timothy k. shih © 2006, idea group inc. aphasic...

Journal of Distance Education Technologies, 4(1), 31-45, January-March 2006 31

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc.is prohibited.

ABSTRACT

This article presents the case study of a cooperative Web-learning environment — SOCRATES —to foster barrier-free learning on the Web. While the growth of the Internet was exponential in thelast years, still many communities don’t benefit from Web-learning technology due to impropertools and constricted communication processes. These problems increase when developingapplications for communities of people with special needs. SOCRATES supports a community oflearning comprising patients suffering from aphasia (aphasics), therapists, researchers onlinguistics, and system developers. Aphasics can improve their conversation skills with a speciallydesigned talk/chat tool, while therapists and linguistic researchers can monitor conversationsfrom automatically generated transcripts. Aphasics in remote areas using SOCRATES are nowable to communicate freely among each other without being afraid that they might get lost inrapid conversation in a distributed privacy protecting virtual community.

Keywords: discourse analysis; learning communities; Web-based learning

Aphasic Communitiesof Learning on the Web

Marc Spaniol, Lehrstuhl Informatik V, RWTH Aachen University, Germany

Ralf Klamma, Lehrstuhl Informatik V, RWTH Aachen University, Germany

Luise Springer, Universitätsklinikum, RWTH Aachen University, Germany

Matthias Jarke, Lehrstuhl Informatik, V, RWTH Aachen University andFraunhofer FIT Institute for Applied Information Technology, Germany

INTRODUCTION ANDRELATED WORK

Acquired language disorders are oneof the most frequent and long lasting con-sequences of stroke or traumatic brain in-jury. Aphasics suffer from a language dif-ficulty because of a brain injury, calledaphasia, that affects one or more aspects

of the complex process of comprehendingand formulating verbal messages resultingfrom newly acquired disease of the centralnervous system (Damasio, 1981). Apha-sia is the Greek word for without speech,which is only partially correct, since thebrain injury might affect speech, under-standing, reading, and writing in different

IDEA GROUP PUBLISHING

This chapter appears in the publication, International Journal of Distance Education Technologies Volume 4, Issue 1edited by Timothy K. Shih © 2006, Idea Group Inc.

701 E. Chocolate Avenue, Suite 200, Hershey PA 17033-1240, USATel: 717/533-8845; Fax 717/533-8661; URL-http://www.idea-group.com

ITJ3029

32 Journal of Distance Education Technologies, 4(1), 31-45, January-March 2006


degrees. Depending on location and degreeof the brain injury, those affects are moreor less serious.

Aphasics are not mentally handi-capped. They can think logically and areable to understand and interpret situationscorrectly. As a rule, aphasic disorders aretreated by speech therapists (logopedists)in clinical settings with either inpatient oroutpatient regimen (Huber, Springer &Willmes, 1993). The frequency of prescrip-tion decreases with increasing duration ofthe handicap, even if quality of life remainslow. Even though many patients need high-frequent, long-term therapy, it often can-not be provided for organizational or finan-cial reasons. Therefore, aphasics — espe-cially those living in remote areas — havebegun to organize themselves in self-helpgroups and meet in regional centers likethe aphasia self-help center in Unterfrankenwith whom we are cooperating (http://www.aphasie-unterfranken.de/). Unfortu-nately, meetings can be organized only afew times a year, because transportationcosts are too high and not funded by theGerman public health system. To avoid iso-lation caused by the lack of money, there isa need for additional communication andlearning/therapy means like the Internet.Especially young aphasics have shownstrong interest in the use of technology forcommunication and learning with otherpeople.

If we speak about learning or elec-tronic/Web-based learning (e-learning) inthis article, we address social learning andcommunity building processes by means ofdigital media. Accessibility for Web tech-nologies and online learning has been rec-ognized as an issue of increasing impor-tance for the HCI (Human Computer In-teraction) community and the EuropeanCommunity (Miesenberger, Klaus &Zagler, 2002). However, aphasics seldom

have been considered as a target commu-nity for accessible Web-based tools up tonow.

One of the few existing tools for apha-sia patients is a package of training soft-ware aimed at teaching aphasics the mostimportant everyday words (Lányi et al.,2004). A master thesis at UBC recentlyproposed a tri-modal approach to overcomeword finding problems of aphasics by com-bining images, text, and sound to representwords and concepts in everyday life appli-cations such as a dictionary, recipe book,or daily planner (Moffatt, 2004). Accessi-bility for e-learning environments and itsconsequences for synchronous and asyn-chronous communication and collaborationtools is analyzed in Guenaga, Burger, andOliver (2004). Research on Web standardsfor supporting accessibility can be found inMohamad, et al. (2004).

