EmergingTechnologies10-31-06

Emerging E-Learning Technologies: Tools for Developing Innovative Online Training

Gary Woodill, Ed.D. Senior Researcher

2006, Brandon Hall Research. All rights reserved. Do not reproduce.

PrefaceIn a 2000 report on e-learning, Trace Urdan and Cornelia Weggan divided the corporate e-learning universe into the sectors of content, technology, and services. I use this tri-part division of the e-learning field to structure a series of three inter-related reports for Brandon Hall Research. The first report, entitled Emerging ELearning: New Approaches to Delivering Engaging Online Learning Content was published as an e-book in December 2005. In it I identified 50 new content formats for e-learning that are now emerging to replace the page-turner models of online pedagogy so prevalent in the first few years of online learning. This report, the second in the series, focuses on emerging innovative technologies for e-learning. In it you will find descriptions of 52 technologies that will have a major impact on elearning over the next five years. I have provided links to online learning examples, lists of online resources, and a bibliography for each of the technologies. A list of companies and organizations that are developing and using these technologies is provided at the end of the report, along with an index. A third report, which focuses on the extensive variety of emerging services that support e-learning, will be available in early 2007. This series of three inter-related reports started with Brandon Hall and Richard Nantel perceiving that the field of e-learning is currently undergoing significant change. They asked me to research and report on these trends and changes, and I thank them for their support and encouragement. My colleagues at Operitel Corporation, where I served as Chief Learning Officer until recently, have supported this research from the beginning. A special thanks to Michael Skinner, Operitels CEO, and the rest of the Operitel management team David Fell, Carlos Oliveira, and Jason Stimers for making my work environment such a positive place to produce this kind of research. Pamela Fragomeli, Lise Bye, Grant Hamilton, Amy Davey, Pierre Cahorn, Jennifer McDowell, and Dan Medakovic were very supportive team mates who allowed me to be more productive. Thanks also to Chad Nolan for checking all the hyperlinks and addresses throughout the report and to Chris Downs for copy-editing the manuscript. None of this would have been possible if my wife, Karen Anderson, had not encouraged me to get into the e-learning field way back in 1992, and had not accompanied me on this journey with input, editing, and support at all points of my career. Thanks, Karen.

Statement of IndependenceBrandon Hall Research Is Independent, Unbiased, and Objective Brandon Hall Research reports and online services are independently written and edited. Brandon Hall Research makes money from publications and online services only by selling research reports and services to the public. Vendors and featured organizations do not pay to be included in any reports or services. Brandon Hall Research does not charge vendors and featured organizations to quote from reports or services for press releases. Brandon Hall Research accepts no advertising or sponsorship of reports or services. Vendors and featured organizations are often asked to complete extensive request for information questionnaires and may be interviewed for inclusion in reports or online services. Once submitted, Brandon Hall Research has editorial control and final approval over the content. Brandon Hall Research emphasizes factual non-marketing-type information in our research reports and online services. Brandon Hall Research does not benefit in any way by the sale of any product included in our research reports or online services. Brandon Hall Research does not provide leads to vendors or assist them in any way in making sales. Brandon Hall Research does respond to requests for consulting from both user organizations and vendor organizations based on the contents of reports and online services. In those cases, the only benefit to Brandon Hall Research is the consulting fee, which is the same rate for user or vendor organizations.

Table of ContentsPart I: E-Learning Architectures and Frameworks Part II: Emerging e-Learning TechnologiesAffective Computing Agents Animation Software Artificial Intelligence Assessment Tools Audio and Podcasting Tools Authoring Tools Avatars Blogs Browsers Classroom Response Systems Collaboration Tools Communications Tools Competency Tracking Software Content Management Systems Data Mining Decision Support Software Displays E-Portfolio Tools Gaming Development Tools Gesture and Facial Recognition Graphics Tools Haptics Interface Devices Learning Management Systems Learning Objects and Repositories Location Based Technologies Mashups, SOAP and Web Services Metadata, Ontologies and Taxonomies Mobile Devices Natural Language Processing Peer to Peer Technologies Personal Learning Environments Personalization Software 12 17 24 28 45 54 59 69 72 76 78 81 97 100 104 107 111 114 117 120 125 128 131 134 138 145 151 155 158 161 167 170 174 176

1 11

Portals Presentation Tools Rapid e-Learning Tools Robotics Search Engines Semantic Web Simulation Tools Smart Labels and Tags Social Bookmarking Social Networking Telepresence Technologies Video and IPTV Virtual Reality Visualization Technologies VoIP and Telephony Wearable Computing Web Feeds Wiki Tools

184 187 192 196 199 209 215 221 224 230 237 241 246 250 259 261 264 268

Part III: Innovation in E-Learning Part IV: List of Companies and Organizations Index

272 277 334

Part I: E-Learning Architectures and FrameworksA Brief History of Learning Technologies Just what are learning technologies? We tend to think of them as the latest wave of computer-based technologies that present educational materials and online assessments to learners sitting in front of a computer. But learning technologies have a much longer history. Humans developed technologies for learning well before the advent of writing over 5000 years ago. Technologies are any technique, material, or device that extends human abilities. Examples from early humans include tools for hunting and fishing, fire for warmth and cooking, marks and other techniques for signaling direction, and language for communication. Educational or learning technologies are anything that extends our ability to teach and learn. They are, to quote Donald Norman (1993), things that make us smart. Some of the earliest learning technologies include marks on a path to indicate directions or danger, oral techniques that are passed on from generation to generation in cultures where speaking is the primary means of transmitting knowledge, early cave drawings that told stories of hunting and warfare, and clay tablets onto which symbols were pressed as the earliest forms of writing. We tend to forget, for example, that lecturing, still used in many institutions of higher education, is a 2000 year old technology. The word comes from the Latin lectura, or reader. Starting in the first century AD, various Christian orders copied manuscripts by hand, making multiple copies by having a reader dictate from a manuscript while others copied it word for word. Medieval universities followed the same practice, as books were scarce. The invention of the printing press by Johannes Gutenberg in 1440 allowed a much wider distribution of knowledge, from a few literate clergy to a much wider group of educated readers. The introduction of textbooks as a technology for teaching was the result of mass education movements in the late 19th century, coupled with faster printing and binding methods. Classrooms themselves can be seen as a form of technology. Schools did not always have classrooms that were organized as we know them today. The modern classroom, with rows, the raising of hands, class periods, detentions, and recess, was first introduced in Prussia (Germany) in the 1770s. With these reforms, the classroom became industrialized, similar to the organization of the burgeoning factories of that time. Learning became standardized and linear, with a principal aim of controlling the learner and the pace of his or her learning. A full history of learning technologies is beyond the scope of this research report. But I mention it in the context of arguing that the first versions of e-learning

Humans developed technologies for learning well before the advent of writing over 5000 years ago.

Do Not Reproduce

1

E-Learning Architectures and Frameworks

consisted of putting industrialized, linear teaching techniques online. Industrial-ized schooling is about the teacher presenting materials to the learner who must take them and prove that he or she has learned through the successful passing of examinations. There are, in fact, many different ways to learn. Perhaps the first way we learn is by imitation. Learning by imitation doesn't require teaching technologies - just the ability to follow an example. The second way we learn from an early age is by listening to stories and repeating them to others. Stories are ways of depositing our thoughts and memories outside of our physical bodies into the larger community. A third way to learn is by seeing. Given that 50 percent of our brainpower is devoted to vision, visualization with pictures and graphics is a powerful tool. The technologies of writing and reading extend our abilities by allowing thoughts to be expressed and received in words, even if the originator of the words is not present. Writing and reading were first developed in Mesopotamia (now Iraq) about 6000 years ago. This new technology was first used by accountants to track crops and inventories. Schools were started to teach accounting. Students in these schools learned by reading, memorizing, and reciting. Sadly, most of the early examples of online teaching still follow this 6000-yearold model. E-learning for many developers has been to simply place materials to read and look at on the

screen, followed by regurgitating this material through online multiple choice tests. This tell-test approach uses little of the possibilities of computer-based learning. E-learning is the latest technology in a long line of extensions of our ability to teach and learn. Like all new technologies, it has been introduced with extravagant claims (hype) of efficacy and efficiency. At the same time, all new technologies have detractors who worry that the new technologies will have a significant negative impact on current practices. Eventually, all new technologies become integrated with previous teaching and learning tools, changing the practice of teaching. When a new technology is introduced, there is a tendency to understand it in terms of what is already familiar. Examples include the horseless carriage (cars), moving pictures (films), and talking machines (phonographs or record players). Today's wireless networks will likely evolve into something without reference to wires. When a new technology first comes into use, it is common for people misunderstand its real impact. For example, in 1876 someone at Western Union, the main telegraph company in the world at the time, stated that this telephone has too many shortcomings to be seriously considered as a means of communication. Similarly, Thomas Watson, chairman of IBM, stated in 1943 that he could envision a world market for maybe five computers. In 1970, the early

Given that 50 percent of our brainpower is devoted to vision, visualization with pictures and graphics is a powerful tool.

