Appendixes



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 59
--> Appendixes

OCR for page 59
This page in the original is blank.

OCR for page 59
--> Appendix A Examples of Capitalization in Fields of Research and Application To gain a better understanding of the capitalization process, the working group examined a number of specific fields of research and application during the course of the study. In several of these cases, a workshop or expert panel discussion was organized and a write-up was prepared on the basis of discussion and background research. The experts who participated in the discussion and others were asked to review the draft write-ups for accuracy. In other cases, the working group prepared write-ups based on telephone interviews with experts and a survey of the relevant literature. The working group has worked to ensure that the write-ups give an accurate picture of a given field, but they inevitably reflect the insights and opinions of the individual experts consulted. In several cases, the working group worked closely with other Academy complex units in organizing the workshops and preparing the write-ups. The working group was only able to cover a limited number of fields and did not attempt a comprehensive assessment of capitalization across all fields in every country. Through experimentation, the working group found that the examination of well-defined subfields and specific applications (e.g., speech recognition and monoclonal antibodies) generated more useful insights than the study of broader fields (e.g., computer science and biology). The examples were selected through consultation among working group and COSEPUP members, staff, and other experts. The working group looked for examples in which success and failure, and the causes of each, could be determined clearly. This proved to be difficult. In most of the examples, a closer examination showed elements of both success and failure. In some instances the success factors and barriers to capitalization were fairly clear; in others, causality was difficult to establish. The examples illustrate a number of important issues related to capitalizing, and are referenced throughout the report. The examples, along with the existing literature that the working group reviewed on topics such as innovation and technology transfer, provided the raw material for the framework of the study, the conclusions, and recommendations. Write-ups of the examples are provided in this appendix and in Box 3-1. Table A-1 summarizes the examples and insights.

OCR for page 59
--> Table A-1 Capitalization in Specific Fields Examined During the Study Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Catalysis [workshops, background research] *The United States is performing at the frontier *The chemical industry has long been characterized by intense global competition. Companies in the United States, Europe, and Japan are effective at capitalizing on research wherever it appears. *It appears that in this field U.S. industry is cutting back on long-term work   *U.S. universities have become world leaders in a number of subfields over the past 15 years *New opportunities are emerging to apply catalysis in advanced materials and more environmentally *Confirms the value of working at the frontier in a global industry.   *Integration of research and education is seen as a U.S. advantage. *Some experts are concerned that, despite a growing commonality of interests, U.S. academia and industry appear to find it easier to work with foreign partners than with each other in these promising areas. *Industry-university differences in perspective emerged over how to promote effective collaboration   *U.S. petrochemical companies have cut back on research in some areas   *Several discussants highlighted the excellence in U.S. advanced education, but raised questions over whether students understand industry problems.   *Excellent academic and industrial research in Europe.   *Government plays a smaller role in this field relative to others examined in the report.       *Start-up companies play a smaller role in this field than in several others examined.

OCR for page 59
--> Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Application of research on cognition and learning to education [interviews, literature survey] *The United States appears to be working at the frontier. *It is inherently difficult to translate research findings into real curricula and teaching practices, and to test new approaches. *Unlike some of the other examples, there are few examples of success, and few established capitalization pathways.   *National Science Foundation, U.S. Department of Education, National Institutes of Health, and private foundations support relevant research. *Some work aimed at capitalization in this area is going on, but the scale of effort is not large. *The policy environment for capitalization is not particularly supportive, and the research community itself is somewhat fragmented   *A number of experts consulted believe that research has produced insights that could be applied successfully in classrooms. *Capitalizing requires cooperation between researchers, education schools, teachers, and school districts, but incentives to collaborate are weak.       *It is well known that other countries outperform the United States in K-12, but it is unclear whether any of this gap is due to more effective capitalization abroad.  

OCR for page 59
--> Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Bioinformatics [workshop, commissioned paper] *The United States appears to be at the frontier in this rapidly emerging field. International companies are ramping up research activities in the United States. *The data produced by the Human Genome Project and related efforts is proving valuable to pharmaceutical research much more rapidly than anticipated. *Although the U.S. appears to be doing as well or better than other countries, this example raises serious future questions about how we invest in human capital.   *At the same time that many Ph.D. holders in the life sciences appear to be underemployed. there is a shortage of people combining experimental expertise and computer skills. This may emerge as a barrier to both long-term progress in research and capitalization. *Utilization of engineering approaches to the study of relationships between gene expression and disease promises to deliver many new therapies. *The life sciences are more insulated from nonacademic labor market demand than most other fields.     *U.S. mobility between academia and industry is an advantage in capitalization, but we may be eating our academic seed corn. *It has been difficult so far to establish new interdisciplinary programs in this field. Liquid crystals displays [interviews, background research] *The original discovery of liquid crystals was made in Austria in the 19th century. *Widespread applications of LCDs emerged in the calculator and watch businesses in the 1970's, with Japanese and U.S. firms both involved. *Even in this area, where other countries have taken the subsequent lead in commercialization, the United States was first in reducing fundamental understanding to practice.   *Key developments in liquid crystal materials were made in England in the 1960's. *Today, the largest market for LCDs is in portable computers. *Illustrates Japanese strengths in improving technologies developed abroad and creating new applications.   *The first prototype LCD was developed at RCA's Sarnoff Lab in the United States. *Japanese firms emerged as leaders in the 1980s because of focus on incremental improvements, cost, and manufacturability.       *Korean firms have entered the LCD business.  