However, recent research in this areafocuses mostly on individuals and peoplesuffering particular motor deficits (e.g., in-put assistance by developing touchpads forhandicapped or elderly people) (Holzinger,2002), adaptive interfaces based on bio-feedback sensors (Velasco et al., 2004),or PC control by eye movement (Fejtová,Fejt & Lhotská, 2004). Mohamad, et al.(2004) present an approach on pedagogi-cal aspects in applications for children withlearning disabilities. Other approaches aimat voice recognition, facing problems dueto misrecognition, depending on the pro-nunciation (Privat et al., 2002). Similarly,predictive systems as well as auditory userinterfaces up to now lack reliability in rec-ognition (Lauruska & Musteikis, 2002;Willis 2002). The system of Baptiste-Jessel, et al. (2004) tries to support blindpeople to access graphical Web-baseddocuments with a special steering device.A similar approach can be found in Rotard,Otte, and Ertl (2004). Alternatively, Chene



and Hoel (2002) try to transform the con-tent in their system to increase readabilitybut lack usability in interactive systems dueto missing input options.

In summary, most of the systems arebound to costly and sensitive equipment.In contrast, SOCRATES pursues the ideathat many communication barriers are over-come best in cooperative learning commu-nities by bringing together end users, thera-pists, researchers, and developers in a set-ting where barriers are overcome by spe-cially designed media with relatively lowcosts.

The rest of the article is organized asfollows: In the next section, we describethe difficulties aphasics face in using newmedia and e-learning applications, in par-ticular. Then, we explain learning and col-laboration processes in aphasic communi-ties. After that, we introduce our commu-nity-centered software development ap-proach and the chat/talk-based e-learningapplication SOCRATES (Simulation of OralCommunication, Research, Analysis, andTranscription Engineering System) that hasbeen conjointly developed with an aphasiccommunity and its therapists. The articlecloses with a summary and an outlook onfurther research.

DIFFICULTIES OF APHASICSIN USING CONVENTIONALAPPLICATIONS

The design of community systems de-pends on a tight interplay between the or-ganization of knowledge and communica-tive processes within communities of prac-tice (Wenger, 1998). In this article, we donot discuss any distinctions between com-munities of practice and communities oflearning (cf. Klamma, Rohde & Stahl, 2003)for a deeper investigation of the concepts.To increase the usability of learning appli-cations, it is necessary to analyze the kind

of difficulty occurring in learning processes.Basically, we distinguish the following typesof difficulties:

• Media-specific. Influences dependingon the medium itself.

• User-specific. Effects caused by theuser’s deficits.

Usually, a mixture of both kinds ofdifficulties can be recognized in aphasiccommunities. Problems emerge due toword finding, comprehension, and spellingproblems. Those problems are common inother communities, too. However, in com-bination with media-specific disturbances,the situation becomes unmanageable withtraditional chat board systems.

Digital media usage is common formost of us, and so, Web-based applicationslike talk tools (e.g., talk) or chat board sys-tems (e.g., IRC) also are applicable, in gen-eral. In the heart of many applications knownin the field of e-learning, these communica-tion tools are used for supporting learning indistributed groups. However, there are peoplewho face severe problems in using newmedia and conventional applications, in par-ticular. These problems not only are re-stricted to poorly designed user interfacesor insufficient design methodologies but areoften inherent in the medium used by anapplication. In the case of aphasic commu-nities, problems occur in not completely syn-chronous communications due to the disease-related delays in word finding processes ofthese people. These delays can be misinter-preted by other communication partners. Forinstance, they might believe that the com-munication partner is not at his or her key-board anymore. This is a well-known prob-lem in asynchronous communication, whichnormally is overcome by quasi-synchronouschat board systems, because lines of textare released and transmitted over the



Internet to all communication partners im-mediately after pressing the return key.

For aphasics, the process of writinga line of text can take several minutes with-out any notification of still being online toany other communication partner in a con-ventional chat board system. Consequently,any communication process or learning pro-cess is in danger of breaking down. Usingtalk systems for fully synchronous commu-nication (every key pressed is releasedimmediately and transmitted over theInternet) is limited to two communicationpartners.