2

2006, Brandon Hall Research

Emerging e-Learning Technologies


days of computer networking, AT&T was given what would become the Internet. The company returned it to the US Department of Defense after a six-month trial saying that it could find no commercial use for computer networking. The new computer-based learning technologies will have their greatest impact when we start to realize their unique advantages. Some of these advantages include the following: High speed computation Interactivity especially for games and simulations Networking with global reach, allowing worldwide collaboration Digital representations/transformations Algorithms repeatable procedures Storage and retrieval extending our memories Individualization/customization/ flexibility resulting in personalized content Constant availability - 24/7 Simulation of complex processes

This is different from reading a printed book or presentations on a screen; mashups can only be done using computer technology. Computers can be programmed and organized in many different ways. However, for one application or data set to work with other applications or data sets, it needs to adhere to architectures, frameworks, and standards. Architectures refer to the overall technical design of a computer system. Frameworks are overall design frameworks for implementing e-learning within a specific architecture. Standards or protocols refer to the design of systems so that they can communicate with each other. It is beyond the scope of this research report to explain the details of computer architectures, as this is a non-technical guide. But e-learning professionals need to be aware that the architecture of a system can limit or expand the possibilities of what can be done. An emerging architecture that is particularly relevant to emerging e-learning technologies is Service Oriented Architecture (SOA). Service Oriented Architectures Chatarji (2004) suggests that service oriented architectures offer the following advantages over traditional approaches to distributed computing: They offer business services across platforms.

To realize these advantages, we need to break from the page metaphor that has dominated the first decade of Web development (Alexander, 2006). The Web is about producing and distributing a variety of content formats. Rather than pages, we are beginning to speak about posts or streams of content, sometimes gathered from multiple sources, and then integrated into a unique online mix of information, sometimes referred to as a mashup (Woodill and Oliveira, 2006).

Rather than pages, we are beginning to speak about posts or streams of content, sometimes gathered from multiple sources, and then integrated into a unique online mix of information, sometimes referred to as a mashup.

Do Not Reproduce

3


They provide location independence. Services need not be conducted at a particular system or particular network. Links are based on loose couplings rather than tight a integration of programs. There is authentication and authorization support at every level. The search and connectivity to other services is dynamic.

The ultimate vision for service oriented architecture is to construct elearning resources in a grid, with access to an enormous variety of learning materials and programs.

Short-term benefits of implementing SOA include the following: Enhanced reliability Reduced hardware acquisition costs Existing development skills leveraged Accelerated movement to standards Provides a data bridge between incompatible technologies

Provides the ability to build composite applications Creates a self-healing infrastructure that reduces management costs Provides truly real-time, decisionmaking applications Enables the compilation of a unified taxonomy of information across an enterprise and its customer and partners Ability to more quickly meet customer demands Lower costs associated with acquiring and maintaining technology Managing business functionality closer to the business units Leverages existing investments in technology Reduces reliance on expensive custom development

Long-term benefits of implementing SOA include the following: Provides the ability to build composite applications Creates a self-healing infrastructure that reduces management costs Provides truly real-time decisionmaking applications Enables the compilation of a unified taxonomy of information across an enterprise and its customer and partners

Benefits from a business value perspective include the following:

The ultimate vision for service oriented architecture is to construct e-learning resources in a grid, with access to an enormous variety of learning materials and programs. Grid technologies define a new computing paradigm by making an analogy to the electric power grid. With applications and content becoming both distributed and interoperable, a learner should be able to plug in to the grid and remotely start any application and/or receive access to any content on the grid. (For a longer description of grid architecture in e-learning, see the call for papers in Learning Grid, Number 3, January, 2005. Find it at: http://kaleidoscope.grid.free.fr/publicatio n.php).

4




For example, the Access Grid is an ensemble of e-learning resources including multimedia large-format displays, presentation and interactive environments, and interfaces to grid middleware and visualization environments (See: http://www.accessgrid.org). E-Learning Frameworks and Standards Several efforts have been started to establish a formal framework for producing e-learning. In particular, the IMS Global Consortium (http://www.imsglobal.org) has a number of published documents that, taken together, could form the basis of a formal e-learning framework. In Europe, the E-Learning Framework (ELF) (http://www.elframework.org) is an international effort to establish a serviceorientated approach to developing and integrating computer systems in the sphere of learning, research, and education administration. Freisen and McGreal (2002) distinguish between e-learning standards and specifications. Standards are formally accepted definitions while specifications are less evolved and contain descriptions that often change over time. Major specifications for e-learning, according to both Freisen and McGreal (2002) and Neuman and Geys (2004), include the following:

Dublin Core The most broadly based metadata specification. http://purl.oclc.org/dc/ IMS Serves as a catalyst for developing instructional software. http://imsproject.com ARIADNE This group has created a European repository for pedagogical documents called the Knowledge Pool System. http://sourceforge.net/projects/ariadnek ps ADL SCORM Specifies the behavior and aggregation of modular, interactive learning components, and makes extensive use of XML. http://www.adlnet.org/Scorm/ IEEE LOM For metadata describing learning objects (LOs), enabling the search for content. http://ltsc.ieee.org/wg12/ AICC An older specification from the Aviation Industry CBT Committee (AICC) for run-time communication between content and learning environments. http://www.aicc.org/ The above Web sites show how the elearning industry is moving to develop a set of common viewpoints that will result in a greater interoperability within the industry. At the same time, relentless change and new innovative technologies make this task difficult.

the e-learning industry is moving to develop a set of common viewpoints that will result in a greater interoperability within the industry. At the same time, relentless change and new innovative technologies make this task difficult.

Do Not Reproduce

5


From Push to Pull in E-Learning For those of us who have been in the business of teaching for a long time (I started in 1971), perhaps the hardest shift is to think of teaching as providing educational resources rather than just instruction. The world is moving away from the model of a teacher as a container of valuable information to be disseminated to learners. Instead, the new model of teaching involves facilitation. Teachers facilitate learners to find what they need to construct their own answers to problems and issues in life. This is especially true for adult education. This theme is found in two recent publications on the shift in e-learning from push to pull. In late 2005, John Hagel and John Seely Brown placed a working paper on the Web entitled From Push to Pull -- Emerging Models for Mobilizing Resources. They noted that in education, we design standard curricula to expose students to codified information in a pre-determined sequence of experiences. In business, we build highly automated plants or service platforms supported by standardized processes seeking to deliver resources to the right place at pre-determined times. The problem with standardized procedures in education and training is that they do not work well in times of rapid change and uncertainty. Rather, what is needed to succeed is the ability to mobilize appropriate resources when the need arises. David Bolliers 2005 report for the Aspen Institute, When Push Comes to Pull: The New Economy and Culture of Networking Technology, reinforces this theme. Bollier says, [a] pull economy - the kind of economy that appears to be materializing in online environments - is based on open, flexible production platforms that use networking technologies to orchestrate a broad range of resources. The trend in e-learning is also to move from push to pull in terms of instructional design of content. Instead of just providing courses, access to a wide range of documents and other online resources needs to be facilitated, along with teaching appropriate search and evaluation strategies. While packaged courses still have a place, the ability to find both human and information resources at a moments notice to resolve an issue has now become a competitive advantage. Moreover, the content that arrives after making a request is becoming more personalized (Werkhoven, 2004). We are now in the era of providing for the long tail, the infrequent requests and desires of many individuals. As Anderson (2004, 2006) has argued, servicing such a large variety of requests is only possible in the online environment. The degree and experience of collaborating and sharing information has also changed with online learning. For example, I recently was working on my computer in a hotel in

The world is moving away from the model of a teacher as a container of valuable information to be disseminated to learners. Instead, the new model of teaching involves facilitation. Teachers facilitate learners to find what they need to construct their own answers to problems and issues in life.

6




Berlin while simultaneously chatting online with one colleague in Canada and another in China. Computing is becoming pervasive and ubiquitous as we move into a world of wireless hotspots, ambient networks, mobile devices, and wearable computers. Thus, the world of learning can now revolve around the individual learner and not the instructor. User-created content, often placed in repositories as open source content or software, is becoming commonplace. Instead of a central administrative office keeping information banks on many learners, those learners are now keeping their own records, often in the form of e-portfolios (Roberts et al., 2005). All this is to say that e-learning is based on a set of emerging technologies. Five years ago, perhaps a dozen technologies could be identified as producing and supporting e-learning materials and experiences. This report identifies 52 distinct technologies that are being used today in online learning. Each technology and its relevance to learning is described. Strengths and weaknesses of each technology are then presented, along with selected examples, online resources, and a bibliography. Because this research report is meant as a reference work, the reader can approach any topic in any order. In the last chapter, I discuss the meaning of these emerging technologies in elearning based on both knowledge lifecycles and technology innovation cycles. From this, I have tried to project what we can expect over the next five years. I hope you will find it useful.

Bibliography Abbas, Z., Umer, M., Odeh, M., McClatchey, R. Ali, A., Ahmad, F. (2005). A Semantic Grid-based E-Learning Framework (SELF). Paper presented at the 5th IEEE International Symposium on Cluster Computing and the Grid, Cardiff, Wales, UK. http://arxiv.org/abs/cs.DC/0502051 Alexander, Bryan (2006). Web 2.0: a new wave of innovation for teaching and learning? EDUCAUSE Review, 41(2), March/April 2006. http://www.educause.edu/apps/er/erm0 6/erm0621.asp?bhcp=1 Amoretti, M., Bertolazzi, R., Reggiani, M., Zanichelli, F. and Conte, G. (2005). Service-oriented Grids for Dynamic ELearning Environments. Paper presented at Communities and Technologies 2005, Milan, Italy, June 2005. http://www.idi.ntnu.no/~divitini/ubilearn 2005/Final/amoretti_ubilearn.pdf Anderson, Chris (2004). The Long Tail. Wired Magazine, 12(10), October. http://www.wired.com/wired/archive/12. 10/tail.html Anderson, Chris (2006). The Long Tail: why the future of business is selling less of more. New York: Hyperion. http://www.amazon.com/gp/product/14 01302378/104-71320800775136?v=glance&n=283155

e-learning is based on a set of emerging technologies.

