the drivers of interdisciplinary research

7/30/2019 The Drivers of Interdisciplinary Research

1/16

FACILITATINGINTERDISCIPLINARY

RESEARCHCommittee on Faeilitating Interdiseiplinary ResearehCommittee on Scienee , Engineering , and Publie Poliey

NATIONAL ACADEMY OF SCIENCES,- - - ~ A T l O N A L ACADEMY OF ENGINEERING, AND

INSTITUTE OF MEDICINE

..U",..... -.. l:!~ ~o o . .

OF THE NA TlONAL ACADEMIES

3 -: c e . , .;;;;;="..BibliotecaEspacio InterdlsclpllnarioUniversidad de la Repblica. UruguayTHE NATIONAL ACADEMIES PRESSWASHINGTON, D.C.www.nap.edu

00 1 O 4


2/16

2The Drivers ofIn terdisciplinary Research

N one can predict the issues that sc ience and society will considermost pressing in the decades to come. But if we loo k at sorne highpriority issues of today- such as world hunger, biomedical ethics,sustainable resources, homeland security, and child deve lopmen t and learning-and pressing research questions, such as the evo lution o vir ulence inpathogens and the relationship between biodiversity and ecosystem func-tions, we can predict that those of the future will be so complex as torequire insights from multiple disciplines. What resea rch strategies areneeded to address such a future ? To what extent will interdisci plinaryresearch (IDR) and interdisc iplinary education be among the strategies?Just what is IOR?

DEFINING INTERDISCIPLINARY RESEARCHNo single definition is likely to encompass the diverse range of activitiesthat have been desc ribed under the heading of IOR. Reflecting the diversityof modes of interdisciplinary work, severa l organizational models haveevolved (see Table 2-1 ). For the purpose of this report, the committee hasdeveloped the following desc ription as a point of departure:Interdisciplinary research (IDR ) is a mode of research by [eams oc indi-viduals rhar integrares nformarion, data, techniques, [ools , perspectives,concepts, and/or rheories from two or more disciplines oc bodies of spe-cialized kn owledge ro advance fundamental understanding or ro salveproblems whose solurions are beyond rhe scope of a single discipline ocfield of resea rch practice.

26


3/16

THE DRIVERS OF INTERD ISCIPLlNAR y RESEARCH 27Research is truly interdisciplinary when it is not just pasting two di sc i

plines together to creare one product but rather is an integrarian and synthes is of ideas and methods. An example is the current exploration of stringtheory by theoretical phys icists and mathematicians , in which the questionsposed have brought fundamental new insights both to mathematicians andro physicists.

Convocation QuateInterdisciplinary research by definition requires the researchers to learn theother discipline. I like to stress vocabulary, but also methodology; I feel verystrongly about it.

Ruzena Bajcsy, director of the Center for Information TechnologyResearch in the Interest of Society, University of California, Berkeley

Other terms used inelude borrowing and multidisciplinary research. Borrowing describes the use of one discipline's methods, sk ills, ortheories in a different discipline. A borrowed technique may be assimilatedso completely that it is no longer considered foreign, and it may transformpractice without being considered interdisciplinary.1 An example of barrowing is the use of physical-science methods in biologic research, such aselec tron mi croscopy, x-ray crysta llography, and spectroscopy. Such borrow ing may be so extensive that the origin of the technique is obscured. ' For purposes of this discussion, multidiscip/inary research is takento mean research that inv olves more than a si ngle disc ipline in which each

discipline makes a separare contribl:1tion. Investigators may share facilitiesand research approaches while working separately on distinct aspects of aproblem.3 For example, an archaeological program might require the participation of a geologist in a role that is primarily supportive . Multidisciplinary

IKlein, J. T. "A Conceptual Vocabu lary of Interdjsciplinary Science," Practising lnter-disciplinarity. Eds. Weingarr , P. and Stehr, N. Universiry of Taranta Press, Taronto, 2000.pp. 3-24.

2See Holton, G., Chang, H. , and Jurkowirz, E. "How a scienrific discovery is made: A casehiswry." American Scientist, Vo l. 84, July-August 1996, pp. 364-75, for specific examples o fborrowing.

3Friedman, R. S. and Friedman, R. C. "Organized Research Unirs of Academe Revisited. "In Managing High Technology: An ln terdisciplinary Perspective. Eds., Mar, B. W., Newell,W. T. and Saxberg, B. O. Amsrerdam: Nonh Holland-Elsevier, 1985. pp. 75-91.