Our solution combines chat-boardtechnology with a synchronous talk modefor up to four people (two more than inconventional talk systems). In order to sup-port aphasics organizing themselves in acommunity of learning, we have to adoptthese conventional communication tools tomake the conversation difficulties of theaphasics productive, not only by simplycombining the strength of both tools but byrealizing community-of-learning processeson top of the information technology. Some-times, we also call these learning processesrepair processes, because they try to over-come communication problems typical foraphasics, like word finding problems. Whenusing the system, all repair processes areinvoked by community members. In par-ticular, users should not be pushed by thetechnology but should develop the softwaretogether with the developers. In our solu-tion, we are trying to find media mixes thatallow us to overcome these problems bycombining media to increase their benefitswhile simultaneously reducing their deficits.

LEARNING PROCESSES INAPHASIC COMMUNITIES

Learning is a social system within thecommunities of practice (Wenger, 1998)and needs a tight interplay between com-

municative acts and the organization of in-formation. Regarding this, the organizationof knowledge is a structuring element,whereas communicative acts often have adiscursive nature. Empirical studies (Daft& Lengel, 1986; Grote & Klamma, 2000)show that even the mental maps of peoplestrongly depend on the media through whichthey communicate effectively. Bringing to-gether those parts that contradict each otherin principle appears to be a challenge forthe recombination of digital media in per-sonalized community software. Neverthe-less, the problem is that one can’t developconcepts separately from the media thatare used for communication, especially inbarrier-free collaboration among learners.Similar to the assumptions by Paavola,Lipponen, and Hakkarainen (2002), theknowledge-building process can be seen asa combination of Nonaka and Takeuchi’s(1995) thesis with the aspects of difficultytriggered knowledge creation (Engeström,1987) and knowledge building communities(Scaramalia & Bereiter, 1994).

Figure 1 shows a modification ofNonaka and Takeuchi’s (1995) cycle ofknowledge creation adapted to the require-ments of community-oriented learning. Thecontinuous spiral of knowledge creationcomprises four modes of knowledge con-version: socialization, externalization, com-bination, and internalization. Implicit knowl-edge is unconscious and procedural knowl-edge which is hard to formalize or commu-nicate to others. Explicit knowledge is for-mal knowledge that is easy to transmit be-tween individuals and groups. In this model,implicit and explicit knowledge are both im-portant for further knowledge creation bytransforming implicit knowledge into explicitand vice versa, dividing up the process intwo sections.

On the left side of Figure 1, we con-centrate on virtual communities, represent-



ing information that is transferred mostlyby human-human interaction (HHI). So-cialization means implicit information be-tween humans based on the social context.In aphasic communities, the transfer ofimplicit knowledge into explicit knowledgeis most likely to be affected by difficulties,since users often have problems external-izing their opinion due to word finding andspelling problems. Consequently, this prob-lem has to be solved, and it is up to re-searchers to gain productivity from the dif-ficulties by initiating monitored repair pro-cesses in SOCRATES. For therapists andresearchers, knowledge within the commu-nity is externalized when conversation pro-tocols are stored in a repository, thus mak-ing difficulties in communication accessiblefor further evaluation.

On the right side of Figure 1, digitalcommunity media represent structured andcategorized information that has been madeexplicit. Combinational operations includestorage, transcription, and retrieval of con-versation protocols. That is the cutting edge,

where virtual communities and digital com-munity media meet. Learning, as observedby internalization, has different meaningsfor the different types of community mem-bers. For the patients, it is a self-initiatedor community-initiated therapeutic meanspossibly amplified by improved human-com-puter interaction. Therapists gain newtherapy concepts by analyzing the tran-scripts and researchers’ new insights intothe mechanisms of language production. Inseveral workshops, aphasics and the de-veloper of SOCRATES discussed require-ments for the next releases of the software.For both sides, this was a real learning ex-perience. In software development, partici-patory methods are well known, but inclu-sive strategies for handicapped people insoftware development are still rather new.Software developers could only understandafter meeting their customers why designwishes or functionalities were so unusualand different. For most aphasics, it was thefirst time that researchers and developersseriously asked them for input. After a

Figure 1. Communication and learning in aphasic communities (adapted fromNonaka & Takeuchi, 1995)



while, they contributed a lot of features thatwere discussed intensively in the work-shops, and some of them already have beenimplemented by us.