Do Not Reproduce

7


Bailetti, T., Weiss, M. and McGinnis (2004). A Service-Oriented Architecture for Creating Customized Learning Environments. Paper presented at workshop for the Semantic Web Interest Group, Nov. 19, 2004, Montreal. http://www.cscsi.org/home/CSCSI/Memb ers/swig/swig04papers/bailetti-weissmcinnis.pdf Bogonikolos, N., Chrysostalis, M., Giotopoulos, K., Likothanassis, S., and Votis, K. (2002). Adaptive E-Learning GRID Platform. Paper presented to the International Workshop on Educational Models for GRID Based Services, Lausanne, Switzerland, Sept. 16, 2002. http://ewic.bcs.org/conferences/2002/1 stlege/session1/paper1.htm Bollier, David (2005). When Push comes to Pull: the new economy and culture of networking technology. A Report of the Fourteenth Annual Aspen Institute Roundtable on Information Technology. http://www.aspeninstitute.org/site/apps/ ka/ec/product.asp?c=huLWJeMRKpH&b= 667387&ProductID=283015 Bouras, C., Hornig, G., Triantafillou, V., Tsiatsos, T. (2001). Architectures Supporting e-Learning Through Collaborative Virtual Environments: The Case of INVITE. In Proceedings of IEEE International Conference on Advanced Learning Technologies - ICALT 2001, Madison, WI, USA, Aug. 6-8, 2001, 13-16. http://ru6.cti.gr/ru6/publications/95795 23.pdf

Cerri, Stefano (2005). Human Learning as a Side Effect of Learning GRID services. Keynote address at the Cognition and Exploratory Learning in Digital Age Conference (CELDA 2005), Dec. 14-16, Porto, Portugal. http://www.iadis.org/celda2005/index.a sp Chatarji, Jagadish (2004). Introduction to Service Oriented Architecture (SOA). Web Services - Dev Shed, Oct. 13, 2004. http://www.devshed.com/c/a/WebServices/Introduction-to-Service-OrientedArchitecture-SOA/ Dede, C. (1996). Emerging technologies and distributed learning. American Jrnl. of Distance Education 10, 2, 4-36. http://www.virtual.gmu.edu/pdf/ajde.pdf Erl, Thomas (2004). Service-Oriented Architecture: a field guide to integrating XML and Web Services. Upper Saddle River, NJ: Prentice-Hall. http://www.amazon.com/gp/product/01 31428985/104-98511511919955?v=glance&n=283155 Friesen, N. and McGreal, R. (2002). International E-Learning Specifications. Athabasca University, Centre for Distance Ed., Software Evaluation Report R11/0203, March 2002. http://cde.athabascau.ca/softeval/report s/R110203.pdf Friesen, Norm (2006). CanCore: connection collections an overview of approaches. Online. http://www.cancore.ca/protocols_en.htm l

8




Hagel, J. and Brown, J.S. (2005). From Push to Pull-Emerging Models for Mobilizing Resources. Working paper. http://www.johnhagel.com/paper_pushpull .pdf He, Hao (2003). What is service-oriented architecture? OReilly Webservices .xml.com, Sept. 30, 2003. http://www.xml.com/pub/a/ws/2003/09/ 30/soa.html Hockemeyer, C. and Albert, D. (2002). Adaptive e-Learning and the Learning Grid. Paper presented to the 1st LEGE-WG International Workshop on Educational Models for GRID Based Services, Lausanne, Switzerland, Sept. 2002 http://ewic.bcs.org/conferences/2002/1st lege/session2/paper1.htm Hohpe, Gregor (2005). Developing Software in a Service-Oriented World. ThoughtWorks White paper, January 2005. http://www.enterpriseintegrationpatterns.c om/docs/SOA_World.pdf Knorr, E., Erlanger, L. and Borck, J. (2005). A field guide to software as a service. Infoworld, April 18, 2005. http://www.infoworld.com/article/05/04/ 18/16FEsasdirect_1.html Kong, W., Luo, J., and Zhang, T. (2005). A Workflow based E-learning Architecture in Service Environment. Proceedings, International Conference on Computer and Information Technology (CIT'05), 1026-1032. http://csdl2.computer.org/persagen/DLAb sToc.jsp?resourcePath=/dl/proceedings/ci t/&toc=comp/proceedings/cit/2005/2432 /00/2432toc.xml&DOI=10.1109/CIT.200 5.56

Marshall, Stephen (2004). E-learning standards: open enablers of learning or compliance straight jackets? Paper presented at the 2004 ASCILITE conference, Perth, Australia. http://www.ascilite.org.au/conferences/p erth04/procs/pdf/marshall.pdf Millea, J., Green, I. and Putland, G. (2005). Emerging Technologies: a framework for thinking. Education.au Limited, for the Australian Dept. of Education and Training. http://www.det.act.gov.au/publicat/pdf/ emergingtechnologies.pdf Neumann, F. and Geys, R. (2004). SCORM and the Learning Grid. Paper presented at the 4th International LeGEWG Workshop: Towards a European Learning Grid Infrastructure: Progressing with a European Learning Grid. Stuttgart, Germany. 27 - 28 April 2004. http://ewic.bcs.org/conferences/2004/4 thlege/session3/paper2.pdf Norman, Donald (1993). Things that Make Us Smart: defending human attributes in the age of the machine. Cambridge, MA: Perseus. http://www.amazon.com/gp/product/02 01626950/104-44064197257534?v=glance&n=283155 Pankratius, V. and Vossen, G. (2003). Towards e-learning grids: using grid computing in electronic learning. Proceedings of the IEEE Workshop on Knowledge Grid and Grid Intelligence, International Conference on Web Intelligence/Intelligent Agent Technology. http://www.aifb.unikarlsruhe.de/BIK/vpa/pankratius_vossen _e-learninggrids.pdf

Do Not Reproduce

9


Pierre, Samuel (2006). E-Learning Networked Environments and Architectures: A Knowledge Processing Perspective. Berlin: Springer. http://www.amazon.com/gp/product/18 46283515/sr=83/qid=1155437505/ref=sr_1_3/1041348092-4859103?ie=UTF8 Riachi, Rhonda (2006). Trends in elearning: education versus entertainment? Presentation, Library+information Show, Birmingham, UK, April 26-27, 2006. http://www.alt.ac.uk/docs/trends_in_elearning_27april2006.pdf Roberts, G., Aalderink, W., Cook, J., Feijen, M., Harvey, J., Lee, S. and Wade, V. (2005). Reflective learning, future thinking: digital repositories, e-portfolios, informal learning and ubiquitous computing. Paper presented at the ALT/SURF/ILTA1 Spring Conference Research Seminar, Trinity College, Dublin, April 1, 2005. http://www.alt.ac.uk/docs/ALT_SURF_ILT A_white_paper_2005.pdf Spaniol, M., Klamma, R. and Jarke, M. (2003). ATLAS: A Web-Based Software Architecture for Multimedia E-learning Environments in Virtual Communities. In W. Zhou et al. (Eds.): Proceedings, ICWL 2003, Berlin: Springer-Verlag, pp. 193 205. http://www-i5.informatik.rwthaachen.de/i5new/staff/spaniol/publicati ons/ATLAS.pdf

Urdan, T. and Weggen, C. (2000). Corporate E-Learning: exploring a new frontier. Research report, W.R. Hambrecht and Co. http://www.astd.org/NR/rdonlyres/E2CF 5659-B67B-4D96-9D85BFAC308D0E28/0/hambrecht.pdf Werkhoven, Peter (2004). Experience machines: capturing and retrieving personal content. ACTeN E-Content Report No. 9. http://www.acten.net/uploads/images/4 31/experience_machines.pdf Woodill, G. and Oliveira, C. (2006). Mashups, SOAP and Services: welcome to Web hybrid e-learning applications. Learning Solutions, May 15, 2006. http://www.operitel.com/publications.aspx Yang, C. and Ho, H. (2005). An e-Learning Platform Based on Grid Architecture. Journal of Information Science and Engineering, 21, 911-928. http://www.iis.sinica.edu.tw/JISE/2005/ 200509_06.pdf

10



Part II: Emerging e-Learning TechnologiesWhat follows are individual reviews of 52 elearning technologies. Included are related terms, a brief description of the technology and the issues surrounding it, selected elearning related examples, online resources to learn more about each technology, and a bibliography for each section. There are over 2000 hyperlinks in this section of the report, allowing the reader to investigate each topic to a much greater depth.

Do Not Reproduce

11

Affective ComputingRelated terms Artificial intelligence, emotional design, emotions, facial recognition, gestures, keystroke patterns, posture, privacy, security, speech patterns Description Affective computing allows computers to interpret, understand, and react to human emotions. Given that 80 percent to 90 percent of human-to-human communication is nonverbal, it is not surprising that researchers are working on software that can recognize the nonverbal cues that indicate specific human emotional states. Without the ability to understand emotions, computers will never be-come human-like or appear natural. Affective computing employs cameras and body sensors to discover clues about what a user is feeling. Specific algorithms interpret these clues and instruct the computer to take appro-priate actions. Affective computing can detect whether a learner is having a problem with a subject and adjust accordingly by offering tutoring or less difficult learning materials. For example, in Italy, the MYSELF project is trying to integrate affective computing into virtual tutors to enhance distance learning and training applications. Affective computing is aimed at giving computers skills of emotional intelligence, including the ability to recognize and express emotions, and to respond to them effectively. (Anolli, et al., 2005). In online therapy, too, affective computing can give the therapist more information on a clients emotional state. Using emotionally realistic characters in an online simulation can make e-learning more effective (Maldonado et al., 2005). Making computers more responsive to a learners emotions should also enhance learning. SRI Consulting (http://www.sricbi.com/Explorer/NGT-AC.shtml) contends that affective computing is an important development in computing, because as pervasive or ubiquitous computing becomes mainstream, computers will be far more invisible and natural in their interactions with humans. Projected benefits include the following: Making people more comfortable with their computers Detecting whether a person is under stress, confused, or sad, then trying to change the user's emotional state Improving the safety of public spaces by detecting a person's malicious intent before he or she commits a crime Learning about the state of employees emotions in order to increase productivity Assessing the reaction of consumers to product offerings Learning about the state of employees emotions in order to increase productivity Assessing the reaction of consumers to product offerings

Affective computing allows computers to interpret, understand, and react to human emotions.