OCR for page 59
--> Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Monoclonal antibodies [workshop, background research] *The fundamental breakthrough by Köhler and Milstein was made in Britian. *Development of human diagnostics and therapeutics has been pioneered by U.S. start-ups. *Illustrates the importance of working at the frontier to capitalize on foreign science.   *The United States has produced many subsequent research gains. *There were capitalization setbacks in the early 1990's, when several septic shock therapies were not successful. *Illustrates the importance of venture capital, university-industry collaboration, and the scientist-entrepreneur as important U.S. strengths.     *More recently, several therapies have proven successful. Monoclonal antibodies remain an important area in the biotechnology industry.   Network systems [workshop] *The United States appears to enjoy clear leadership in many areas. *Taking into account all areas of networking, this is an enormous, growing market. *Illustrates the serendipitous spin-off benefits of government mission-related investments.   *Government investment in the Internet and related areas has played a major role. *U.S.-based companies are the leaders in capitalization. *Illustrates the importance of venture capital and engineer-entrepreneurs as U.S. strengths.   *Applications and commercialization are moving so rapidly that it is difficult to interest talented people in more fundamental areas of research.   *Illustrates the possible problems of imbalance between long-term and short-term research investments.   *Ironically, government investment in fundamental research is seen to be lagging, partly because of the rapid growth in market success for U.S. companies.   *Illustrates the shortening time horizons of U.S. industry.   *A number of the prominent U.S. companies do not do research, just product development.    

OCR for page 59
--> Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Numerical-control machine tools [background research] *The original breakthrough was made at MIT on an Air Force contract. *While capitalization in the United States focused on higher-end defense applications, Japanese companies developed lower-cost applications of NC technology in general-purpose tools. *Illustrates the effectiveness of Japanese institutions that acquire, diffuse, and improve technology, such as industry associations.       *Illustrates challenges to the United States to capitalize in fields requiring incremental improvement over the long-term. Optical sensing [workshop] *The focus of the discussion was on fiber optic sensors. This is an interdisciplinary and somewhat fragmented field, where applied work is targeted at particular needs. *Optical sensors have a number of advantages for specific applications, but have not emerged in a large commercial market. *Participants credited the SBIR program with making important contributions to capitalization efforts by smaller U.S. companies.   *The United States is clearly ahead in areas where there is significant funding, such as space and health care applications. The United States also publishes more papers. One expert believed that Europe might be ahead overall, however. *In some areas it has been difficult to reach cost-performance levels necessary to displace older technologies in established applications. *Concern was expressed that U.S. government funding is becoming too focused on applications work with short time horizons.     *Both small and large companies play important roles. *Participants expressed the view that greater familiarity by students with industry needs would enhance U.S. ability to capitalize.

OCR for page 59
--> Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Oxygen steelmaking [background research] *The process was developed in Austria, but was widely available worldwide through licensing. *Capitalization on this technology was an important factor in Japan's emergence as a leader in the global steel industry during the 1960's and 1970's. *Illustrates the Japanese government's role in facilitating foreign technology acquisition during the high-growth period.     *The Japanese government coordinated licensing of the technology, and encouraged collaboration in developing generic technology. Firms competed in implementation.       *U.S. companies did not capitalize as effectively, and fell seriously behind in the steel industry.   Piezoelectric ceramics [workshop] *Most groundbreaking work has been done in the United States, funded by the DoD and performed at U.S. universities. *Applications are in actuation and sensing in a number of areas. *Illustrates that U.S. advantages in venture capital, flexible human resources, and strong government research support do not translate into capitalization success in all fields. In fields with low margins, a significant existing infrastructure, and a need for incremental improvement, Japanese companies are still formidable.   *Japanese companies that manufacture ultrasonic imaging machines are also supporting long-term work. *U.S. companies lead in the military and mission-critical applications supported by government funding (Hubble Telescope).       *Commercial applications in large markets are emerging, such as in hard disk drive manufacturing. Japanese companies appear better positioned to capitalize.  