SOCRATESCommunication difficulties are com-

mon in our daily life. When designing com-munity software, it is necessary to analyzethe nature of a difficulty in order to adjustthe repair strategies. With respect to apha-sic communities, novel media mixes shouldincrease their strength by simultaneouslyreducing their deficits. SOCRATES com-bines digital media to make difficulties insingle media usage less visible. On the onehand, it creates a meeting point for theaphasics, and on the other hand, the sys-tem also can be used by therapists and re-searchers for a further analysis on conver-sation structures. Research in oral aphasiccommunities is still in its infancy, which iseven more true for simulated orality like on

chat boards. For linguistic research andtherapy development, for example, one ofthe most interesting features is automaticextraction of time delays between pressedkeys, because this might indicate the gradeof disability.

Figure 2 gives an overview of theSOCRATES system. The left side showsthe screen dumps of the Java client appletscommunicating via Apache relay servers.The relay servers are used to coordinateconversations and to enrich them with ad-ditional metadata, which is necessary forfurther computation. In the case ofSOCRATES, time delays between lettersto be spelled are especially captured. Thedigital community media are captured inXML trace files to be placed in the con-versation repository. The underlying com-munity software architecture ATLAS (Ar-chitecture for Transcription, Localization,and Addressing Systems) is a communitymanagement system especially handling

Figure 2. SOCRATES community support and analysis software



digital multimedia created by communitiesusing networked information systems(Spaniol, Klamma, & Jarke, 2003). Toevaluate the dialogs, transcripts and statis-tics are generated from XML files (e.g., inthe Adobe PDF format). The SOCRATEStalk/chat protocol is similar to Internet Re-lay Chat (IRC) (Oikarinen & Reed, 1993),which offers a good possibility for a real-time, text-based communication in com-munities. An IRC server relays all incom-ing messages to the other participants tak-ing part in the same conversation. Servercode has been added to our chat applica-tion, which is partially located on the re-lay server. The SOCRATES core is simi-lar to other user interface applications inthe IRC area but adapted for people withconcentration deficits. Triggered by userinteraction, the system creates certain mes-sages that are passed to a dispatcher. Thedispatcher invokes a callback function,which processes the message and performscorresponding actions like creating, cancel-ing, or changing files.

To reduce the need for user-side com-puting resources, SOCRATES clients aredesigned as thin applications and only usedfor presentation and filtering of events gen-erated by the human-computer interaction.Where events require the execution ofsome server code, it is being sent to therelay server. The processing of such mes-sages on the server side is done by the dis-patcher of the relay server. The type ofmessage is checked, corresponding func-tions are called, and actions are performed,if needed. Message processing betweenclient and server is performed asynchro-nously. This means that the client is notblocked after having sent a message but isable to perform other actions like receiveother messages as a result of actions ofthe other users. In addition, the client maydiscard messages due to a new state of

the system. Access to SOCRATES isgranted by a login procedure in order tomake the system exclusive to persons suf-fering from aphasia, their therapists, andresearchers. The underlying communitysoftware ATLAS handles message ex-change by using Secure Socket Layers(SSL). So, members are protected frompeople who might molest or amuse aboutaphasic stories. Even more, the system be-comes somewhat unique, since only thoseare allowed to participate who share a com-mon value. To provide that, passwords aregiven only to those persons who are activein or known to self-help groups.

Designing Media Combinations forAphasic Communities of Learning

Figure 3 shows a multi-user talk inthe SOCRATES Java applet. Its design issimilar to conventional chat boards, but itallows aphasics to initiate a decoupled con-versation of up to four people in case theyencounter difficulties to participate in thechat-board communication. Our approachtries to overcome the word finding and spell-ing problems of aphasics by invoking thecommunity in its learning process.

This process is initiated by an apha-sic pushing the button on the right side ofthe applet, indicating the interest in startinga Talkrunde (Eng. multi-user talk). Afterlaunching a multi-user talk, a screen at thebottom of the Java client applet is high-lighted. This screen offers a synchronouscommunication of up to four persons simi-lar to a face-to-face talk. For keeping fo-cus on the user’s own window, the chatwindows of other participants are coloredbluish. The command line is automaticallyset to the lower section on the screen, al-lowing input for the multi-user talk. In themeantime, all other chatters are informedof the newly initiated multi-user talk by dis-playing a designated sign behind the



aphasic’s name. Those chatters willing tojoin the conversation click on the sign toparticipate. Due to the synchronicity in amulti-user talk, the others are now awareof any difficulties in the spelling processand may help to find the right words byproposing them. The activation of aSOCRATES multi-user talk is described indetail in Table 1.