12



Affective Computing

Major difficulties with using computers for the above benefits are concerns with privacy, accuracy, and legality. Selected Examples Rosalind Picard, author of the 1997 groundbreaking book, Affective Computing, heads the MIT lab on affective computing. The labs Web site has many resources to check out at: http://affect.media.mit.edu The MYSELF project coordinates a number of researchers in several European countries who are working on affective computing. See: http://www.myself-proj.it/ Ditto the Donkey software rates the niceness or nastiness of messages and responds emotionally. Meet Ditto at: http://www.convo.co.uk/x02/ Online Resources The Cognition and Affect Project at the University of Birmingham, UK, maintains a list of papers and doctoral dissertations on the topic of affective computing. Access it from: http://www.cs.bham.ac.uk/research/cogaf f/0-INDEX.html The Emotion Home Page is a listing of various research studies on emotion, including studies of emotions in computers. See the links at: http://emotion.nsma.arizona.edu/emotion. html

The first international conference on Affective Computing and Intelligent Interaction was held in Beijing, China, October 22-24, 2005. It is instructive to read the list of papers presented and to see the advances that have been made in this field. Contents of the proceedings of ACII 2005 are at: http://www.informatik.unitrier.de/~ley/db/conf/acii/acii2005.html In Germany, Dr. Christoph Bartneck maintains an Affective Computing Portal, listing many links to interesting resources. Check it out at: http://www.bartneck.de/link/affective_p ortal.html

and Emotional Roots of Cognition and Action, University of Hertfordshire, Hatfield, UK, April 12-15, 2005, are available online: http://www.aisb.org.uk/publications/proce edings/aisb05/2_Agents_Final.pdf

Given that 80 percent to 90 percent of human-tohuman communication The Geneva Emotion Research Group at is nonverbal, it is the University of Geneva maintains a Web not surprising site on this topic, with many resources. that researchFind it at: ers are working http://www.unige.ch/fapse/emotion/ on software that can recognize The Humaine Project has a portal with reports, bibliographies, and demonstrations the nonverbal on affective computing in Europe. See it at: cues that http://emotion-research.net/ indicate specific human The Proceedings of the Symposium on emotional Agents that Want and Like: Motivational states.

Do Not Reproduce

13

Affective Computing

Bibliography Anolli, L., Mantovani, F., Balestra, M., Agliati, A., Realdon, O., Zurloni, V., Nortillaro, M. Vescovo, A. and Confalonieri (2005). The Potential of Affective Computing in ELearning: MYSELF project experience. Paper presented to the Workshop on eLearning and Human-Computer Interaction: Exploring Design Synergies for more Effective Learning Experiences, September 13, 2005 http://www.dis.uniroma1.it/~lhci/009.pdf Cahour, B., Salembier, P., Brassac, C., Bouraoui, J., Pachoud, B., Vermersch, P., and Zouinar, M. (2005). Methodologies for Evaluating the Affective Experience of a Mediated Interaction. Paper presented at the workshop on Evaluating Affective Interfaces, CHI 2005 conference, Portland, Oregon, April 2-7, 2005. http://www.sics.se/~kia/evaluating_affecti ve_interfaces/Cahour.doc Chateau, N. and Merisol, M. (2005). AMUSE: a tool for evaluating affective interfaces. Paper presented at the workshop on Evaluating Affective Interfaces, CHI 2005 conference, Portland, Oregon, April 2-7, 2005. http://www.sics.se/~kia/evaluating_affecti ve_interfaces/Chateau.pdf Diamond, David (2003). The Love Machine: building computers that care. Wired Magazine, Issue 11/12, December. http://www.wired.com/wired/archive/11.1 2/love.html

Fallman, D. and Waterworth, J. (2005). Dealing with User Experience and Affective Evaluation in HCI Design: A Repertory Grid Approach. Paper presented at the workshop on Evaluating Affective Interfaces, CHI 2005 conference, Portland, Oregon, April 2-7. http://www.sics.se/~kia/evaluating_affe ctive_interfaces/Fallman.pdf Goren-Bar, D., Graziola, I., Pianesi, F., Rocchi, C., Stock, O. and Zancanaro, Z. (2005). I Like It - Affective Control of Information Flow in a Personalized Mobile Museum Guide. Paper presented at the workshop on Evaluating Affective Interfaces, CHI 2005 conference, Portland, Oregon, April 2-7, 2005. http://www.sics.se/~kia/evaluating_affe ctive_interfaces/Goren-Bar.doc Hook, K., Isbister, K., and Laaksolahti, J. (2005). Sensual Evaluation Instrument. Paper for Evaluating Affective Interfaces, CHI 2005, Portland, Oregon, April 2-7. http://www.sics.se/~kia/evaluating_affe ctive_interfaces/Hook.pdf Kaye, Joseph (2005). Intimate Objects: a site for affective evaluation. Paper for Evaluating Affective Interfaces, CHI 2005, Portland, Oregon, April 2-7. http://www.sics.se/~kia/evaluating_affecti ve_interfaces/Kaye.pdf Maldonado, H., Lee, J., Brave, S., Nass, C., Nakajima, H., Yamada, R., Iwamura, K. and Morishima, Y. (2005). We Learn Better Together: Enhancing eLearning with Emotional Characters. In Computer Supported Collaborative Learning 2005: Next 10 Years! Mahwah, NJ. L. Erlbaum. http://www.stanford.edu/~kiky/CSCL200 5Maldonado.pdf

14



Affective Computing

Mandryk, Regan (2005). Evaluating Affective Computing Environments Using Physiological Measures. Paper presented at the workshop on Evaluating Affective Interfaces, CHI 2005 conference, Portland, Oregon, April 2-7, 2005. http://www.sics.se/~kia/evaluating_affecti ve_interfaces/Mandryk.pdf Masum, S. and Ishizuka, M. (2005). An affective role model of software agent for effective agent-based e-learning by interplaying between emotions and learning. Paper for WEBIST 2005, Miami, May 26, 2005. http://www.miv.t.utokyo.ac.jp/papers/mostafa/WEBIST2005 _Mostafa_Japan_Final.pdf Mentis, Helena (2005). Insight into Strong Emotional Experiences through Memory. Paper for Evaluating Affective Interfaces, CHI 2005, Portland, Oregon, April 2-7, 2005. http://www.sics.se/~kia/evaluating_affecti ve_interfaces/Mentis.pdf Norman, Donald A. (2004). Emotional Design: Why We Love (or Hate) Everyday Things. New York: Basic. http://www.elearningreviews.org/topics/human-computerinteraction/usability/2004-normanemotional-design/

Picard, R. and Daily, S. (2005). Evaluating affective interactions: Alternatives to asking what users feel. Paper for Evaluating Affective Interfaces, CHI 2005, Portland, OR. http://www.sics.se/~kia/evaluating_affe ctive_interfaces/Picard.pdf

Ruebenstrunk, Gerd (1998). Emotional Computers: Computer models of emotions and their meaning for emotion-psychological research. E-book http://www.ruebenstrunk.de/emeocom p/content.HTM Steele, M. and Steele, J. (2002). Applying affective computing techniques to the field of special education. Journal of Research on Technology in Education, 35(2), Dec. 22, 2002. http://www.iste.org/Content/Navigatio nMenu/Publications/JRTE/Issues/Volu me_351/Number_2_Winter_2002_200 31/Applying_Affective_Computing_Tech niques_to_the_Field_of_Special_Educat ion.htm

Tao, J., Tan, T. and Picard, R. (2005). Affective Computing and Intelligent Interaction: Proceedings of the First International Conference, ACII 2005, Beijing, China, October 22-24. Berlin: Picard, Rosalind (1997). Affective Springer. Computing. Cambridge, MA: MIT. http://www.amazon.com/gp/product/3 http://www.amazon.com/gp/product/02626 540296212/sr=861152/sr=82/qid=1155436264/ref=sr_1_2/1041/qid=1152930312/ref=sr_1_1/1041348092-4859103?ie=UTF8 9851151-1919955?ie=UTF8

Do Not Reproduce

15

Affective Computing

Wiberg, Charlotte (2005). Affective Computing vs. Usability?: insights of using traditional usability evaluation methods. Paper presented at the workshop on Evaluating Affective Interfaces, CHI 2005 conference, Portland, Oregon, April 2-7, 2005. http://www.sics.se/~kia/evaluating_affec tive_interfaces/Wiberg.doc Wright, Ian (1997). Emotional Agents. Doctoral Dissertation, University of Birmingham. http://www.cs.bham.ac.uk/research/cog aff/Wright.thesis.pdf

16



AgentsRelated terms: Artificial intelligence, autonomous agents, avatars, intelligent agents Description Agents are intelligent software programs that can act on behalf of an individual or a group. Agent-generated content can be utilized in several different ways. First, software agents can retrieve content on the Internet for an individual user. Second, software agents can watch for new items of interest to a learner and send an alert when one appears. Third, personal agents can negotiate with other agents to produce a personalized learning environment. Fourth, animated agents can be used to speak and present learning materials in an online application. These pedagogical agents serve as the role of teacher by presenting the materials to learners online. Software agents act on behalf of users to accomplish their goals. Being goal orientated is a key character of agents (Yan, 2004). Agents are autonomous and can act independently within the limits their programming. An intelligent agent is a computer system capable of flexible autonomous action in some environment (Wooldridge, 1999). In this context, flexible means reactive (responds to its environment), proactive (goal directed), and social (able to communicate and interact with other agents). Other possible qualities of online agents include mobility, veracity, benevolence, rationality, and, learning/adaptation. Wright (1997) has even suggested that virtual agents can have emotions, and many of them can actually learn. For example, Ueno (2005) describes an agent that learned from the log data of a Web site. Learning by computers is sometimes called machine learning, which is a sub-field of artificial intelligence. Intelligent computer aided instruction or tutoring programs often use agent technology. An agent can act as a personal assistant for a teacher and as a personal assistant for a student, and both may be found in the same program (for example, see the paper by Far et al., 1999). Sometimes multiple agents can work together. Luengo (1999) describes students interacting with three agents while constructing a mathematical proof. Dick Stenmark, of the University of Goteberg in Sweden, has classified intelligent agents as follows: Interface agents System agents Advisory agents Filtering agents Retrieval agents Navigation agents Monitoring agents Recommender agents Profiling agents http://w3.informatik.gu.se/~dixi/agent/cl ass.htm)

Agents are intelligent software programs that can act on behalf of an individual or a group.

