research involves understanding how people learn, how experts organize information, and the skills of effective learners. Research goals include:
  • Developing a framework to integrate research results;
  • Understanding how exploration, discovery, and games can be used to improve learning; and,
  • Understanding the comparative advantage of environments of libraries and museums (unique expectations, capabilities, people).

2. Building Simulations and Exploration Environments

This research involves creating the simulated worlds and equipment that can support discovery based learning. Goals include:
  • Ensuring interoperability of software developed by different designers and different academic disciplines;
  • Designing ways ontologies and metadata that facilitate communication of concepts and information among people, between people and automated devices, and between software components operating on these devices;
  • Specifications of geometries in ways that would facilitate sharing of components such as virtual buildings for virtual cities and that would allow work to be maintained through many generations of operating systems and hardware;
  • Improved methods for navigating through complex simulated worlds.

3. Question and Answer Systems

This includes:
  • Methods for stimulating learner questions;
  • Interfaces that make it easier for students to ask questions;
  • Tools for interpreting, comprehending and answering questions;
  • Context sensitive search of multimedia data;
  • Preparation of multimedia answers; and,
  • Tools to advance the discussion with the learner and to summon teachers and other experts as needed.

4. Learner Modeling and Assessment

This includes:
  • Models of content expertise, competency and pedagogy;
  • Automated modular assessment design, development, delivery and analysis; and,
  • Multi–dimensional learner models and measurement methods.

5. Tools for Combining the Component Tools into Integrated Learning Systems.

This includes engineering strategies for using learning system tools to build learning systems. Tasks include:
  • Course building tools for designing scenarios, creating assignments, designing response to information gathered from student observer tools, and programming avatar behaviors;
  • Shareable Content Objects;
  • Tools and services to assist developers in the application of metadata; and,
  • Tools for worldwide collaboration on building and maintaining learning environments.



Why be optimistic?

A well funded research program built around the goals just listed could yield enormous benefits. But impressive systems will not help if there is no market for them. Given the obstacles to progress I discussed earlier, why is there any reason to believe that a market will develop? It is because the economics and the politics of the issue have shifted in fundamental ways during the past few years:

  1. We now have a generation of kids and young adults who have been weaned on video games, Web browsing, and other new information tools. They want to learn using new tools and are frustrated by "IT free zones" they find in schools.
  2. Tools for using powerful simulation and visualization software that once cost millions are now crammed in $200 game boxes and inexpensive computers.
  3. Advanced systems for answering questions, building multi–user environments with instant messaging and chat technology are familiar tools for connecting people to people and creating tailored answers.
  4. The Department of Defense and other organizations that have made significant investments in development and testing have spectacular and convincing proof of what can be achieved with well designed simulation based training and other IT training systems.



The role of museums and libraries

The very real possibility of a revolutionary change in the way people can be connected with ideas and creative works, coupled with an energetic constituency for change, has created an exciting and unstable situation. A few creative acts could catalyze a dramatic change in the way people learn. Given the huge constraints facing schools at all levels, we need to look elsewhere for this change to occur. Museums and libraries are perfectly positioned for the task. They are respected parts of the educational system, but free of many of the political constraints (school boards, standardized testing, and endless faculty meetings) that can make it difficult for traditional educational institutions to act creatively. Of all the learning institutions, they have the greatest freedom to define new rules and break old ones.

Museums and libraries are expected to use creative new approaches to learning and have a long tradition of working with artists and creative designers to explore new ways to make material come alive and appeal to diverse audiences. They are not afraid to experiment with new media or work with multi–disciplinary teams of experts, artists, and curators to create unique experiences. They are expected to break the barrier separating scholarship and the real stuff of history, science, and the natural world. Most importantly, they are places designed both for serious study and fun.

Museums and libraries have already set precedents for leadership in information technology. They were among the first institutions to recognize many of the opportunities we take for granted in using digital formats and networked resources. And the results have been spectacular in:

  • Allowing near universal accessibility of rich collections of texts and images;
  • Allowing even remote schools and homes to have affordable access to materials that would have taken specialists with enormous budgets decades to assemble;
  • Breaking the boundaries separating collections (geographic, subject matter, text/artifact/image);
  • Developing powerful tools for searching diverse collections (text/image/artefact); and,
  • Solving the riddle of how to protect precious archival material while making it easily available to a broad audience.