Computer-Assisted CommunityLearning Analysis in SOCRATES

In SOCRATES, XML files captureall necessary information to cover the con-text of conversation (i.e., participants, glo-bal time, delays, and the conversation it-self). By keeping compatibility to XML,SOCRATES data easily can be interpretedby applications compliant to the XML stan-dard. Moreover, automatic processing ofthe files is backed by their hierarchical

structure. Hence, queries can be used toaccess selected elements of a documentvia XPath/XQuery (Berglund et al., 2002).Conversations in SOCRATES can be con-verted automatically into different outputmedia, allowing the therapists to adapt theirtherapy to the user’s individual needs. Fora better traceability of the conversationprocess, the transcripts contain additionalinformation about the spelling process, likedelays and repairs. The transcripts are anextension of the analysis standard in lin-guistics (GAT) (Selting, 1998) for writtentext pool (Springer, 2004). They cover writ-ing delays of more than one second, timeof conversation, corrections, usage ofsmileys, and the nicknames of the partici-pants in addition to the spelled words.

An example of an automatically gen-erated transcript from a SOCRATES XMLfile is shown in Figure 4. Each interaction

Figure 3. SOCRATES: Combining chat and multi-user talk (in German)



Table 1. SOCRATES activation protocol

1a) A user pushes the 'Gesprächsrunde' (Eng. multi-user talk) button. 2a) The callback function is called by the GUI. 3a) A new message is prepared and being sent to the relay server. 4a) The message is received by reading thread and queued for further processing. 5a) A processing thread removes the message from the queue and passes it to the dispatcher. 6a) The dispatcher checks the type of the message and calls the suitable callback function. 7a) A new conversation is being created, and the user is added as a participant, which means

that corresponding records are created in the database. 8a) A message is sent containing information about a newly created multi-user talk. 8b) To all other participants currently active on the chat board, a message is sent that a multi-

user talk has been started. 9ab) The client receives the message and processes it. 10a) The function tries to connect the user to this conversation. 10b) All other users can now recognize the newly started multi-user talk by highlighting a

corresponding sign behind the initiator’s name. 11a) If the connection is successful, the fields at the bottom of the chat applet get visible, and the

multi-user talk starts. 12b) Joining the multi-user talk by others is done by clicking the designated sign. 13b) Steps 2a)-6a) are executed for users joining the multi-user talk. 14b) The function tries to add the user to this conversation, since it is limited to four participants.

If this was successful, the last two steps are executed. 15b) A message is sent to the client containing all relevant information (IDs, port, etc.). 16b) The last steps are the same as 9a)-11a) but now on the helper’s side. Legend: a) Indicates actions of the user initiating a multi-user talk b) Indicates actions of users joining the multi-user talk

Figure 4. Conversation XML file and its Adobe PDF transcript (in German)



with the chat system is contained in thetranscript as well as the delays while typ-ing. In addition, therapists and researcherscan replay the whole conversation basedon the information contained in theSOCRATES XML transcript to (re-)expe-rience the discourse situation. Hence, theynow can seek recurring patterns of spell-ing problems and monitor the learning pro-cess of individuals and their community viaa Web form displaying statistical data. Inaddition, via the Web interface, therapistsand researchers are able to access the sta-tistical information about the overall sys-tem or an individual community member.In Figure 5, the evaluation of TobiasMeyer’s activities (name changed) withina period of about the four months is beingdisplayed. By comparing this informationwith those spanning other time intervals,

therapists can gain valuable evidencewhether results of the therapy indicate im-provements of the disease pattern. Surely,this information is used for research andtherapy related issues only. In the work-shops, this has been discussed with theaphasics, and there was consensus thatthese data are important for the furtherdevelopment of the system and should betracked for that purpose.

CONCLUSION AND OUTLOOKIn this article, we presented our

SOCRATES chat/talk environment foraphasic communities. Currently, we arecooperating with logopedists andneurolinguists from the clinical center ofRWTH Aachen University and psycholo-gists of the aphasia self-help center inUnterfranken, Germany (http://

Figure 5. Statistical information about SOCRATES usage (in German; namechanged)



www.aphasie-unterfranken.de/). Commu-nity members come mostly from Aachen,Cologne, and Unterfranken, all located inGermany. SOCRATES usage and learningexperiences are very promising. One of theparticipants suffering mild aphasia and se-vere speech apraxia wrote in his book(Grefe, 2004) (translated into English bythe authors):

With chat we are all on the same level,and the thing that’s important in my eyesis that we are amongst ourselves andunderstanding each other. [...] As personsuffering speech apraxia it’s good forme to be able to concentrate on writingand not pronunciation. A chatting sideeffect is that to keep up you have to beable to read fast. Initially, I had a lot ofproblems with that, but now it’s more orless OK. That’s why I think this kind of“therapy” is a very good idea.