Do Not Reproduce

17

Agents

Multiple adaptive agents act as a complex adaptive system to reproduce social dynamics with feedback loops and uncertain outcomes.

Most online agents in e-learning play the role of teacher or tutor. However, a Learning-by-Teaching approach can also be effective (Leelawong, 2005; Katzlberger, 2005). Viswanath, et al. (2004) report that teaching a computer agent can be effective in terms of learning. A software simulation for Grade 5 students called Bettys Brain learned by students teaching her about concept maps. She made mistakes, and the students had to continue to teach her. A second agent in the simulation, Mentor, told the students when Betty was wrong and how they could teach her properly. A student agent and an environment agent allowed interactivity and change within the environment. Agents can be used to model social systems (Guessoum, 2004) and are, therefore, useful in educational simulations. Multiple adaptive agents act as a complex adaptive system to reproduce social dynamics with feedback loops and uncertain outcomes. In such situations, agents need to cooperate with each other to solve collective problems. Sahin (2000) says that self-organization of intelligent agents is accomplished because each agent models other agents by observing their behavior. Agents have beliefs, not only about environments, but also about other agents. Therefore, an agent takes its decisions according to the model of the environment and the model of the other agents. Even though each agent acts independently, they take the other agents' behaviors into account to make a decision. This permits the agents

to organize themselves for a common task (Sahin, 2000). Stone (1998) reports on another study where multiple agents were organized in teams, acting against other teams of agents. One issue for further study involves how close to a human being a software agent needs to be to comfortably interact with people. Massaro et al. (1998) developed a conversational agent, Baldi, which could show realistic facial expressions to convey emotions on a computer screen. The agent was successful in language tutoring with children with hearing loss. Baylor and Kim (2003) applied the same thinking to the interaction effects between student ethnicity and agent ethnicity. Their study revealed that students working with agents of the same ethnicity perceived the agents to be significantly more engaging and affable. Baylor and Ebbers (2003) examined the question of whether it is more effective to have one pedagogical agent with combined expertise and motiva-tional support or two separate agents one with expertise and one with motivational support. They found that having two separate pedagogical agents representing the two roles had a significantly more positive impact on both learning and the perceived value of the agents.

18



Agents

Selected Examples Nel is an agent based tutoring system that teaches introductory physics. See the article by Williams et al. (2004): http://www.editlib.org/index.cfm?fuseactio n=Reader.ViewAbstract&paper_id=11101 A research group in Italy has used XML and the Java Agent Development Framework to develop a prototype e-learning system using multiple agents. MASEL (Multi-Agent System for E-Learning), uses seven different types of agents: 1) Chief Learning Officer (CLO) Assistant Agent 2) Skills Manager Agent 3) Student Assistant Agent 4) Learning Paths Agent 5) Content Agent 6) Chief Content Officer (CCO) Assistant Agent 7) User Profile Agent. For details see: www.old.netobjectdays.org/pdf/02/papers /malceb/0623.pdf Animated characters from Extempo Systems can be used in online teaching and coaching. They are available as adaptive coaches, expert role-players, and expert guides. Go to: http://www.extempo.com/ The simulations from Redwood e-Learning Systems make extensive use of pedagogical agents. See a demo at: http://www.redwoodelearning.com/ CodeBaby Corp. has a virtual studio for programming actions and gestures of a variety of online characters. See: http://www.codebaby.com

Online Resources For an online primer on pedagogical agents, go to: http://ldt.stanford.edu/~slater/pages/ag ents/ Professor Michael Wooldridge of the University of Liverpool has written over 200 articles and 13 books on the behaviors of software agents and on multi-agent systems. See: http://www.csc.liv.ac.uk/~mjw/ Professor Wooldridge also maintains a large bibliography on agents, See: http://liinwww.ira.uka.de/bibliography/Ai /agents.html For a set of papers on pedagogical agent research by Dr. Amy Baylor and her colleagues, go to: http://ritl.fsu.edu/_Website/ Research on animated agents with programmed social skills is being carried out at the Center for Advanced Research for Technology in Education (CARTE) at the University of Southern California. For details of this research: http://www.isi.edu/isd/carte/ Bibliography Baylor, A., & Ebbers, S. (2003). The Pedagogical Agent Split-Persona Effect: When Two Agents are Better than One. In Proceedings of World Conference on Educational Multimedia, Hypermedia, and Telecommunications 2003, 459462. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=111 22

Agents have beliefs, not only about environments, but also about other agents. Therefore, an agent takes its decisions according to the model of the environment and the model of the other agents.

Do Not Reproduce

19

Agents

Baylor, A., & Kim, Y. (2003). The Role of Gender and Ethnicity in Pedagogical Agent Perception. In Proceedings, ELearning 2003, 1503-1506. http://www.editlib.org/index.cfm?fusea ction=Reader.ViewAbstract&paper_id=1 2158Cao, L., & Bengu, G. (2000). Developing Web-based Tutoring Agents Using CORBA. In Proceedings of WEBNET 2000 Conference, 75-80. http://www.editlib.org/index.cfm?fuseacti on=Reader.ViewAbstract&paper_id=6341 Choy, S., Ng, S. and Tsang, Y. (2005). Software Agents to Assist in Distance Learning Environments. Educause Quarterly, 28(2), 2005. http://www.educause.edu/ir/library/pdf/ eqm0523.pdf Clarebout, G., Elan, J., Johnson, W. and Shaw, E. (2002). Animated Pedagogical Agents: An Opportunity to be Grasped? Journal of Educational Multimedia and Hypermedia, 11(3), 267-286 http://www.editlib.org/index.cfm?fuseacti on=Reader.ViewAbstract&paper_id=9270 Elen, J., Clarebout, G., & Johnson, W. (2002). Animated pedagogical agents: Where do we stand? In Proceedings of World Conference on Educational Multimedia, Hypermedia, and Telecommunications 2002 (pp. 306-311). Norfolk, VA: AACE. http://www.editlib.org/index.cfm?fuseactio n=Reader.ViewAbstract&paper_id=9288

Far, B., Koono, Z., & El-Khouly, M. (1999). Agent-Based Computer Tutorial System: An Experiment for Teaching Computer Languages (ATCL). Jrnl. Inter-active Learning Res., 10 (3), 275-285. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=884 3 Garro, A. and Palopoli, L. (2002). An XML Multi-Agent System for e-Learning and Skill Management. Paper presented, International Symposium on Multi-Agent Systems, Large Complex Systems, and EBusinesses (MALCEB'2002), Erfurt, Germany. http://www.old.netobjectdays.org/pdf/02 /papers/malceb/0623.pdf Guessoum, Zahia (2004). Adaptive agents and multi-agent systems. IEEE Distributed Systems Online, 5(7), July. http://csdl.computer.org/comp/mags/ds /2004/07/o7004.pdf Jafari, Ali (2002). Conceptualizing Intelligent Agents for Teaching and Learning. Educause Quarterly, No. 3. http://www.educause.edu/ir/library/pdf/ eqm0235.pdf Kao, G., Sun, C., & Lin, S. (2005). A Heterogeneous Agent Model for Distributed Constructionism. In Richards, G. (Ed.), Proceedings, e-Learning 2005, 1353-1356. http://www.aace.org/newdl/index.cfm?fu seaction=Reader.ViewAbstract&paper_id =21382

20



Agents

Katzlberger, Thomas (2005). Learning by Teaching Agents. Doctoral Dissertation, Vanderbilt University, Nashville, Tennessee. www.teachableagents.org/papers/Thomas KatzlbergerDissertation.pdf Kim, Y. (2003). Things that Make Agent as Learning Companion Effective. In Proceedings, E-Learning 2003, 16591666. http://www.editlib.org/index.cfm?fuseactio n=Reader.ViewAbstract&paper_id=12193 Kutay, C. and Ho, P. (2005). Designing Agents for Feedback Using the Documents Produced in Learning. International Journal on E-Learning, 4(1), 21-38. http://dl.aace.org/16948 Leelawong, Krittaya (2005). Using the Learning-by-Teaching Paradigm to design intelligent learning environments. Doctoral Dissertation, Vanderbilt University, Nashville, TN. www.teachableagents.org/papers/krittayathesis-sp.pdf Lin, Fuhua Oscar (2005). Designing Distributed Learning Environments with Intelligent Software Agents. Hershey, PA: Information Sciences Publishing. http://www.amazon.com/gp/product/159 1405009/sr=88/qid=1155437833/ref=sr_1_8/1041348092-4859103?ie=UTF8