4/16

28 FACILlTATING INTERDISCIPLINA RY RESEARCHSTRUCTURE/POLlCIES

TABLE 2-1 Interdisciplinary Research StructuresAs a direct response ro ane component of es charge, "ldenrify and analyze currentstructucal models of interdisc iplinary rescarch," che cornmittee collected informaran 0 0abour 100 existing IDR groups and centers. The cornmittee rested che caregorizationproposed by Epton er al.a and found char, alrhough ir is largel y applicable, (here areimportant additional IDR strucrural categories and characteristics, including nationallabs, space alloearan, and fluidity of tcams.SMALL ACADEMIC 10 persans)

Bottom-up initiaran Research is primary; training is byproducr Loase management structure Many participants have disciplinary research commitments as well

LARGE ACADEMIC Bonom-up initiation, top-down incubarion and management Research and training components Management by directors who report directly to vice president for resea rch or

equivalent Tend to be permanent features: new building, instrumentation Some centers "co-hire" faculty, but faculty are affiliated with departments Space allocation: mix of permanent and "hote!" facilities

INDUSTRY Top-down, product-driven research Focused on research, not training Structured management Discrete time!ines and end points Fluid movement of researchers between teams

NATIONAL LABORATORIES Blend of top-down, mission-driven research and bottom-up initiation

Research and training components Structured management Discrete timelines and end points Fluid movement of researchers between teams

INTERINDUSTRY, INTERUNIVERSITY, UNIVERSITY-INDUSTRYTop -down, societal needs -driven research (can be basic and applied)

Research and training components Part-time directors with advisory boards Often initiated with large starting grants (such as National Science Foundation

funded Science and Technology Centers and Engineering Research Centers ) Except for seed grants, faculty must provide own grant money Programs may offer an "immersion" IDR opportunity

dEpton, S. R., Payne, R. L., and Pearson, A. W. (1985) "Contextual Issues in Managing,. Cross-Disciplinary Research." In Managing High Technology: An Interdisciplinary Perspec-tive. Eds. Mar, B. W., Newell, W. T. and Saxberg, B. O. New York: Elsevier. pp. 209-29.


5/16

THE DRIVERS OF INTERDISCIPLINARY RESEARCH 29A \ / AB / \ BA eB

Multidisciplinary:Join together towork on common problem,split apart unchangedwhen work is done.

Interdisciplinary:Join together towork on common question or problem.Interaction may lorge a newresearch lield or discipline.

FIGURE 2-1 Difference between multi- and interdisciplinary.SOURCE: Adapted froro L. Tabak, Director, NINDS, NIH. Presentation at Convocarian 0 0 Facilitating Interdisciplinary Research, Washington, D.C., January 29,2004.

research ofren refers ro effarts thar are additive but nor necessarily integrative (see Figure 2-1 )4,5IDR can also be described in terms of modes of participation. In one

mode an individual investigarar masters and integrares several fields. Theinvestigarar may conceive a new problem oc methad or may venture farenough from his or her original discipline to create a new field. For example, Albert Einstein ventured from his field of physics into Riemanngeometry to describe his new General Theory of Relativity.In a second mode, a group of investigarars, each with mastery in anefield, leam to communicate and coHaborate on a single problem.' In sornecases, such groups may be quite large, as in high-energy physics and genom

ies research.

4Porter, A. L. and Rossini, F. A. "Multiskll Research ," Knowledge: Creation, Diffusion,UtiJizaton , Vol. 7, No. 3, March 1986, p. 219.

SK lein, J. T. Interdisciplinarity: History, Theory, and Practice . Detrait: Wayne State University Press, 1990, p. 56.61n ane formulatan, this made is termed consilience: the "jumping together of knowledge"across disciplines "ro create a camman groundwork of explanarian". Wilson, E. O. Con-silience: The Unity of Science, New York: Alfred A. Knopf, 1998, p. 8.


6/16

30 FACILlTATING lNTERD lSCIPLlNARY RESEARCH

Convocation QuoteIf you think of disciplines as organs, Irue interdisciplinarity is something likeblood. It Aows. It is a liquid . It is not contained. There is no inside andoutside.