OCR for page 59
--> Field [Source of information] U.S. Standing in Research Capitalization Situation and Trends Key Points for the Study Speech recognition [workshop] *Speech recognition has been a focus of long-term government and industry research support since the 1960's. *Despite progress in fundamental research over many years, speech recognition did not find its way into applications until the 1980s. *Large U.S. companies are more focused on results and products than previously. In this field the shift has produced benefits.   *The United States is working at the frontier, and has more extensive activity. Europe and Japan also have well-funded programs. *Capitalization appears to be accelerating, and occurring fastest in the United States. *As in other rapidly growing fields in information technology, there is a problem of retaining some of the best talent in academia.     *The increased power and lower cost of computer hardware and other external factors contributed to capitalization. *Illustrates that long-term efforts, a combination of progress in a range of fields, and a period of "research gestation" may be required for capitalization to occur. Fuzzy logic [interviews, background research] *The United States produced the first breakthrough. Research communities in Europe and Japan embraced this work, and now excellent research is ongoing around the world. *First applied by European and Japanese companies in industrial controls, transportation systems, and home appliances. *A rare example of a U.S. research advance first reduced to practice and commercialized abroad.   *A substantial group in the U.S. research and applications community believes that the benefits claimed for fuzzy logic are easier to come by using traditional logic. *Slowly being applied by global U.S.-based companies.   Applications of economics: options pricing and spectrum auctions [interviews, background research] *U.S. academic economists produced the relevant breakthroughs. *Research in both fields was first utilized in the United States. *Economics research is not a top priority of national policy, but the field has generated a number of advances that have been capitalized on over the years.     *Economics research does have established pathways for capitalization by financial services firms and policy makers.  

OCR for page 59
--> Monoclonal Antibodies Antibodies are soluble proteins produced by the immune system in response to potentially harmful antigens such as viruses and bacteria (Haber, 1996). They bind to specific antigens and help to destroy them. Antibodies to even a single antigen are highly diverse and heterogeneous, produced by many different types of cells. Some antibodies, once activated by a disease, help to provide continuing resistance to that disease. This characteristic makes it possible to develop vaccines, which consist of killed or weakened bacteria or viruses that stimulate the production of antibodies against those antigens (Biotechnology Industry Organization, 1989). For many years, scientists tried to produce antibodies in pure form. As part of their research on the genetic basis of antibody diversity, Georges Köhler and Cesar Milstein developed a method of producing large amounts of pure, monoclonal antibodies (MAb), in 1975 (Raiten and Berman, 1993). In this method, tumor cells that reproduce endlessly are combined through cell fusion with mammalian cells that produce an antibody. The resulting line of fused cells, or hybridoma, are immortalized and produce only one type of antibody. Köhler and Milstein won the Nobel Prize in 1984 for this work. The discovery of MAb technology has been a boon to research and public health, although at various times, expectations have been higher than what could be delivered in the short-term. MAb/hybridoma research and applications, both past history and current trends, illustrate a number of the strengths and issues for the United States and its ability to capitalize on research leadership, particularly in biotechnology and biomedical fields. Initial applications and commercialization Although Köhler and Milstein had done their work in Great Britain, the strong U.S. research base in immunology was quickly able to understand the implications of the discovery and begin developing applications. Much of this work was done in universities and was funded by the National Institutes of Health (NIH) and other government agencies. Close collaboration between small high-technology start-up companies and universities characterizes commercial biotechnology in the United States. The U.S. environment for research and commercialization also allowed for relatively free movement of skilled researchers from universities to industry, and for the recruitment of experienced managers for start-up operations. One prominent example of the importance of university-industry collaboration and "people linkages" is Centocor, Inc., and its founder, Hubert J. P. Schoemaker. Schoemaker immigrated from the Netherlands and received a Ph.D. in biochemistry from Massachusetts Institute of Technology (MIT). He then went to work for the medical products group at Corning Glass Works (now Corning, Inc.), which was using polyclonal antibodies for diagnostic applications (H. J. P. Schoemaker, Centocor, personal communication, November 1996). Schoemaker's scientific and business background provided good preparation for launching a

OCR for page 59
--> but there were differences in perspective over how to accomplish this. Some participants were positive about initiatives such as the ATP Focused Program on Catalysis and Biocatalysis. Others favored initiatives that would more directly involve universities. For example, catalysis is not the focus of any special continuing federal research effort such as the Engineering Research Centers or Science and Technology Centers of NSF Finally, a number of participants observed that education and human resource issues are critical, as was true for just about all of the other cases. The general sense of the workshop discussion was that the United States does a good job of training students relative to that of other countries. Although demand for chemists and chemical engineers focused on some areas of catalysis has been slack, their skills and knowledge are often transferable to other areas, such as surface science, where the electronics industry has a growing demand for talent. There were differences of perspective over whether the lack of jobs in catalysis research should be seen as a negative, or whether the flexibility of students should be seen as a positive. Examples of Japanese Capitalization on External Research This section describes several cases in which Japan has capitalized on research performed elsewhere: the basic oxygen steel process, numerically controlled (NC) machine tools, and LCDs. Basic oxygen steel process Developed in Austria in the early 1950s, the basic oxygen process uses pure oxygen rather than air to convert molten iron into steel. 6 It allows higher productivity and the utilization of a wider range of raw materials than earlier processes. During the 1950s, Japanese engineers did not have as many resources to stay abreast of global technological developments through travel and technical journals as did Western engineers. Japan's trading companies played a significant role by gathering information about the oxygen steel process and disseminating it to steel companies. By 1955, Nippon Kokan and Yawata Steel had learned enough about the process to became interested in licensing it, and approached the Ministry of International Trade and Industry (MITI) for foreign exchange approval to conclude licensing agreements. MITI brokered an agreement whereby Nippon Kokan would be the principal licensee, but would sublicense the technology to other Japanese steelmakers. This was a common MITI practice, which lowered the overall price to Japanese industry of critical foreign technologies. MITI and the steel industry also set up the Basic Oxygen Committee in 1956 to act as a clearinghouse for information exchange about the new process. The committee held regular meetings and facilitated informal contacts among engineers. 6.   This account is based on the account by Tessa Morris-Suzuki (1994, pp. 189-191), who, in turn, bases much of her account on that of Lynn (1982).