Mahmood, A.K., and Ferneley, E. (2006). Embodied agents in e-learning environments: an exploratory case study. Journal of Interactive Learning Research, 17(2), 143-162. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=628 5 Massaro, D., Cohen, M., Beskow, J., Daniel, S., and Cole, R. (1998) Developing and Evaluating Conversational Agents. Paper presented at WECC'98 conference. http://cslu.cse.ogi.edu/publications/ps/ MassaroCole_WECC98.pdf Menczer, Filippo (1998) Life-like agents: Internalizing local cues for reinforcement learning and evolution. Doctoral dissertation, U. of California. http://www.informatics.indiana.edu/fil/ Padgham, L. and Winikoff, M. (2004) Developing Intelligent Agent Systems. New York: John Wiley & Sons. http://www.amazon.com/gp/product/04 70861207/sr=11/qid=1155438530/ref=sr_1_1/1041348092-4859103?ie=UTF8&s=books Pankratius, V., Sandel, O. and Stucky, W. (2004). Retrieving Content with Agents in Web Service E-Learning. In The Symposium on Professional Practice in AI, IFIP WG12.5 - First IFIP Conference on Artificial Intelligence Applications and Innovations (AIAI). Toulouse, France, August. www.aifb.unikarlsruhe.de/BIK/vpa/109.pdf

Luengo, V. (1999). Cooperative Agents to Learn Mathematical Proof. In Proceedings, ED-MEDIA, 1999, 1632. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=728 3

Do Not Reproduce

21

Agents

Perez, R. and Solomon, H. (2005). Effect of a Socratic Animated Agent on Student Performance in a Computer-Simulated Disassembly Process. Journal of Educational Multimedia and Hypermedia, 14(1), 47-59. http://dl.aace.org/16968 Sahin, Ferat (2000). A Bayesian Network Approach to the Self-Organization and Learning in Intelligent Agents. Doctoral Dissertation, Virginia Poytechnic and State University. http://scholar.lib.vt.edu/theses/available /etd-09202000-00230057/ Sheremetov, L., and Neza, G. (1999). Multi-stage cooperation algorithm and tools for agent-based planning and scheduling in [a] virtual learning environment. Paper presented at the 1st International Workshop of Central and Eastern Europe on Multi-Agent Systems (CEEMAS), June 1-4, 1999, St. Petersburg, Russia. http://citeseer.ist.psu.edu/565321.html Shim, S., Atreya, S., Wesley, L., & Booth, R. (1999). ROADS: An Environment for Developing Automated Intelligent Agents to Support Distance Learning. Journal of Interactive Learning Research. 10 (3), 321-333. http://www.editlib.org/index.cfm?fuseacti on=Reader.ViewAbstract&paper_id=8833 Stone, Peter (1998). Layered Learning in Multi-Agent Systems. Doctoral Dissertation, Carnegie Mellon University, Pittsburgh, PA. http://www.cs.utexas.edu/~pstone/thesis /

Sung, J. & Lim, D. (2005). Intelligent Learning System Based on Tutoring Agent and VR Training Agent (TAVTA). In P. Kommers & G. Richards (Eds.), Proceedings, ED-MEDIA 2005, 14151420. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=202 77 Ueno, M. (2004). Animated agent to maintain learners attention in e-learning. In Proceedings of World Conference on ELearning in Corporate, Government, Healthcare, and Higher Education 2004, 194-201. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=112 88 Ueno, M. (2005). Intelligent LMS with an agent that learns from log data. In Richards, G. (Ed.), Proceedings, EDMedia 2005, 3169-3176. http://www.aace.org/newdl/index.cfm?fu seaction=Reader.ViewAbstract&paper_id =21687 Viswanath, K., Adebiyi, B., Biswas, G. & Leelawong, K. (2004). A Multi-Agent Architecture Implementation of Learning by Teaching Systems. Paper for Conference on Advanced Learning Technologies, Finland, 61-65. http://www.teachableagents.org/papers/ 158bis.pdf Williams, M., Gilbert, J., & Madsen, N. (2004). Nel: An Interactive Physics Tutor. In Proceedings, ED-MEDIA 2004, 30003002. http://www.editlib.org/index.cfm?fuseact ion=Reader.ViewAbstract&paper_id=111 01

22



Agents

Wooldridge, Michael (2002). Introduction to Multi-Agent Systems. New York: John Wiley & Sons. http://ca.wiley.com/WileyCDA/WileyTitle/p roductCd-047149691X.html Yan, Chun. (2004). Agent Mediated Grid Services in e-Learning. Paper at CLAG2004, Collaborative Learning Applications of Grid Technology, Chicago, April 18-21, 2004. http://research.ac.upc.edu/clag/cy_lanegoal_grid_service_conference.ppt

Do Not Reproduce

23

Animation SoftwareRelated terms Flash, motion graphics Description Animation has been a staple of e-learning since the start of computer assisted learning in the 1970s and 1980s. Animations in e-learning range from simple swapping of successive images to highly complex 3-D motion graphics. Toth (2003) identifies three major formats for online graphics: It is generally thought that adding animations to online materials can help as a learning aid. However, recently researchers have questioned whether animations make a difference. The results of many experiments have been mixed. Lowe (2004) argues that despite the plausibility of cognitively based arguments for the benefits of animation, research to date has failed to provide unequivocal evidence that it is superior to static depiction. Lowe adds that in some cases, animations may even prejudice learning. This is echoed in research by Hegarty et al. (2003), who suggest that stimulating a learners mental animation capacity may be more important for learning than watching a moving picture. Hegarty and his co-researchers found no advantage to using external animations. Rather, a static diagram, coupled with imagining how something worked, produced the best learning results. Also, animations can be complex and move quickly through showing a process without real understanding being achieved by the learner. Visual cues such as arrows pointing to relevant parts of an animation, improved the understanding of animated explanations. (Huk, et. al., 2003) Finally, producing animations can be very costly, with long hours spent to produce even a short sequence. Given that, and the uncertainty of its effectiveness, animations should be used sparingly in e-learning.

Animations in e-learning range from simple swapping of successive images to highly complex 3-D motion graphics.

Animated GIFs: A series of still images shown in sequence, like pictures in a flip book. This is an older animation technique that is not used as much today. Flash and Shockwave animations: Macromedias Flash and Shockwave have extensive abilities to produce sophisticated animation sequences. Flash animations are perhaps the most common form of animation used in elearning. Dynamic 3-D Web graphics: Threedimensional motion graphics draw on large datasets to visualize dynamic processes. Software tools for 3-D Web graphics are more expensive and complex to learn. In addition to adding eye candy to the presentation of educational materials, animation can add real educational value by illustrating a dynamic procedure that is relevant to understanding. However, animation for its own sake can often be distracting or misleading when implemented poorly.

24



Animation Software

Selected Examples Based on Toths (2003) types of animations listed above, animation tools can be divided into three groups: Software for Producing GIF Animations Software for Producing Flash and Shockwave Animations Software and Hardware for Producing 3-D Web Graphics GIF animations can be constructed with several different shareware or low cost programs, including the following: Gif Construction Set Professional http://www.mindworkshop.com/alchemy/g ifcon.html Real GIF Optimizer 3.05 http://www.topshareware.com/Real-GIFOptimizer-download-2965.htm Ulead GIF Animator 5.0 http://www.topshareware.com/Ulead-GIFAnimator-download-11513.htm For a selection of over 400,000 pre-built animated GIFs, go to the Animation Factory: http://www.animationfactory.com/animati ons/ Software for Producing Flash and Shockwave Animations One of the most popular software packages on the market is Macro-media Flash (Macromedia is now owned by Adobe). It is popular because it is easy to

use and is cross-browser compatible (Hess and Hancock, 2004). Macromedia Flash has based digital animation on traditional animation techniques. More sophisticated procedures in Flash may require the use of Action-Script, the builtin programming language. To use ActionScript, some knowledge of computer programming is necessary. For information on Flash, go to: http://www.macromedia.com/software/fl ash/flashpro/ Shockwave is an older technique from Macromedia, connected with its Director and Authorware content creation packages. Shockwave is a program that takes Director movies or Authorware animations and com-presses and readies them for play-back on the Web. A downloadable plug-in is required to play Shock-wave, whereas Flash plays automat-ically within the latest versions of the most popular Web browsers. For more information on Shockwave, go to: http://macromedia.com A low cost alternative to authoring in the Flash format is SWISH. Its at: http://www.swishzone.com/ Eighty-five pre-built Flash animations for Physics are available under a Creative Commons license from the University of Toronto. To try them: http://faraday.physics.utoronto.ca/Gener alInterest/Harrison/Flash/

in some cases, animations may even prejudice learning.