Alice GoHlieb , professor of medicine and di rectar, Clinical ReseorchCenter at the Robert wood Johnson Medical School

The cornmittee paid special attention to interd isciplinary education,viewing it as a central component of IDR. Students are prepared for thecomplexities of IDR when they are encouraged ro understand and I'ursuemultiple disciplines and to address complex problems from the perspectiveof multiple fields in their undergraduate and graduate stud ies. Specificsuggestions for strengthening interdisc iplinary education are presented inChapters 4, 5, and 9.

CHALLENGES DRIVING INTERDISCIPLINARY RESEARCHTo understand the natural world, sc ientists are drawn toward the un

known , especially toward the "grand challenges" of research. How did theuniverse originate? What physical processes control c1imate? What is thecarrying capacity of the biosphere? Such challenges almost a lways invitejourneys across disciplinary frontiers. A sc ientist may respond ro manykinds of motivation, or "drivers," in undertaking interdisciplinary projects.We Iist four such drivers below, providing examples and exploring why thepractice of modern science and eng ineering requires interdisciplinary work.

The Inherent Complexity of Nature and SocietyHuman society in ts natural setting contends with enormously com-plex systems that are influenced by myriad forces. It is not possible to study

rhe earth's clima te , for example, without considering the oceans, rivers, seaice, atmospheric consrituents, solar radiation , transport processes, land-use, land-cover, and other anthropogenic practices and the feedback mechanisms that link this "system of subsystems" across sea les of space and time.A full predictive or even descript ive understanding requires the use of manydisciplines (see Box 2-1 l.

Nature's complexity often leads ro surprises that require much thoughtand experimentation to unravel. An exa mple is the unexpected emergen ceof the Antarctic ozone hole in the austral springtime, a phenomenon foundto be the consequence of complex chemical and dynamic pathways attributable to the use of chl orine- and bromine-bearing compounds in commercial


7/16

THE DRIVERS OF INTERDISCIPLlNAR y RESEARCH

EVOLUTIONBOX 2-1 The International Geosphere-BiosphereProgram (IGBP)

The real connections tha! link the geosphere and biosphere to each otherare subtle, complex, and aften synergistic; their study transcends Ihe bounds ofspecialized, scientific disciplines and Ihe $COpe of limited, natienal scientific endeavars. For these reasans progress in fundamental areas of ocean-atmosphereinteractions, biogeochemical cycles, and solar-terrestrial relationships has comefar more slowly han in specialized fjelds, in spile of the obvious practical imparlance of such sludas. If, however, we could launch a cooperativa interdisciplnaryprogram in the earth sciences, on an inlernational seale, we might hopa lo take amajor step toward revealing the physical, chemical, and biological workings 01 theSun-Earth system and the mysteries 01 Ihe origins and su/Vival 01 life in the biosphere. The concept 01 an International Geosphere-Biosphere Program (IGBP),as outlined in this report, calls for this sort 01 bold " h o l i s t i c ~ ventura in organizedrasearch-the study 01 whoJe systems 01 interdisciplinary science in an effort tounderstand global changes in the terrestrial environment and ils living systems.aSo begns the prelace to a 1983 workshop report that would help lo launchthe IGBP, which 20 years later is one 01 the largest interdisciplinary international

research efforts ever undertaken. In lts origns, the program reflected all the majordrivers 01 IDA. It begins with the complexity of nature, the interactions between theland mass, the oceans 01 air and water, and the lite torms of Earth. II finds thatmuch of the most exciting science lakes place on the boundaries 01 both systemsand disciplines, such as the biogeochemical flows 01 the major life-support elements. Encouraging such explorations are powerful socetal needs to understandhow humankind is transforming the earth and the threats and opportunities thatsuch transformation poses. Making possible such ambition are generative technologies, particularly computer simulation and modeling, remote sensing trom space,and recovering the past trom cares of oeean bottom, ice, lakes, and trees.In scale, the program refleets both big scienee and local investigation. Sorne10,000 scientists in 80 eountries and fllore than 20 disciplines take part in IGBPseientific activites.b They nclude agricultural scientists, archaeologists, atmospheric chemists, and dynamicists, biologists, climatologists, ecologisls, economists,environmental historians, geographers, geologisls, hydrologists, mathematicians,meteorologists, planl physiologists, political scientists, physical and chemicaloceanographers, remote sensing scientists, and sociologists.