OCR for page 59
--> At the same time, individual companies were competing to refine and adapt the process. Japanese steel companies worked with firms in related industries to develop complementary innovations. For example, the new process caused the refractory bricks lining the new converters to wear out very quickly. Yawata Steel and Kurosaki, a refractory brick maker, developed an improved brick. More rapid adoption of the basic oxygen process than that by U.S. and European steel companies was a factor contributing to the success of the Japanese steel industry over the next several decades. In 1960 the Japanese steel industry was only half as productive as the European industry and one-third as productive as the U.S. industry. By the early 1980s, Japanese productivity in steel had overtaken that of the United States and Europe. The Japanese steel industry had several advantages, such as rapid growth in demand for steel, which gave Japanese firms more opportunity to build new plants with modern technology. Still, this case illustrates the effectiveness of Japanese institutions such as trading companies, industry associations, collaborative research, and government coordination, in scouting, importing, diffusing, and improving foreign technologies during the postwar period. NC machine tools Numerical control, which allowed machine tools to be automated, was developed in the early 1950s by a subcontractor to the U.S. Air Force in cooperation with researchers at MIT.7 An MIT report on NC machinery was brought to Japan by a Japanese professor working at the University of California, and publicized by an industry research association. Several companies and universities in Japan started working on NC technology. Fujitsu, a telecommunications equipment company, set up a team to work on the technology, and produced a prototype NC turret punch press in 1956. Fujitsu began to work with other machinery companies to develop the technology further. Japan's first commercial NC tool was developed by Fujitsu, Hitachi, and Mitsubishi Heavy Industries for use in the latter's Nagoya aircraft factory. Fujitsu set up its Fanuc subsidiary to focus on NC technology in the late 1950s. During the 1960s, Fanuc played a key role in incorporating advanced electronics, first transistors and then integrated circuits, into NC controls. Fanuc and other Japanese companies also continued to stay abreast of research developments in the United States and actively licensed technologies. Because of cost reductions enabled by use of microelectronics and other factors such as rapidly rising labor costs and growth in Japan's machine-tool demand, a significant market for relatively inexpensive general-purpose NC tools developed among Japan's small manufacturers during the 1960s and 1970s. MITI and regional governments set up programs to promote technical information exchange and provide assistance to small manufacturers, which also fed this growth. 7.   This account is based on Morris-Suzuki (1994, pp. 199-202), who in turn cites Friedman (1988), as well as Japanese language sources.

OCR for page 59
--> In contrast to developments in Japan, U.S. machine tool makers focused their NC product offerings on highly sophisticated customers, a profitable but relatively small market. By the time smaller U.S. manufacturers began to demand NC tools, Japanese companies were better positioned to supply them. The Fanuc NC controller became the industry standard. By 1983, Japan led the world in machine tool production. Lcds Liquid crystal materials were discovered in 1888 by F. Renitzer, an Austrian botanist.8 In 1963, George Heilmeier and other researchers at RCA discovered that electrical charges affect how light passes through liquid crystal materials, and they began work to develop an electronic display that would utilize liquid crystals. By 1966, they had demonstrated the first prototype liquid crystal alphanumeric displays in instruments and cockpit applications as well as digital voltmeters and digital clocks. These prototypes were shown to the world in 1968. The ultimate goal was to develop a liquid-crystal flat-panel television that could be hung on a wall. Although RCA was able to demonstrate the technology, it lacked a liquid crystal material that would remain stable at room temperature in the nematic phase in which the display could function. George Gray, a professor at Hull University in England, made the key discovery of cyanobiphenyl materials that exhibited room-temperature nematic phases. Several European firms developed and patented these materials, and continue to hold a strong position in supplying liquid crystal materials today. At the same time that RCA was developing its flat television prototype, the electronic calculator industry was growing rapidly, enabled by developments in microelectronics. American and Japanese companies were at the forefront of this industry, and extensive business and technological ties developed. For example, Intel developed the first microprocessor for use in a calculator made by Busicom, a Japanese firm that has since gone out of business. Rockwell International sold key calculator components to Japan's Sharp, which assembled them. In the early 1970s, leading calculator companies were searching for an appropriate display technology to use in hand-held units. The display would need to be visible in ambient light and not consume an excessive amount of power. Rockwell and Texas Instruments both did work on LCDs. Combining insights from its own work, exposure to Rockwell's work, and technology licensed from RCA, Sharp produced the EL-8025, which it claims is the world's first electronic calculator using an LCD. Rockwell also produced a calculator with an LCD at around the same time, but soon exited the calculator business because it was a low-margin activity outside the company's core military and space work. 8.   This account is based mainly on a telephone interview with Lawrence Tannas on November 25, 1997. It is supplemented by material from Tannas et al. (1992) and material from the Sharp Corporation World Wide Web page.