Since SOCRATES is an access-au-thorized community system, our softwareis an extension of conventional chat boardtechnology to fit the requirements of anaphasic community. Therefore, the userinterface was designed in cooperation withaphasics and their therapists. Design issueswere discussed in workshops bringing to-gether aphasics, therapists, and computerscientists, giving us, as the developers ofthe SOCRATES system, valuable insightsand never before considered problems inthe usage of digital media. Resulting fromthese workshops, several changes in theapplication were integrated. Some of thesefindings are discussed in the following.

The user interface functionalities werereduced to a minimum to avoid aphasicsgetting lost in SOCRATES due to informa-tion overload. Size and colors of the letterson the chat board were adapted to theaphasics’ needs. Also, the naming of but-

tons has been adapted to the community’sinterests. For instance, the button Talkrunde(Eng. multi-user talk) was initially namedHilfsgespräch (Eng. help conversation); thisled to unsatisfying usage, since most par-ticipants were embarrassed to push it. An-other change derived from the workshopswas to integrate smiley buttons in the sys-tem. The reason was twofold. Aphasicswith a minor degree of disability liked touse smiley symbols frequently for fun;aphasics still suffering from a higher de-gree of aphasia wished to have the buttonsintegrated into the system for a more seri-ous reason: they wanted to express theiremotions quickly without writing long sen-tences and probably getting lost in overlyfast conversation.

An important aspect of introducing anew system to aphasic communities waspersonal support. In the beginning, supportwas needed to promote the system and fortroubleshooting technical aspects (e.g., theinstallation of a browser), since theaphasics in the communities we workedwith had a tendency to give up quickly incase of any malfunction. Now,SOCRATES is self-promoted by the mem-bers of the community as their own devel-oped system and giving support in case oth-ers encounter problems using it. In themeantime, SOCRATES has become verypopular in the aphasics community, enablingthem to communicate freely with otherswithout being afraid that they might get lostin rapid conversation. Since they have theoption to indicate their need for a conver-sation in a smaller circle, other communitymembers become aware of their media-dependent difficulties. Participants are nowable to meet others in a privacy-protectingvirtual community. By integrating other self-help groups of aphasics nationwide intoSOCRATES, we hope to support evenmore aphasics in the near future.



In SOCRATES, the emerging prob-lems of an exclusive usage of a not com-pletely synchronous digital medium (chat)— interrupts based on word-finding pro-cesses of aphasics — can be made visibleto other community members and therebyrepaired by invoking the functionalities ofthe SOCRATES system. By making theiractions transparent, community membersare addressed to participate and collabo-rate. When coupling the medium of a chatand a talk, media-related difficulties alsoare made visible. This concept is new, sinceso far, chat tools as well as talk toolsworked perfectly as stand-alone applica-tions for non-handicapped users, but thesynergetic power of their combination isnow advantageous for aphasics. In addi-tion, the tracing of conversations allows lin-guistic researchers and therapists to ana-lyze sessions and to customize users’ thera-pies. Due to the options given by the XML-based trace file, therapists can select fromversatile media formats for further analy-sis. In addition to statistics and replays offace-to-face talks, SOCRATES offers tran-scripts in XML, Adobe PDF, and RTF for-mat. By relying on XML as interchangeformat, we preserve flexibility for furtherdevelopments.

The next step will be to develop adap-tive user interfaces to make the contentpresentation dependent on predefined set-tings and users’ needs. Hence, we are do-ing research on optimizing the comfort inhuman-computer interaction and have thusstarted joint research projects to improveaddressing and personalization in our com-munity software (Spaniol, Klamma, &Jarke, 2003). We hope to detect a modelcore common to different learning pro-cesses, which can be extended to fulfillspecific requirements in a versatile area ofapplications.

ACKNOWLEDGMENTSThis work was supported by the Ger-

man National Science Foundation (DFG)within the collaborative research centerSFB/FK 427 “Media and cultural commu-nication” and by the 6th Framework ISTprogramme of the EC through the Net-work of Excellence in Professional Learn-ing (PROLEARN) IST-2003-507310.We’d like to thank our colleagues for theinspiring discussions.