The program has transformed the disciplines initially involved. Disciplines thatwere primarily focused on local and sma!! scales, such as ecology, now addresslarge-seale processes and conduct extensive experiments including in situ carbonenrichment and experimental deforestation. Disciplines that were primarily curiosity-driven such as the many paleosciences, have acquired important societal 'relevanee. Natural and social sciences have come lo need and value each other.

continuasaFriedman, H. Prelace. Toward an International Geosphere-Biospher9 Program: A Study

of Global Change. Repar! of a National Research Council Workshop, Woods Hole, MA July25-29, 1983. Washington: National Academy Press, p. viL

bFor these details and insights we are gratelul lO WiU Steffen, Executive Director of theIGBP.

31


8/16

32 FACIL/TATING INTERDISCIPL/NARY RESEARCH

BOX 2-1 ContlnuedDisciplines have discovered camman nterests, such as how to relate wholes toparts, macro processes to micro behavior, and global lo local. Indeed, globalchange seienee now exhibits many interdisciplinary aspects, with a secend generalan 01 scientists Iranscending their disciplines and schooled in problem-drivencamman knowledge.

But mast important are the major scientific findings. The program has transformed our understanding 01 both nature and humankind. A recent summary volumec finds that:

The earth is a system that life ilself helps to modulate. 8iol09ical processesnterael with chemical and physical processes to ereate the planetary environment. Human activities are influencing the functioning of the earth system inmany important ways.

The earth is operating in a no-analogue state. The magnitudes and rates ofchanges occurring simultaneously in the earth system are unprecedented. The earth's dynamics are characterized by critical thresholds and abrupt

changes.CSteffen, W., Sanderson, A. , Tyson, P. , Jager, J .. Matson, P Moore 111, B., Oldfield, F.,Richardson, K., Schellnhuber. H.J., Tumer 11, B. L. , Wasson, R. Global Changa and the Earth

System: A Planet Under Pressure. IGBP Global Change Series.New York: Springer-Verlag,SerUn Heidelburg, 2004, 336 pp.

produets. Pinpointing that eause required the eombined efforts of manysc ientific and technical disciplines; solving the problem itself required thecollaboration of physical scientists, engineers, economists, and social scientists.Similarly, the human-genome mapping project was a complex undertaking that depended on extensive collaboration across many fields, including the biological and computat ional sc iences. Basic ques tions of life-howliving beings grow, how the brain functions, why many animals need rosleep, how retroviruses function-share the characteristic of complexity,and understanding them, even in part, depends on multiple disciplines.Gaining sueh understanding will almost certainly require deep expertiseboth at the subsystem leve l and at the interdisciplinary leve l-and the integration of these two levels. Ir is important to note that depth in researeh isnot confined to single-discipline inves tigations. Statistical mechanics, forexample, unites physicists and mathematicians in studies of substantialdepth7If science and enginee ring deal with extremely complex systems, thesame is true for studies of human society. How human societies evolve,

7Kafaros and Eisner, ibid. p. 1257.


9/16

THE DRIVERS OF INTERDISCIPLlNARY RESEARCH 33make decisions, interacr, and solve problems are all matters that call fordiverse insights. Very fundamental questions are inherently complexoForexample, why do humans kili each other? Why does hunger persist in aworld of plenty? Answering such questions successfully requires collabora-rio n across che natural sciences, social sciences, and humanities .

The Orive to Explore Ba sic Research Problemsat the Interfaces of Disciplines

Sorne of che most interesting sc ien rific questions are found at che inter-faces between disciplines and in the white spaces on organi za tional charts.Exploring such interfaces and interstices leads inves tigators beyond theirown disciplines ro invite che participaran of researchers in adjacent orcomplementary fields and even to srimulate the development of a newinterdisciplinary field. Examples inc\ude the fo llowing:

Biochemistry was long ago considered an interdisciplinary activity;today ir has departmental, prograrn, or similar structural statu s in mos!major universities . The field of cogn itive science has evo lved in respo nse ro questionsthat could not be answered by single disciplines. Today the Cognitive Sci-ence Society embraces anthropology, art ificial intelligence, neuroscience,education, lingu istics, psychology, and philosophy. ' As biology has become more quantitative, its points of overlapwirh [he mathematical sciences and che physical sciences have become morenumerous and important. Today, the computa tional and statistical powerof mathematics and the research facilities of the physical sc iences are re-quired for mak ing sense of, for example, genomics, proteomics, epidemiol-ogy, structural biology, and ecology. Ecology and economics (and other social sciences) have a commonorigin , at least in name, and, in creasingly, a common fi eld --eco logic eco-nomics- th at aspires to facilitate "undersranding between economists andecologists and the integrarion of their thinking" with the goal of developinga sustainable world.9

That many of the most interesting scientific questions are lodged in theinrerstices between disciplines can also be seen in various acriviries thathonor outstanding creativity. For example, a lthough the MacArthur Foun-dation fellow awards are not given on the basis of interdiscip linarity, to

8See Appendix D on {he development of disciplinary sociecies.9The Web site of che International Society of Ecological Economics is http://www.

ecologicafeconomics. org.