OCR for page 59
--> Although light-emitting diodes emerged as the display technology of choice for hand-held calculators during the 1970s, LCDs emerged again in the late 1980s with major advantages as a display technology in the rapidly growing electronic watch business. A number of U.S. and Japanese companies entered the LCD business during the latter half of the 1970s. However, the technology was widely available and straightforward, and so, U.S. companies tended to move manufacturing to offshore locations when it became cost-effective to do so. Japanese companies, including Suwa-Seiko (now Seiko Epson), Sanyo, Canon, and others, were more inclined to see such component technologies as fundamental capabilities for a range of consumer-oriented electronics businesses and long-term growth. As was the case with NC tools, Japanese companies continued to supplement their own incremental improvements of LCD technology with insights gained from foreign research. In 1983, Sanyo and Suwa-Seiko demonstrated the first twisted nematic active matrix LCD (TN-AMLCD). The use of a poly-silicon semiconductor substrate was a key improvement pioneered by Sanyo and Suwa-Seiko. However, amorphous silicon soon dominated the industry. The target market at that time was hand-held televisions, first black and white and then color. The super-twisted nematic LCD (STN-LCD) was first reported by European researchers Terry Scheffer and Juergen Nehring in the early 1980s (Scheffer and Nehring, 1990). STN-LCD became the dominant technology for manufacturing portable computer displays, an application that emerged in the second half of the 1980s and is now a multibillion-dollar business. The TN-AMLCD using amorphous silicon became the major display technology in the 1990s for notebook computers and hand-held televisions. Technologies other than LCDs were tried in portable computers, but all have been replaced with LCDs as STN-LCDs and TN-AMLCDs have continued to improve their performance as costs have declined gradually. Japanese firms have dominated this business, although Korean, and to a lesser extent Taiwanese, companies have entered in recent years and appear to be enjoying significant success. In the LCD case, as in basic oxygen steel and NC machinery, Japanese companies displayed adeptness in incorporating new component technologies in a variety of products, enabling the emergence of larger mass markets. Long-term efforts on complementary technologies enabled Japanese industry to capitalize on foreign research. The role of Japanese government and industry research laboratories in gathering information and diffusing technology to individual companies is apparent in the LCD case. Fuzzy Logic Fuzzy logic is a field of research and application in which fundamental discoveries made in the United States were first reduced to practice and capitalized on overseas. Fuzzy logic is a system for representing and manipulating values associated with vague or uncertain concepts, such as "large," warm," and "fast," which can be seen simultaneously to belong partially to two or more different, contradictory sets of values (JTEC, 1993). In contrast to traditional logic, which represents

OCR for page 59
--> objects in terms of sharp distinctions, fuzzy logic allows an object to be represented as a member of a class in a graded way. Fuzzy logic was invented by Lotfi Zadeh, a professor at the University of California at Berkeley, in the 1960s. Researchers in the United States and abroad began developing applications for fuzzy logic. In 1973, at Queen Mary College/ London University, Ebrahim Mamdani and Sedrak Assilian applied fuzzy logic to the control system of a small steam engine. Lauritz Peter Holmblad and and Jens Jorgen Østergaard, corporate engineers at F. L. Smidth (now FLS Automation), learned of this work and began research on an automatic cement kiln control system utilizing fuzzy logic in the mid 1970s (McNeill and Freiberger, 1993). In 1980 the first high-level kiln control system became commercially available, supplied by FLS.9 Today, most cement kilns use fuzzy logic control (L. Zadeh, University of California at Berkeley, personal communication, June 30, 1998). Fuzzy logic caught on quickly in Japan, perhaps because of a cultural tolerance for uncertainty. In 1968, papers on fuzzy logic began to appear in Japanese journals. In 1972, Professor Toshiro Terano of Hosei University introduced fuzzy logic to the research community in Japan and several study groups were formed. This led to research and applications mainly in the area of physical systems control. In 1987, after eight years of development, the fuzzy-controlled Sendai Subway system went into operation (McNeill and Freiberger, 1993, p. 155). The system was developed by Hitachi. Besides featuring an extremely smooth ride, the subway stops and starts more accurately than a human-operated train, and cuts energy usage by 10 percent. By 1990, fuzzy logic had been implemented in a wide range of home electric appliances in Japan (Munakata and Jani, 1994). In contrast to researchers in Europe and Japan, who were receptive to applying fuzzy logic, progress has been slower in the United States (JTEC, 1993). Important segments of the U.S. research community have been indifferent or hostile to fuzzy logic. Although mathematical work on fuzzy logic continued in the United States, the interested community was isolated. More practical, engineering-oriented work was slow to develop. Zadeh himself, who has remained an active and effective advocate for his ideas, believes that discomfort with the word "fuzzy," the American tradition of respect for precision, and an entrenched establishment of control system techniques all prevented the U.S. research community from embracing the theory (L. Zadeh, University of California at Berkeley, personal communication, June 30, 1998). Engineers in industry working on controls for various products have been skeptical that fuzzy logic could deliver better performance than effective implementation of traditional "crisp" logic. By contrast, there was less entrenchment in Japan and an eagerness for new ideas, which facilitated commercialization. Currently, fuzzy logic is applied in a broad range of commercial products, such as automobile climate control and transmissions, microwaves and dishwashers, and other control systems. U.S. industry has become more receptive to utilizing the 9.   FLS web site, http://www.flsautomation.dk