However wonderful these advances have been, they should be seen as only a beginning. This work has made existing text and image collections available. The challenge now facing the museum and library community is how to catalyze a real revolution in learning by making materials available in the entirely new forms I outlined briefly above. Success will depend on their ability to continue to break traditional boundaries separating scholars, information scientists, artists, and computational scientists. It will also depend on finding the creative energy to invent and test dramatically new methods of presenting complex information. But their track record is good.

Importantly museums and libraries also have unique experience in meeting the challenge of presenting information in a way that is accessible to people of all ages, cultural and educational backgrounds. These institutions have explored ways of enriching the experiences of groups as diverse as high school friends or a multi–generational family. This gives them a particular advantage in exploiting another of the powerful capabilities of new information technology: the ability to let a group of people share the same virtual environment and work together to meet a challenge — discussing options, teaching, discovering, and encouraging each other.

Having done so much to define the power of information technology in learning, museums and libraries have a unique responsibility to keep the process moving.



But will there be any money?

Achieving the revolution in learning made possible by new technology will require an organized plan for action and some creative institutions to lead the way in developing and testing new ideas. But it also needs money — something not in great supply during these fiscally constrained times.

There is, however, some good news on this front. A growing number of educational leaders and leaders in the Executive branch and the Congress are beginning to recognize that use of new technology must play a critical part in meeting the enormous and growing challenge of providing people in the U.S. and the world with the lifelong learning services. Such skills are essential to prosper in a fast–paced, technologically–sophisticated economy. Clearly the new technologies aren’t sufficient in themselves, but they are a key part of the solution. The potential impact of a comparatively modest federal investment in the development and testing of new learning concepts can be enormous.

The Bush administration is exploring opportunities for learning research in a new interagency group. It is beginning by trying to understand how federal agencies are now investing in the kinds of research described by the Learning Federation. This includes research supported by DoD, the National Science Foundation (NSF), the Department of Education, IMLS and many others. The NSF is actively pursuing new concepts in technology and will shortly announce several new "Science of Learning Centers" at least one of which will almost certainly focus on many of the research challenges in instructional technology.

Congress is actively considering an ambitious proposal that would represent a major breakthrough. The Digital Opportunity Investment Trust or DO IT (S. 1854) would use interest from funds realized from the sale of public spectrum to fund exactly the kinds of projects needed to accelerate development and use of the revolutionary learning tools I’ve been discussing. The bill is completely neutral about all of the complex issues surrounding spectrum sale, and is limited to the proceeds of currently planned sales. In brief DO IT would do three things:

  1. It would support the ambitious program of research, development, demonstration and testing recommended by the Learning Federation;
  2. It would fund development of key exemplar applications including both critical tools (such as simulations of historic sites) and integrated learning systems built on simulations, question management, evaluation, and other tools; and,
  3. It would support an ambitious program to digitize and catalogue collections in museums and libraries around the country.

It is hard to think of any innovation more important than one that could make learning more productive and more engaging for Americans of all educational backgrounds. And it is thrilling to consider that the new tools could bring the riches of history and creative contributions from collections around the nation to every American. Museums and libraries have a long tradition of experimenting with new learning strategies and have a unique responsibility to lead a critical national effort. I’m optimistic that political leaders in Washington will provide these institutions with the resources they need to create and develop technologies that will transform the museum and library experience. I can’t think of a more productive use of federal educational funds. End of article


About the Author

Henry Kelly is the President of the Federation of American Scientists (FAS). Kelly received a PhD in physics from Harvard University and is a fellow of the American Physical Society. Before joining FAS in July 2001, he spent over seven years as Assistant Director for Technology in the White House’s Office of Science and Technology Policy. Prior to his work in the White House he was a Senior Associate at the Congressional Office of Technology Assessment, Assistant Director for the Solar Energy Research Institute, and worked on the staff of the Arms Control and Disarmament Agency. Kelly is the author of numerous books and articles on issues in science and technology policy.

Editorial history

Paper received 15 April 2004; accepted 22 April 2004.

Copyright ©2004, First Monday

Copyright ©2004, Henry Kelly

Can technology make the recommendations of learning science practical and affordable? by Henry Kelly
First Monday, Volume 9, Number 5 - 3 May 2004

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2019. ISSN 1396-0466.