10/16

34 FACIUTATING INTERDISCIPLINARY RESEARCHjudge from brief biographies, two-thirds to three-fourths of MaeArthurfell ows in seienee appear to work in interdiseiplinary fields.

The Need to Solve Soeietal ProblemsHuman society depends more [han ever on sound science foc sounddeeision making. The fabrie of modern life-its food, water, seeurity, jobs,energy, and transportation-is held together largely by teehniques and tools

of seienee and teehnology. But the applieation of teehnologies to en hancethe quality of life can itself ereate problems that require teehnologieal solutions. Examples inelude the buildup of greenhouse gases (henee globalwarming), the use of artificial fertilize rs (water pollution and eutrophiearion ), nuclear-power generaran (radioacrive waste), and automotive [cans-portation (highway deaths, urban sprawl, and air pollution).

EVOLUTIONBOX 2-2 The Developmenl 01 Mlcrowave Radar alMIT's Radlallon LaboraloryB

The development 61 radar (radio detection and ranging) during the 19405 waslargely accelerated by military needs in Wortd War 11. Members of the scientificcommunity recognized the value of radar lo the war effort. In the United Slates, theetfort to expand microwave radar capabilities was concentrated al MIT's RadiationLaboratory, which was staffed by civilian and academic scientists in many disciplines. Projects ncluded physical electronics, microwave physics, electromagneticpraperties of matter, and microwave cammunicalion principies.The "Rad Lab" was responsible tor almosl half Ihe radar deployed in WorldWar 11 and al one point employed almost 4,000 people working on several continenls in government, industrial, and university laborataries. What began as aBrltishAmerican effort to make microwave radar work evolved into a centralizedlaboratory committed lo understanding the theories behind experimental radarwhile solving lts engineering problems.The Rad Lab was formally shut down afler Ihe end of World War 11 in 1945,but in 1946 the Basic Research Division was incorporated into the new ResearchLaboratory 01 Electronics al MIT. Research continued on problems in physicalelectronics and micrawave physcs. Modern techniques were applied to physicsand engineering research, and engineering applications were emphasized in microwave communication.8MIT Radiation l aboratory series Volume 28. Ed. Henney, K. Avai lable al http. llwww.

brewbooks.com/reflrfireCradlab_v28.htmt, G.Goebel, Microwave Radar & The MIT Rad lab.Available at http.l/www.vectorsite.netlttwiz3 .htmf.LabsMicrowave Traditions al RLE . ALEcu rrents, Vol. 4, No. 2-$pring 1991. Available at hftp:/ lrfeweb.mit.edu/radlablradlab.HTM.


11/16

THE DRIVERS OF INTERDISClPLlNARY RESEARCH 35An indication 01 interdisciplinariry in response ro societal needs is the

success oE large, sustained endeavors, many DE which continue ro chis day.During World War JI, lor example, sc ience and engineering demonstraredrhe abil ity ro srrengrhen military power rapidly (see Box 2-2). The 3-yearManhanan Project (1942-1945) ro deve lop an aromic bomb was an interdisciplinary eflore requiring resea rchers Irom many lields and sublields 01science and engineering, from (h e wide sweep of chemisrry and physics roche specific skills of uranium refinemenr , sorape separaria n, plutoniumpurificarian, nuclear decay measurement, nuclear-waste dispasal, and radiation biology.Anorher example is rhe National Ca ncer Acr, signed by Pres ident Nixonin 1971. The acr au th orized an inrerdisciplinary research eflore involving ava sr sweep 01 biomedical disciplines, lram ge netics and ce ll biology thraughc1inical care, bioerhics, and biosrarist ics. Caneee research has alwa ys beeoamong rhe mosr interdisciplinary 01 lields, mirraring the complexity 01 themany diseases ir addresses.Researchers continue to apply the 20th century's revolutionary geneticinsights to unravel the structures and lunctions of proteins (see Box 2-3).This investigarian influences every as pecr DE (h e life sciences, at everylevel, from molec ular arrangements ro clnical, poputarian, and ecologicstudies. 1O