OCR for page 59
--> technology in recent years. U.S. firms, such as Otis Elevator and Motorola, that are active in the Japanese market and eager to respond to their Japanese customers' interest in fuzzy logic, are most advanced. The United States is still among the leading centers of research, with excellent work being done at institutions such as Georgia Institute of Technology and the University of New Mexico. Capitalizing in the Social Sciences Although this report focuses on capitalization in the natural sciences and engineering, research capitalization also occurs in the social sciences. Two examples from the field of economics illustrate the successful application of social science theory to real-world problems. 10 Research in the area of options pricing has been applied to risk management and has helped to make a new exchange system successful. Game theory has been applied to spectrum license distribution, with the result of increased profits for the government and more efficient distribution of licenses. Options pricing In 1997, Robert C. Merton and Myron Scholes won the Nobel Memorial Prize in Economic Sciences for their work on the pricing of options. Myron Scholes and Fischer Black created a formula, first published in the Journal of Political Economy in 1973, relating options pricing to asset price volatility and time. Simultaneously, Robert Merton had applied these results to other types of financial assets. The Chicago Board of Options Exchange (CBOE) opened for business that same month.11 This theory provided an efficient way to manage risk in stock portfolios, which has increased participation and improved liquidity. The presence of the theory makes the market more predictable and therefore more easily and more widely used, enhancing the value of exchanges. "Corporate strategists use the theory to evaluate business decisions; bond analysts use it to value risky debt; regulators use it to value deposit insurance; wildcatters use it to value exploration leases. In fact, the model can be used to examine any 'contract' whose worth depends on the uncertain future value of an asset" (The Economist, 1998). The first application and growth in this area took place in the United States. 10.   In the social sciences, as in the natural sciences, a research advance that is successfully capitalized is not necessarily a success in every single use of what was learned from the research. Some attempted applications of fundamental knowledge founder even when there have been many successful uses. Two such examples have arisen with the applications discussed here. Long-Term Capital Management (LTCM), a hedge fund whose partners included the economists awarded the Nobel Memorial Prize for their work on options pricing, experienced huge losses and almost collapsed before the Federal Reserve Board worked with LTCM's creditors to work out a rescue plan. See The Economist (1998). Also, the Federal Communications Commission's successful auction program suffered a setback when procedures aimed at encouraging bidding by industry newcomers in a May 1996 auction backfired. A number of the winning bidders were subsequently unable to pay for the licenses. See Mills (1998). 11.   www.cboe.com/cboe25th/news.html

OCR for page 59
--> The basic research was done in Boston during the 1970s. The CBOE opened in 1973 and, by 1997, four U.S. options exchanges were trading more than 350 million options contracts on 2,400 individual stocks. There are over 50 options exchanges in the world where options pricing based on the Black-Scholes equation is widely used. Spectrum auctions From July 1994 to May 1996, the Federal Communications Commission (FCC) conducted six auctions for the distribution of radio spectrum licenses for wireless technologies. The system devised for these auctions was based on economic theory. The FCC enlisted John McMillan, an expert in game theory and economist at the University of California at San Diego, to apply the principles of game theory to help optimize the sale of licenses. Game theory was created by mathematician John von Neumann and economist Oskar Morgenstern during the 1940s. Research on game theory was funded during the 1950s and 1960s by DoD, and it has become increasingly important in political science.12 Game theory is suited to highly structured situations, such as auctions. Auction theory is an application of game theory that was first developed for single-item auctions. William Vickrey received a Nobel Memorial Prize for critical analysis in this area. Recent advances in auction theory are for simultaneous multiple-item auctions and the use of experimental economics to design an auction in a way that helps people to reach the predicted equilibrium. Auction theory can be used to help people decide how to bid in an auction and also to design an auction so that the equilibrium will be as efficient as possible. Several other academics helped to design the FCC spectrum auction system. The potential bidders hired consultants who filed briefs to the FCC and then, after the system was devised, advised their clients on the best methods to use during the auctions. Stanford University professors Jeremy Bulow, Paul R. Milgrom, and Robert B. Wilson, Yale University professor Barry J. Nalebuff, and University of Maryland economist Peter Cramton consulted for major telecommunications firms (O'Toole, 1994). After considering all the input gathered on the auction process, the FCC decided on an electronic simultaneous multiple-round auction system. This system was chosen because many items' values were interdependent and an asynchronous auction might undervalue particular licenses. "This auction form proved remarkably successful. Similar items sold for similar prices, and bidders successfully formed efficient aggregations of licenses" (Cramton, 1997). The FCC has demonstrated its auction system to representatives of Argentina, Brazil, Canada, Hungary, Peru, Russia, South Africa, and Vietnam. Mexico has licensed this system and has used it in a spectrum auction. 12.   A Nobel Memorial Prize was given to John Nash, John Harsanyi, and Reinhard Selten for work in game theory.