Do Not Reproduce

25

Animation Software

Software and Hardware for Producing 3-D Web Graphics Numerous 3-D authoring packages range from relatively inexpensive to tens of thousands of dollars. Following is a list of leading packages with company Web sites: AfterEffects An industry standard from Adobe. More details at: http://www.adobe.com/products/aftereff ects/main.html CAT - Character Animation Technol-ogies (CAT) has a set of advanced animation tools that work in 3ds Max. http://www.catoolkit.com/gallery/default. asp?pageID=1 EI Technology Groups Animation System and Amorphium, their 3-D character creation environment, have been used in a number of Academy Award winning films. See more at: http://www.eitechnologygroup.com Enliven - ViewPoints Enliven provides a simple visual interface for creating 3-D interactive Web content quickly and easily without programming. http://www.viewpoint.com/pub/products/ enliven.html Falling Bodies - Falling Bodies is a special purpose plug-in for Softimage|3D. It animates fall stunts, using accurate dynamic simulation techniques. For a free demo, go to: http://www.animats.com/dynamics/dem oversion.html

Lightwave 3D is a modeling, animating, and rendering tool. Version 9 is available: http://www.newtek.com/lightwave/lw9_b etafaq.php Massive software is used to add animated crowds to movies. See: http://www.massivesoftware.com Maya is a high-end modeling, animation, effects, and rendering solution from Autodesk (formerly Alias). See why it has won Academy Awards at: http://usa.autodesk.com/adsk/servlet/in dex?id=6871843&siteID=123112 Motion2 is professional level animation software from Apple that runs on both Macintosh and Intel platforms. It is part of Final Cut Studio, a high-end editing and visual effects suite. For details see: http://www.apple.com/finalcutstudio/mo tion/ SoftImage|XSI - Softimage Co., a subsidiary of Avid Technology, Inc., has one of the worlds most advanced 3-D content creation tools. Learn more at: http://www.softimage.com/ Novices at animation can try the HTML and Flash templates from Animation Online. Their VisionBlazer product is described as easy to use. For more information, see: http://www.animationonline.com/ Ascension Technologies has a wide variety of motion capture tools that will turn any sequence of movements into an animated 3-D character with the same moves. Go to: http://www.ascension-tech.com/

stimulating a learners mental animation capacity may be more important for learning than watching a moving picture.

26


Animation Software

Online Resources The ACM SIGGRAPH Industry Directory lists hundreds of firms that develop animations or have animation software. Find more at: http://esub.siggraph.org/cgibin/cgi/idCatResults.html&CategoryID=8 For a comprehensive list of animation software, go to AllWorldSoft.com. You will find a listing of over 60 software packages that can be used to develop animations: http://www.allworldsoft.com/folders/page 2/graphic-apps/animation-tools/ Hundreds of tools exist for video production and animation. FreeDownloads Center.com lists almost 700 free tools: http://www.freedownloadscenter.com/Mul timedia_and_Graphics/Video_and_Animati on_Tools/ Bibliography Castillo, S., Hancock, S. and Hess, G. (2004). Using Flash MX to Create eLearning. Lehi, Utah: Rapid Intake Press. http://www.rapidintake.com/books_catalo g.htm Hegarty, M., Kriz, S., and Cate, C. (2003). The roles of mental animations and external animations in understanding mechanical systems. Cognition and Instruction, 21(4), 325360. http://www.psych.ucsb.edu/~hegarty/C&I %20HKC.pdf Hess, G. and Hancock, S. (2004). Using Macromedia Flash MX 2004 as an ELearning Authoring Environment. Learning Circuits, July 2004. http://www.learningcircuits.org/2004/jul2 004/hess.htm

Hess, G. and Hancock, S. (2004). Using Macromedia Flash MX 2004 as an ELearning Authoring Environment. Learning Circuits, July. http://www.learningcircuits.org/2004/jul 2004/hess.htm Huk, T., Steinke, M., and Floto, C. (2003). The educational value of cues in computer animations and its dependence on individual learner abilities, Proceedings of the ED-Media 2003 Conference. http://projekte.learninglab.unihannover.de/pub/bscw.cgi/d17506/Huk _EDMedia2003.pdf Lowe, R.K. (2004). Animation and learning: value for money? In R. Atkinson, C. McBeath, D. Jonas-Dwyer & R. Phillips (Eds.) Beyond the comfort zone: Proceedings, ASCILITE Conference. http://www.ascilite.org.au/conferences/p erth04/procs/lowe-r.html Malheiro, T. (2003). Flash Interactive Session. Paper presented at ED-Media 2003, (1), 1046-1048. http://dl.aace.org/12937 Mayer, R. and Moreno, R. (2002). Animation as an aid to multimedia learning, Educational Psychology Review, March 2002, 14(1), 87-99. http://www2.sjsu.edu/depts/it/edit235/ handouts/mayer_mmlearn.pdf Toth, Thomas (2003). Animation just enough, never too much, Learning Circuits, July 18, 2003. http://www.learningcircuits.org/2003/au g2003/toth.htm

animation for its own sake can often be distracting or misleading when implemented poorly.

27


Artificial IntelligenceRelated terms Adaptive systems, agents, AI, cognitive informatics, data mining, expert systems, intelligent tutoring, machine learning, multi-agent systems, personalization Description Artificial intelligence uses computer programming to simulate reasoning and thought processes similar to those in human beings. The success of artificial intelligence is sometimes measured against the Turing Test, whereby human beings interact with a computer interface that may have a human or computer hidden from view. The test is considered successful if the person is unable to tell whether there is a computer or another human being on the other end. So far, no computer program has been able to pass the Turing Test. Artificial intelligence initiatives encompass a wide range of computer programming techniques and systems. While it is beyond the scope of this research report to get into the technical details, here is a list of some of the many applications to which artificial intelligence is being put: Adaptive or Intelligent Tutoring Affective Computing Agents Bayesian Models Bioinformatics Business Intelligence Systems Case-based Reasoning Causal Models Chaos and Complexity Theories Cognitive Processes Connectionist Models Context-aware Computing Cooperative AI Systems Data Mining and Web Mining Distributed Artificial Intelligence DNA Computing E-business and E-commerce Evolutionary Engineering Expert Systems Fuzzy Logic and Systems Game Design Genetic Algorithms and Programs Human-centered Computing Hybrid Systems Information Retrieval Intelligent Control Systems Intelligent Databases Intelligent User Interfaces Knowledge Representation Logic Programming Machine Learning Man-Machine Interfaces Mobile Computing and Systems Model-based Reasoning Multi-agent Systems Neural Networks Neuro-Computing Probabilistic Reasoning Simulations Software Tools Temporal Reasoning User-profiling for personalization Virtual Reality Visualization Given the high expectations, artificial intelligence has not lived up to its initial promise or hype. Nevertheless, there are important and useful applications of artificial intelligence to online learning.

Artificial intelligence uses computer programming to simulate reasoning and thought processes similar to those in human beings.

28



Artificial Intelligence

Many of the topics listed above are included in this research report, making it clear that artificial intelligence is having a major impact on emerging e-learning techniques and technologies. A central topic of artificial intelligence is learning. Having a computer learn is termed machine learning as opposed to human learning, but many of those working in the field of artificial intelligence see these two types of learning as converging and becoming the same thing. Others are skeptical and believe that another kind of intelligence will emerge from artificial intelligence, one that is different from the intelligence of human beings and other intelligent life forms. To achieve their goals, AI scientists try to model how experts solve problems in a given domain. Once the solutions to problems are encoded in the computer program, algorithms are written to have the computer act as a tutor in that subject area. Intelligent tutoring systems (ITS) that provide direct feedback to learners are part of an emerging and intense area of research in the use of artificial intelligence in educational environments. An ITS may use a variety of technologies, including collaborative filtering, recommender systems, and data mining. ITS systems create several different user models - profiles of the learner, the subject matter expert, and the teacher. To seem humanlike, ITS systems often use some version of natural language processing. The vision of a computer taking the place of a teacher has been around for quite some time. The reality is that, in spite of progress in artificial intelligence, intelligent tutoring systems are not yet ready to replace human instructors.

Some current issues with using artificial intelligence and education include: Gaming the System Aleven et al. (2004) found that 72% of all student actions represented unproductive help- seeking behavior[W]e found a proliferation of hint abuse (e.g., using hints to find answers rather than trying to understand). We also found that students frequently avoided using help when it was likely to be of benefit and often acted in a quick, possibly undeliberate manner. Hedging and Hostility Bhatt et al. (2005) say that students hedge and apologize often to human tutors but very rarely to computer tutors. The type of expressions also differedovert hostility was not encountered in human tutoring sessions but was a major component in computertutored sessions. On the other hand, for Johnson and Rizzo (2004), a major issue was too much politeness between the learner and the online tutor. Emotional Effects Chaffar and Frasson (2004) note that emotions play an important role in cognitive processes and especially in learning tasks. Moreover, there is some evidence that the emotional state of the learner is correlated with his performanceits important that new Intelligent Tutoring Systems involve this emotional aspect; they may be able to recognize the emotional state of the learner, and to change it so as to be in the best conditions for learning. Ochs and Frasson (2004) also discuss how emotions affect learning with intelligent tutoring systems, and Aist et al. (2002) contend that provision of human emotional scaffolding made a positive difference (increased persistence and

Having a computer learn is termed machine learning as opposed to human learning, but many of those working in the field of artificial intelligence see these two types of learning as converging and becoming the same thing.

Do Not Reproduce

29


learning) for students using an intelligent tutoring system. Complexity According to Thomsen-Gray et al. (2003), intelligent tutoring in nondeterministic and dynamic domains can be very complex and can lead to unexpected results. Whereas human tutors must teach students how to respond to unexpected results in a timely and appropriate manner, computer based systems usually have limited ability to do this. Context Kinshuk and Patel (1997) suggest that one weakness of intelligent tutoring systems is their lack of ability to understand the context of the learner. While an ITS inherits powerful functionality at the points of convergence between its objectives and the capabilities of the methodology employed, it also inherits a context gap at the points of divergence between the purpose of the tasks performed within an ITS and the purpose of the methodology. Degree of Personalization and Use of User Profiles Personalization using artificial intelligence depends on the set of assumptions made about the users and how user models are constructed. While many systems purport to be personalized, they can be frustratingly wrong about what a user wants and needs at any given time. There is debate in the literature over the use of user profiles vs. a building a system that infers tutoring suggestions from assessing the users interactions with the system (see Smid et al, 2002).