The Stimulus 01 Generarive TechnologiesGenerative technologies are those whose nove!ry and power not onlylind applicarions 01 grear value but also have the capacity to transform

existing disciplines and generare new ones. An early momenrous examplewas the use of microscopes by Ho.oke and van Leeuwenhoek ro view "cubieles," or cells, in animal and plam bod ies and ro make ir possible ro seeliving "an imalcules" (bac teria) wirh rheir own eyes-borh critical stepsalong the path ro modern molecular biology.A recent example of a generarive tec hnology has been rhe developmenr01 rhe Internet, whose popular form is only about 10 years old. The Internet

10Yet another example ca n he found io Brao scomh, L, Holroo, G., and Sonn err, G. Cutting-edge Bas;c Research in the Serv;ce of Publie Ob;eet;ves: A Blueprint for an Intellectually Boldand Socially Beneficial Science Po/iey. Consortium for Sc ience Po liey Ourcomes, ArizonaSu te Univers iry, Ma y 2001. Avai lable online at http://www.cspo .org/products/reports/scienceforsociety.pdf (Based 00 a workshop sponsored by the David and Lucile PackardFoundation aod [he Alfred P. Sloao Fouodation.) The report makes the case for use-inspiredor "Jeffersonian" ha sic rcsearch and ineludes a mas ter list of qu estioos in science aod technology, most of which requ ire interdisciplinary approaches. Holton, G., "Whar Kin ds ofSc ience are Worrh Supporting?" The Grear Ideas Today, Encyclopcdia Britannica, Chicago,1998.


12/16

36 FA CIL lTATING INTERDISC IPLINA RY RESEARCH

EVOLUTIONBOX 2-3 Prolein Slruclure Delerminalion UslngX-Ray CrystallographyB,b

The knowledge of protein structures is criticar lo fighting disease with drugs.In recenl years, the development of new techniques to determine protein structure,combinad with rapid improvement in computer technology, has allowed proteinstructure determination lo proceed al arate that is keeping pace with advances inbiomedical sclenC9. In the case of x-ray crystallography, ils development and wideuse in p r o t e i n ~ s t r u c t u r e determination-which spanned a century-began with noknowledge ot ilS valua for biomedicine.X rays were firsl discovered in 1895, and the diffraction 01 x rays byelectronsin crystals was firsl demonstrated in 1912. In the 19305, x rays were aimed atcrystals of biological molecules, but il was not until Perutz and Kendrew determinad Ihe molecular structure of hemoglobjn and myoglobin in 1960 that the valueof x-ray crystallography in protejn science was realizad. In the 1970s, synchrotronradiation (see Box 2-5) was harnessed as a source of x rays for protein crystallography, and the 1990s saw a great increase in the number of protein structuresdetermined with this technique. Research to develop the technology was an interdisciplinary endeavor. Its long-term nature should remind those who facilitate IDRthat support 01 basic research can often have payoffs that are not immediatelyvisible and are often outside the field in which theywere initially envisioned.

aDUI, K. Strengthening Biomedicine's Roots. Nature22 400:309-310. July 1999.bJ-iistory of X-ray Crystallography and Associated Topies. Available al htfp: lwww.dr.ac.uk/SRSlPX/h;story/history.hfmf.

has both enhanced connectivity between people and revolurionized acce ssto information, transforming the ability to interaet and co llaborate acrossspace and rime. Ir has special relevance ro rh e world of research, for whichit offers ways ro work in large, disrribured teams, enlarge rhe educationalenterprise, provide access to data on time and spatial scales never possiblebefore, and design powerful new tools lo transform the processes of discovery, learning, and communication (see Box 2-4 ).Dramatic declines in the cost of processing, storing, and rransmirringinfarmation a re transforming science and engineering discip lin es. Sorneexpens have ca lled on rhe National Sc ience Foundalion and orh er sc ienceagencies ro launch a bold new iniriarive in cy ber infrasrrucrure, which wouldplay Ihe same role in supporting the knowledge economy rhar roads, powergrids, and rail lines have played in supporting the industrial econorny.l1

11Revolutionizing Science and Engineering through Cyberinfrastructure. Report of the National Science Foundation Advisory Panel on Cyberinfrastructure. February 3, 2003. Available a l http://www.cise. Ils f.gov/scilreportsltoc.c{m.