OCR for page 59
--> Applying Research on Cognition and Learning in Education A number of experts consulted by the panel believe that there is great potential for expanding capitalization on recent research on cognition and learning, which has developed important insights into the functioning of the human mind. This work is ongoing in a number of disciplines, including developmental psychology, linguistics, mathematical logic, philosophy, computer science, and neuroscience, as well as the relatively new interdisciplinary field of cognitive science. Education is seen as a particularly promising area of application, given the content of the research and the pressing educational problems facing the United States, particularly in early care and learning during the preschool and early elementary school years. Yet research on cognition and learning is not making a measurable contribution to early care and education in the United States today, and even strong proponents of the research believe that prospects for the immediate future are mixed at best. This example illustrates the special challenges of capitalizing on research to address certain pressing national needs. Research on cognition and learning The recent wave of research on cognition and learning has its roots in the mid 1950s when a "cognitive revolution" began in American psychology and an interdisciplinary field of cognitive science began to develop.13 The hallmark of this wave of research is the effort to build understanding of human cognition and behavior from models of unobservable mental constructs related to information processing. 14 Research in these fields is being capitalized upon in a number of areas. Work by Herbert Simon and others underlies developments in artificial intelligence, for example. Another example that is interesting because it is a clear case of capitalization success is conjoint measurement (also known as conjoint analysis), a technique based on insights from mathematical psychology and psychometrics. R. Duncan Luce, now a professor at the University of California at Irvine, and others developed conjoint measurement during the 1960s.15 The technique allows for the quantitative characterization of how two or more independent variables affect a psychological dependent variable. This class of problems is recurrent in psychological research. Several years after Luce's work, Paul E. Green of the University of Pennsylvania's Wharton School and others showed how conjoint measurement could be applied to analyzing consumer preferences as an aid to developing and 13.   For an overview of research on cognition and learning, see Bransford et al. (1998). For information on cognitive science as a discipline, see Stillings (1993) (http://hamp.hampshire.edu/-nasCCS/ nsfreport.html). Widespread use of the term "cognitive science" and the appearance of distinct educational and research programs has occurred only over the past 20 years. 14.   John Bruer, "President's Statement," John S. McDonnell Foundation homepage (www.jsmf.org). 15.   The seminal paper is Luce and Tukey (1964, p. 1).

OCR for page 59
--> marketing new products (Green and Wind, 1975). Over the past several decades, the technique has come to be widely used by corporate marketing departments and consulting companies. Research on cognition and learning also has generated a number of insights that have important implications for education and training.16 Research efforts also are being focused on the cognitive development and learning processes of young children. For example, research shows that "naive understanding," the often mistaken prior beliefs and concepts of children, plays an important role in learning.17 If new or contradictory information is introduced without addressing these prior beliefs and concepts, a child may construct a logical loophole to accommodate the contradiction rather than learn the correct concept. The implications of this and other research insights for education are far reaching. Experts consulted by the panel believe that new approaches informed by research could significantly improve science and mathematics education in the early grades. In general, approaches informed by research on cognition and learning focus on developing a deep understanding of basic concepts that corrects the naive understanding of children by guiding them through a carefully structured process of discovery. This often implies much less emphasis on memorization of facts and information than traditional educational methods. Efforts to capitalize The panel was able to uncover several examples of efforts to apply research on cognition and learning in the classroom during the course of exploring this issue.18 Research on how children learn mathematics concepts has informed efforts to develop tools for "cognitively guided instruction," focusing on early elementary mathematics.19 Efforts also are being made to incorporate this improved understanding into training programs for teachers at professional schools of education and associated centers for educational research, often with support from the U.S. Department of Education or NSF. Private foundations play a key role in this area as well. The John S. McDonnell Foundation supports work aimed at applying cognitive science insights to education, and the Alfred P. Sloan Foundation played a catalyzing role in the original development of education and research programs in cognitive science. A capitalization effort in the area of early science learning is under way at the Institute for Research in Cognitive Science (IRCS) at the University of Pennsylvania.20 IRCS is one of NSF's Science and Technology Centers [see COSEPUP (1996a)]. A group of professional curriculum developers, classroom teachers, cog- 16.   For a review of major insights and potential applications, see NRC/CBSSE (1994). 17.   Bransford et al. (1998). 18.   One prominent example is the Learning Research and Development Center at the University of Pittsburgh (www.lrdc.pitt.edu). The examples here are illustrative, and not meant to present a comprehensive picture of developments in this field. 19.   Telephone interview with Thomas Cooney, September 4, 1998 20.   Telephone interview with Christine Massey, September 4, 1998.