REFERENCESBaptiste-Jessel, N., Tornil, B., & Encelle,

B. (2004). Using SVG and a force feed-back mouse to enable blind people toaccess “graphical” Web based docu-ments. In Proceedings of the ICCHP2004 (pp. 228-235), Paris, France.

Berglund, A., et al. (2002). XML path lan-guage (XPath) 2.0. Retrieved April2005, from http://www.w3.org/TR/xpath20/

Chene, D., & Hoel, M. (2002). Web siteaccessibility auditing tool for visuallydeficient persons OCAWA. In K.Miesenberger, J. Klaus, & W. Zagler,(Eds.), Proceedings of the ICCHP2002 (pp. 27-32), Linz, Austria.

Daft, R. L., & Lengel, R. H. (1986). Or-ganizational information requirements,media richness and structural design.Management Science, 32(5), 554-571.

Damasio, A. R. (1981). The nature of apha-sia: Signs and syndromes. In M. T. Sarno(Ed.), Acquired aphasia (pp. 51-64).New York: Academic Press.

Engeström, Y. (1987). Learning by ex-panding. Helsinki, Finland: Orienta-Konsultit Oy.

Fejtová, M., Fejt, J., & Lhotská, L. (2004).Controlling a PC by eye movements: TheMEMREC project. In Proceedings ofthe ICCHP 2004 (pp. 770-773), Paris,France.



Grefe, U. (2004). 3 + 4 = 8 Vergrabenund Verschüttet Sind Meine Worte!(2nd ed.) Idstein: Schulz-Kirchner-Verlag (in German).

Grote, K., & Klamma, R. (2000). Mediaand semantic relations: Comparison ofindividual and organizational knowledgestructures. In Proceedings of theDagstuhl Seminar Knowledge Man-agement. An Interdisciplinary Ap-proach (pp. 9-14), Schloß Dagstuhl,Germany.

Guenaga, M. L., Burger, D., & Oliver, J.(2004). Accessibility for e-learning en-vironments. In Proceedings of theICCHP 2004, Paris, France.

Holzinger, A. (2002). User-centered inter-face design for disabled and elderlypeople: First experiences with designinga patient communication system(PACOSY). In Proceedings of theICCHP 2002, Linz, Austria.

Huber, W., Springer, L., & Willmes, K.(1993). Approaches to aphasia therapyin Aachen. In A. L. Holland, & M. M.Forbes (Eds.), Aphasia treatment:World publishing. San Diego: SingularPublishing Group.

Klamma, R., Rohde, M., & Stahl, G. (2004).Community-based learning: Explorationsinto theoretical groundings, empiricalfindings and computer support. Introduc-tion to the Special Issue, SIGGROUPBulletin, 24(3), 7-13.

Lányi, S. C., Bacsa, E., Mátrai, R.,Kosztyán, Z., & Pataki, I. (2004). Thedesign question of development of mul-timedia educational software for apha-sia patients. In Proceedings of theICCHP 2004, Paris, France.

Lauruska, V., & Musteikis, T. (2002). De-veloping of predictive communicationsystem for the physically disabledpeople. In Proceedings of the ICCHP2002, Linz, Austria.

Miesenberger, K., Klaus, J., & Zagler, W.(2002, July 15-20). In Proceedings ofthe 8th International Conference onComputer Helping People with Spe-cial Needs (ICCHP 2002), Linz, Aus-tria.

Moffatt, K. (2004). Designing technol-ogy for and with special populations:An exploration of participatory de-sign with people with aphasia.Master’s thesis, University of BritishColumbia, Vancouver, Canada.

Mohamad, Y., Stegemann, D., Koch, J., &Velasco, C. A. (2004). Imergo: Support-ing accesibility and Web standards tomeet the needs of the industry via pro-cess-oriented software tools. In Pro-ceedings of the ICCHP 2004 (pp. 310-316), Paris, France.

Mohamad, Y., Velasco, C. A., Damm, S.,& Tebarth, H. (2004). Cognitive train-ing with animated pedagogical agents(TAPA) in children with learning disabili-ties. In Proceedings of the ICCHP2004 (pp. 187-193), Paris, France.

Nonaka, I., & Takeuchi, H. (1995). Theknowledge-creating company. Oxford:Oxford University Press.