13/16

THE DRIVERS Of INTERDISCIPLlNARY RESEARCH

INNOVATIVE PRACTICEBOX 2-4 The Knowledge and Dislribuled Inlelligence (KDI)Funding Initialive

The rise in computar power and connectivity i5 reshaping relationships amongpeople and organizations and transforming the processes 01 discovery, learning,and communicatio.n . The knowledge and distributed intelligence (KDI) funding initiativa at the National Science Foundation (NSF) was created in 1998 to find waysto model and make use 01 complex and cross-disciplinary scientific data.a KDIsupported interdisciplinary projects 01 individuals or groups that took advantage 01changes in how research was being done, such as ncreases in computing powerand connectivity among researchers. The nitial solicitation had three foei 01research: knowledge networking, learning and intelligent systems, and new computational challenges. The KDI iniliative has sponsored research that analyzesliving and engineered syslems in new ways, and il encourages investigators toexplore Ihe cognitive, elhical, educational, legal, and social implications of newIypes of learning, knowledge, and interaclivity.A program assessment was carried out in 2002. NSF recognized thal metricshave to be developed that match the goals 01 the researeh programo To Ihat end,KDI granlees were invited to a workshop lo determine how prajeets were organized and managed, to identify the projecls, outeomes, and lo eatalog suggestionsIhat might help future grantees in their exeeulion of KDI-sponsored projects.The evaluationb,c provides interesting infarmalion about tools, researeh direetions, outreach, and student training. Management 01 collaborative and mullidiseiplinary researeh projects was a substantive issue. Project success dependedlargely on eaordinating interaetions among researehers. Dispersion 01 participants,ralher than interdiseiplinarity, was the most problematic aspect of KDI projeets.Projeets with principal investigators in multiple universities were subslanlially lesswell eoordinated and reported lewer favorable outeomes. Project-related conterenees, workshops, and other regular meetings appeared lo reduce Ihe adverseeffects 01 dispersion. The assessment identilied a number 01 needs for furthersupport, including management lools' that would increase the ease with whiehprojeet participants inleract aver the lifelime ot Ihe project.

a"fhe original KDI solicitation is available a: http://www.nsf.gov/pubs/19981nsf9855/nsf9855.pdf.bCummings, J. and Kiesler, S. (2004) KDllniliative: Multidisciplinary Scienlific Collaboratons. NSF Report. Available on the NSF KOI Home Page: http://www.cise.nsf.gov/kdi/about.html. Cfaking stock of the KOI: Science of Evaluation. http://www.cise.nsf.gov/kdi/eval.html.

37

This cyberinfrastructure might be composed of distributed, high-performanee campllters, online scientific instrllments and sensor arrays, mllltidisciplinary collections of scientific data, software toolkits for modelingand interactive visualization, and toals that enable close collaboration byphysically distributed reams of researchers (see Box 2-5 and Box 9-7).


14/16

38 FA C ILlTATING IN TERDISCIPLlNARY RESEARCH

EVO LUTIONBOX 2-5 Tool-Driven Inlerdlselplinary Researeh:The Advaneed Pholon Souree (APS) alArgonne Nallonal Laboralory

The Advanced Photon Source (APS) at Arganne Natlonal Laboratory is anatianal synchrotronradiation light-source research facility. Commissioned in 1995,the APS is funded by the US Department 01 Energy, Office of Science, Office 01Basic Energy Sciences.a Members 01 the intemational research community usehigh-brilliance x-ray b e a m s f r o m the APS to carry out basic and applied researchin materials science; biology; physics; chemistry; environmental , geophysical , andplanetary science; and innovative x-ray instrumentation.Aesearchers come lo the APS as members 01 col1aborative access teams(CATs) or as ndependent investigators. CATs comprise larga numbers of investigators with common research objectives and are responsible for design, construclion, funding, and operalion 01 beamlines al the facility. CATs must allocate 25pereent 01 their x-ray beam time to independent invesligators or groups not affmat-ed with CATs.The APS was designed to accommodale up lo 32 CATs , of which over 20 arein operation. One 01 the interdisciplinary industry-university collaborations established to take advantage of APS resources ls the University 01 Michigan-HowardUniversity-AT&T BeU Laboratories (MHATT) CAT, formed in 1989. MHATT-CATstudies range from baslc protein dynamics to the behavior of solid-slate lasers.According to one ofthe directors of the MHATI-CAT, University ofMichigan Physics Professor Roy Clarke, "a very important part of the projeet is lo establish highspeed communicalions that link participating institutions and the facility al the APS ,so that our sludents, particularly our undergraduate researchers, can participatea.;:tively in the research while attending classes on their respective campuses."bOthers CATs are run by university or industry leams. To enhanee communication among and between leams, the APS Web site provides a linked list 01 CATsand offers a listserver for nter-CAT communication. The APS Web site also lisismeetings 01 interest lo lacility users and highlights recent research by posting ab-stracts and figures on lts home paga.