OCR for page 59
--> nitive developmental and educational researchers, and university scientists is developing, field testing, and evaluating science curricula to meet the developmental and practical needs of children in early elementary classrooms (kindergarten through second grade). The unit on perception, Science Makes Sense, illustrates the overall approach of this initiative (Massey and Roth, 1997). Rather than utilize the traditional kindergarten approach linking the "five senses" with associated body parts, the IRCS-developed curriculum focuses on how we experience and get information about the world through different modalities. Children are guided through a sequence of carefully structured exercises that isolate various sensory modalities and help them become aware that immediate sensory experience can be incomplete or misleading. This approach addresses the naive understanding of five- and six-year olds, such as the common belief that the color of an object can be determined by touch. Science Makes Sense and other components of the IRCS curriculum are being field tested in Philadelphia elementary schools, with encouraging preliminary results. Issues and barriers to capitalization The poor relative performance of U.S. students in international comparative studies of mathematics and science education is well known.21 Improving education has been a major issue on the U.S. national agenda since the 1983 publication of A Nation at Risk.22 Despite the promise of basic and applied research in cognition and learning to improve educational outcomes, there are several significant barriers that need to be addressed in order to realize this promise. This discussion is meant to be suggestive and illustrative rather than comprehensive and conclusive. A full assessment of these barriers would require a separate study. 23 1. Learning how to apply scientific insights requires focused effort. The general insights and principles developed from the sciences of cognition and learning do not apply in exactly the same way in all fields of instruction. Thus, to move from controlled laboratory applications to particular educational settings is a major step that often requires focused research. Even where research provides clear and unambiguous direction for applications work, developing and testing concrete approaches for the classroom can be an arduous process. Effects on sustained learning of different approaches may take years to measure. 21.   The Third International Mathematics and Science Study (TIMSS), which was recently completed, involved collection of data on half a million students from 41 countries, and is the largest, most comprehensive, and most rigorous international study of schools and students ever. The National Center for Education Statistics website is a useful starting point for finding out about TIMSS (http:/ /nces.ed.gov/TIMSS/). The TIMSS results contain no information bearing on the utility of learning research in K-12 education. 22.   Respondents to a opinion poll named education as the issue most likely to influence their voting in the 1998 elections (Balz and Deane, 1998). 23.   An extended discussion of the barriers to knowledge utilization is found in Bransford et al. (1998).

OCR for page 59
--> 2. Up to now, funding for the necessary focused efforts has been limited. Research on cognition and learning is funded by agencies such as NSF, NIH, and the Department of Education. The latter agency also funds research on new educational techniques. Although some programs have funded work to apply research insights in educational settings, the amount of dedicated long-term funding is limited. For example, the IRCS effort described earlier benefits from access to funding through NSF's Science and Technology Centers (STC) program. STC funding has a limited duration, however, and it is not obvious how continued work will be supported after STC support ends. 3. Institutional incentives to transform useful research into classroom practice are lacking. Additional funding may not be enough to change educational decision-making routines. Experience from educational reform initiatives suggests that the most promising efforts involve bridge building between education researchers, scientists, education schools, teachers, and communities. This is inherently difficult because these groups have different incentive and reward structures, and none is tasked with application of research in the classroom as a primary mission. For example, researchers advance their careers through successful publication of their research, which leads to tenure and status in their fields. They are not rewarded for making efforts to apply their research insights in the classroom. Teachers face numerous challenges in the classroom, and may have little time or incentive to learn about new approaches based on research. A lack of clear market signals in education may contribute to this institutional inertia. 4. Other possible barriers could be encountered in disseminating new approaches. Some of the difficulties encountered in applying research on cognition and learning to the classroom are similar to those encountered in other interdisciplinary fields examined by the panel, such as bioinformatics. These barriers include the lack of dedicated funding sources and institutional structures. The application field of education itself could be the source of additional problems in the future. Even if the barriers discussed above can be overcome and new research-based approaches to early education are developed and tested, additional obstacles may be encountered in promoting the widespread adoption of new methods. Several of the experts interviewed by the panel pointed to the advantages that other countries might have over the United States in areas such as stronger systems for funding early care and education, and a stronger national government role in the education system. It is not clear that other countries are applying research on cognition and learning to the classroom more successfully than the United States, however.