Oikarinen, J., & Reed., D. (1993, May).Internet relay chat protocol. InternetNetwork Working Group RFC 1459.Retrieved from ftp://ftp.isi.edu/in-notes/rfc1459.txt

Paavola, S., Lipponen, L., & Hakkarainen,K. (2002). Epistemological foundationsfor CSCL: A comparison of three mod-els of innovative knowledge communi-ties. In Proceedings of the Computer-Supported Collaborative Learning2002 Conference (pp. 24-32), Hillsdale,NJ: Erlbaum.

Privat, R., Vigouroux, N., Truillet, P., &Oriola, B. (2002). Accessibility andaffordance for voice interactive systemswith the VoiceXML technology. In K.



Miesenberger, J. Klaus, & W. Zagler(Eds.), In Proceedings of the ICCHP2002 (pp. 61-63), Linz, Austria.

Rotard, M., Otte, K., & Ertl, T. (2004). Ex-ploring scalable vector graphics for vi-sually impaired users. In Proceedingsof the ICCHP 2004, Paris, France.

Scaramalia, M., & Bereiter, C. (1994).Computer support for knowledge-build-ing communities. Journal of the Learn-ing Sciences, 3(3), 265-283.

Selting, M., et al. (1998). GesprächsanalytischesTranskriptionssystem (GAT) (In German).Linguistische Berichte, 173, 91-122.

Spaniol, M., Klamma, R., & Jarke, M.(2003). ATLAS: A Web-based softwarearchitecture for multimedia e-learningenvironments in virtual communities. InProceedings of the ICWL 2003,Melbourne, Australia.

Springer, L. (2004). Mediendifferenz undSprachverwendung - Eine empirischeStudie zur medienspezifischenProzessierung des Sprachwissens beiAgrammatikern und Sprachgesunden (inGerman). In L. Jäger, & E. Linz (Eds.),

Medialität und Mentalität -Theoretische und empirische Studienzum Verhältnis von Sprache,Subjektivität und Kognition (pp. 221-249). München: Fink-Verlag.

Velasco, C. A., et al. (2004). IPCA: Adap-tive interfaces based upon biofeedbacksensors. In Proceedings of the ICCHP2004, Paris, France.

Wenger, E. (1998). Communities of prac-tice: Learning, meaning, and identity.Cambridge, UK: Cambridge UniversityPress.

Willis, T., Pain, H., Trewin, S., & Clark, S.(2002). Informing flexible abbreviationexpansion for users with motor disabili-ties. In Proceedings of the ICCHP2002, Linz, Austria.

Marc Spaniol is a doctoral researcher in the Information Systems group at RWTHAachen University. Within the Collaborative Research Center on “Media andCultural Communication”, he works on the project on “Impacts of NetworkedMultimedia Information Systems on Cooperation and Organization of Knowledgein Cultural Science Communities”. He holds a diploma in computer science. Hisresearch interests include e-learning and multimedia information systems forcommunities, multimedia metadata standards such as MPEG-7, and cross-mediaanalysis.

Ralf Klamma is a senior researcher within the Information Systems group at RWTHAachen University. He has visited the Massachusetts Institute of Technology,Cambridge and has been a substitute professor at the universities of Chemnitz andPassau. He is project manager in the collaborative research center “Media andCultural Communication” and core member of the European Network of ExcellencePROLEARN in professional training. His research interests include theory and



utilization of information systems, organizational memories and workflowmanagement, virtual community support, electronic learning, and professionaltraining.

Luise Springer is head of the School of Logopedics at the hospital of RWTH AachenUniversity and works as a researcher within the collaborative research center on“Media and Cultural Communication”. One of her projects deals with “Computer-mediated Communication in Aphasic Communities”. Also, she is involved inneurolinguistic and media-specific projects at the Departments of Medicine andPhilosophy at RWTH Aachen University. Her research focuses on media-specificlanguage performances of healthy and aphasic persons. In addition to this, she isconcerned with the evaluation of pragmatic and model-orientated therapyapproaches in neurogenic disorders.

Matthias Jarke is professor of information systems at RWTH Aachen Universityand director of the Fraunhofer FIT Institute for Applied Information Technologyin Sankt Augustin, Germany. Jarke studied computer science and businessadministration at the University of Hamburg and held faculty positions in NewYork University and the University of Passau prior to joining RWTH in 1991. Hisresearch interest is information systems support for cooperative applications inengineering, business, and culture. Jarke served as chief editor of informationsystems from 1993-2004 and is currently president of GI, the German Informaticssociety.

timothy k. shih © 2006, idea group inc. aphasic...

Documents