aAPS: Advanced Photon $ource al ANL. Home page: http://epics.aps.anl.gov/aps.php.~ I g a J .R. (1994) Ctarke co-directs project at Argonne ph oton facitity. Th e Unlversily

Record. Universily of Mlchlgan. March 28 , 1994. Acces sed March 29. 2004 at http://www.umich.edu/- urecord/9394/Mar28_9412.htm.

Ad voeates of eyberinfrastrueture believe that it will allow a growingnumber of researehers ro colleet , proeess , analyze, and make ava ilable volumes of information that trigger shifts in the kinds of seientifie questionsthat can be pursued; simulate systems of greater eomplexity and importance ; and more easily work across scientific disciplines. For exa mple, theNational Se ienee Foundation has funded a "National Virtual Observa-


15/16

THE DRIVERS OF INTERDISCIPLlNARY RESEARCH 39tory"12 that is likely ro transform asrronomy. Within a few years, comprehensive sky surveys will be generating petabyres (quadrillions of bytes) ofdata every year. The long-term goal is ro make this data available ro everyresearcher, along with the databases, data mining algorithms, and visualizarian toals needed ro make sense of ir. Researchers believe thar this informarian abundance willlead to qualitatively new science, such as statisticalastronomy chal analyzes the large-scale structure of rhe universe, and automated searches far e;mtic or previously unknown types of celestial objects.

Magnetic resonance imaging (MRI) is another example of a generativetechnology. The Nobel Prize in medicine and physiology for 2003 wasawarded to chemists Paul Lauterbur and Peter Mansfield ro honor theirwark that led to MRI. Their research grew out of a fundamental interest inusing che magnetic resonance effee! ro produce images in proton-containingmatter. MRI and positron-emission romography (PET), with ancillary math-emarical advances in romographic analysis, ha ve revolutionized many aspecrs of medical diagnosis and opened opportunities for safe experimentation with human subjects in the cognitive sciences.13

CONCLUSIONSThe potemial power of IDR to produce novel and even revolutionaryinsights is generally accepted. Ultimately, however, the value of IDR to the

scientific enterprise depends on the extent to which individual researchersare free to engage in it. IDR musr be not only possible but also attractive forstudents, postdoctoral fellows, and faculty members.

FINDINGSInterdisciplinary research (IDR) is a mode of research by teams orindividuals that integrates information, data, techniques, tools, perspectives, concepts, and/or theories from two or more disciplines orbodies of specialized knowledge to advance fundamental understand-ing or to solve problems whose solutions are beyond the scope of asingle discipline or area of research practice.

12US Narional Virtual Observatory. http://www.us-vo.org/.131n one view, "new technologies are now driving scienrific advances as much as rhe orherway around. These technologies are enabling novel approaches ro old quesrions and are

posing brand-new ones." Leshner, A. 1. "Science at the leading edge," Science Vol. 303:729.Feb. 6, 2004.


16/16

40 FACILlTATlNG INTERDISC IPLlNARY RESEARCH

"1 CA>l RV1l:Med< WI\W ALL "" p WAj 5 o M U > ~ "'1\0CO\110 CARIt. MD ~ w\tO' c.ovlD s e : ~ : \

IDR is pluralistic in method and focus . It may be conducted by individuals or groups and may be driven by scientific curiosity or practicalneeds.Interdisciplinary thinking is rapidly becoming an integral feature ofresearch as a result of four po:-verful "drivers": the inherent complexiryof nature and society, the desire to explore problems and questions thatare not confined to a single discipline, the need to solve societal problems, and the power of new technologies.Social-science research has not yet fully e\ucidated the complex socialand intellectual processes that make for successful IDR. A deeper understanding of these processes will further enhance the prospects forcreation and management of successful IDR programs.

the drivers of interdisciplinary research

Documents