Talking Head Tutor vs. Voice Only Tutor Craig et al. (2004) show that while a talking head displaying facial expressions, gestures, and gaze during dialog does not produce a split attention effect and concomitant decrements in performance, it also does not enhance performance when compared to a condition that includes only spoken narration. The talking head agent metaphor may be more trouble (and expense) than it is worth. Difficulties in Representing Knowledge Hatzilygeroudis and Prentzas (2005) provide a comprehensive review of different schemes for representing knowledge. They advocate for a hybrid approach to knowledge representation, rather than using a single type of knowledge. They divide knowledge into the following types: Structural knowledge is concerned with types of entities (i.e. concepts, objects, etc) and how they are interrelated. Relational knowledge concerns relations between entities of the domain. Heuristic knowledge is knowledge in the form of rules of thumb, practical knowledge about how to solve problems based on experience. Schemes for knowledge representation from Hatzilygeroudis and Prentzas (2005) include the following:

While many systems purport to be personalized, they can be frustratingly wrong about what a user wants and needs at any given time.

Do Not Reproduce

30


Schemes for knowledge representation from Hatzilygeroudis and Prentzas (2005) include the following: Single schemes: Semantic nets Conceptual Graphs Ontologies Symbolic rules Expert systems Case-based representations Neural networks Belief networks Hybrid schemes: Fuzzy rules Connectionist expert systems Integration of rules and cases Description logics Terminological knowledge Assertional knowledge Neurules (integration of symbolic rules with neurocomputing) As the above list of issues shows, using artificial intelligence in e-learning is not a simple matter. A great deal of further development needs to occur before this technology becomes mature. Selected Examples Artificial intelligence in e-learning has generated a wide range of approaches to improving computer-based teaching. Approaches include the following: Dialogue-Based Intelligent Tutoring Systems Yang (2001) describes a system for taking turns in a dialogue-based intelligent tutoring system. http://www.cs.iit.edu/~circsim/documents /fydiss.pdf

Reasoning About Actions and Changes Baldoni et al. (2004) use an agent logic language (DyLOG) to implement reasoning capabilities of agents to dynamically build study plans and to verify the correctness of user-given study plans with respect to the compet-ence that the user wants to acquire. http://www.di.unito.it/~argo/papers/200 4_JAIR.pdf Natural Language Processing Di Eugenio et al. (2005a, 2005b) developed two natural language generators and ... found that the generator which intuitively produces the best language does engender the most learning. http://www.cs.uic.edu/~bdieugen/PSpapers/ACL05.pdf Latent Semantic Analysis (LSA) LSA is a technique used for automatic scoring of essays. Steinhart (2001) used this e for tutoring writing. http://lsa.colorado.edu/papers/daveDiss ertation.pdf For a portal on the latent semantic analysis, see: http://lsa.colorado.edu/ Bayesian Networks Butz et al. (2004) describe Bayesian networks as a formal framework that uses probability techniques for uncertainty management. Web intelligence researchers have applied Bayesian net-works to many tasks, including student monitoring, ecommerce, and multiagents. http://www.cs.uregina.ca/~butz/publicati ons/wi04.pdf

using artificial intelligence in e-learning is not a simple matter. A great deal of further development needs to occur before this technology becomes mature.

Do Not Reproduce

31


Far (2006) describes the use of Bayesian techniques in the development of a multiagent learning and tutoring system. http://www.enel.ucalgary.ca/People/far Precision Teaching/Programmed Learning Precision teaching is a very systematic approach to teaching based on behaviorism. Infrature, describes this in a white paper on learning theories. http://www.infrature.com/Library/WhiteP apers/LearningTheories.htm Ontology Based Systems Day et al. (2005) propose an Intelligent Tutoring Agent (ITA) that uses ontology, question answering (QA) techniques, and INFOMAP, a knowledge representation framework that can be used to extract important concepts from a natural language text. http://www.iis.sinica.edu.tw/IASL/webpdf /paper-2005-Designing_an_Ontologybased_Intelligent_Tutoring_Agent_with_In stant_Messaging.pdf Oguejiofor et al. (2004) also discuss an ontology-based approach to the design of intelligent tutoring systems. http://www.kicinger.com/publications/pd f/OguejioforIT-AEC2004.pdf Hierarchical Graphs Gutierrez et al. (2004) note that courses tend to have a high number of learning objects. As a result, designing a personalized sequencing strategy for each student quickly becomes unmanageable. They propose using an approach called hierarchical graphs. http://bach.gast.it.uc3m.es/~sergut/publ ications/Gutier04b.pdf

Side-By-Side Example Tutoring - Davidovic (2001) describes and evaluates the Structural Example-based Adaptive Tutoring System (SEATS) and a number of other intelligent tutoring systems. http://ariic.library.unsw.edu.au/unisa/ad t-SUSA20050922-010120/ Student Log Files McLaren et al. (2004a) argue that a potentially powerful way to aid in the authoring of intelligent tutoring systems is to directly leverage student interaction log data. They propose an approach called bootstrapping novice data (BND) in which a problem-solving tool is integrated with tutor development software through log files and that integration is then used to create the beginnings of a tutor for the tool. http://ctat.pact.cs.cmu.edu/pubs/ITS20 04-BND-Camera-Ready.pdf Teaching Metacognitive Strategies by Computer Graesser et al. (2005) describe some of recent computer systems that were designed to facilitate explanation-centered learning through strategies of inquiry and metacognition while students learn science and technology content. http://www.leaonline.com/doi/pdf/10.12 07/s15326985ep4004_4 Quantum Intelligent Tutoring Engines develop software for others to build intelligent tutoring applications. http://quantumsimulations.com/index.ht ml

To achieve their goals, AI scientists try to model how experts solve problems in a given domain. Once the solutions to problems are encoded in the computer program, algorithms are written to have the computer act as a tutor in that subject area.

32




Founded in 1988, Stottler Henke Associates, Inc. applies artificial intelligence and other advanced software technologies to solve problems that defy solution using traditional approaches. Stottler Henkes products include the following: SimBionic - A visual authoring tool and runtime engine for creating complex behaviors in computer-based training simulations and games more quickly and easily, so that these systems become more realistic, challenging, and engaging. Task Tutor Toolkit - A set of Java software libraries and applications for creating intelligent tutoring system scenarios quickly and easily, without programming. Aurora - A sophisticated scheduling system that combines a variety of scheduling techniques, intelligent conflict resolution, and decision support to make scheduling faster and easier. For more information on Stottler Henke Associates, see: http://www.stottlerhenke.com/ Gemini Performance Systems used artificial intelligence to build the SWIFT adaptive learning environment as an intelligent tutoring system comprised of an adaptive learning environment, an adaptive testing algorithm, and an interactive intelligent tutor. http://www.gemini.com Virtuel Age International has an artificial intelligence-based intelligent tutoring system that dynamically adapts the course according to the learner's existing

knowledge base, skill gaps, preferred cadence, and learning style, taking personalized and adaptive learning to a new level. http://www.virtuelage.com The Reusable Artificial Intelligence Tutoring System Shell (RAITSS) from Knowledge Engineering allows users to build intelligent tutoring systems. http://www.ke-corp.com/s10.htm Carnegie Mellon University is a leading research institution that uses artificial intelligence in education. Its Pittsburgh Advanced Cognitive Tutor Center (PACT) develops cognitive tutors that have been used widely in constructing intelligent tutoring systems in a variety of settings. http://pact.cs.cmu.edu/ Carnegie Mellon researchers are also developing a suite of authoring tools called Cognitive Tutor Authoring Tools (CTAT) to make tutor development easier and faster for developers and to make it possible for educators without technical expertise to develop such systems. Find out more at: http://ctat.pact.cs.cmu.edu/ Other universities with research groups in intelligent tutoring and artificial intelligence include the following: University of Sydney - Intelligent Tutoring Systems Research Group http://www.it.usyd.edu.au/~netsys/resea rch/current_computer_science_educatio n_research.htm

The vision of a computer taking the place of a teacher has been around for quite some time. The reality is that, in spite of progress in artificial intelligence, intelligent tutoring systems are not yet ready to replace human instructors.

Do Not Reproduce

33


University of Memphis Tutoring Research Group (developers of AutoTutor) http://www.autotutor.org/ Worcester Polytechnic Institute Tutor Research Group http://web.cs.wpi.edu/Research/trg/ The Intelligent Tutoring Systems Conference is held every two years. The 2006 conference is in Taipei, Taiwan. http://www.its2006.org/cfp.htm The IEEE International Conference on Cognitive Informatics is held every year. The 2006 ICCI conference was held in July in Bejing, China. For more info: http://www2.enel.ucalgary.ca/ICCI2006/ Online Resources The International Artificial Intelligence in Education Society (AIED) is an interdisciplinary community that organizes conferences and publishes a journal on AI in learning. The AIED conferences are held every two years, with the next one in 2007. For more information, go to: http://aied.inf.ed.ac.uk/aiedsoc.html The International Journal of Artificial Intelligence in Education (IJAIED) is the official journal of the International Artificial Intelligence in Education Society (AIED). It publishes papers on applying artificial intelligence techniques and concepts to the design of systems to support learning. http://aied.inf.ed.ac.uk/

The American Association for Artificial Intelligence (AAAI) maintains a listing of Intelligent Tutoring resources. See: http://www.aaai.org/AITopics/html/tutor. html The LICEF Research Centre in Montreal is dedicated to cognitive informatics and training. http://www.licef.teluq.uquebec.ca/eng/in dex.htm For an introduction to intelligent tutoring, see the article by Ong and Ramachandran (2000) in Learning Circuits entitled Intelligent Tutoring Systems: The What and the How. Learn more at: http:/

EmergingTechnologies10-31-06

Documents

Transcript of EmergingTechnologies10-31-06