<p>This thesis presents an approach for automatically generating signing animations from a sign language notation. An avatar endowed with expressive gestures, as subtle as changes in facial expression, is used to render the sign language animations. SWML, an XML format of SignWriting is provided as input. It transcribes sign language gestures in a format compatible to virtual signing. Relevant features of sign language gestures are extracted from the SWML. These features are then converted to body animation pa- rameters, which are used to animate the avatar. Using key-frame animation techniques, intermediate key-frames approximate the expected sign language gestures. The avatar then renders the corresponding sign language gestures. These gestures are realistic and aesthetically acceptable and can be recognized and understood by Deaf people.</p>

本研究使用眼動儀器與神經科學的分析方式，探討網站設計時運用虛擬替身與個人化推薦而設計出不同的個人化購物網站，是否會影響使用者對網站親密度與產品購買意願上有不同感受。研究問題共有三個：1.網站個人化設計對網站親密度的影響程度。2.網站個人化設計對於產品購買意願的影響程度。3.使用者是否會注意到網站的個人化設計。研究結果發現個人化推薦會對網站親密度及購買意願產生正向影響，而虛擬替身則只會對親密感產生正面的影響。虛擬替身及個人化推薦都會提高消費者的對該區域的首次瀏覽時間。 / The purpose of this study is to use eye tracking analysis to investigate whether different designs of shopping websites using personalization and avatar will influence users’ website intimacy and product buying attention. First, we want to know the relationship between personalized design and website intimacy. Second, we want to know the relationship between personalized design and product buying attention. The last one is whether users pay attention to personalized designs on websites. An experiment was conducted to evaluate the effect of personalized recommendation and avatar. The results indicate that personalization had positive impacts on users’ perceived intimacy, and users’ total gaze duration has positive effects on both website intimacy and product buying attention.

The dynamic development of new technologies influences consumers in many different ways reaching far beyond the shift in consumption patterns, challenging the way consumers live their lives. The role of new information technologies is continually growing in our daily lives changing the way we see the self and the world around us. Consequently, the advent of the computer culture incites a radical rethinking of who we are and the nature of being human, which clearly illustrates the postmodern age. As a result, over the past decades consumer research has moved away from simply viewing consumers as information processors to consumers as socially conceptualized beings. This Consumer Culture Theory (CCT) movement views consumers and consumer behaviour as articulations of meanings and materiality within the productive of complex cultural milieu. This ethnographic thesis focuses on the three-dimensional virtual world of Second Life, which is a 'Real Life' simulation and where the residents represent themselves through 'avatars', creating a kind of virtual materiality. This raises interesting questions for consumer researchers, not just about how consumption is enacted, produced and articulated within this environment, but also in relation to theoretical and methodological issues. More specifically, this thesis critically examines the development of interpretive consumer research and the emergence of the Consumer Culture Theory framework in the context of the juxtaposition of reality and hyperreality and takes a position which goes beyond the 'body in the net/physical body' binary. Therefore, this thesis places the 'avatar-as-consumer' at the centre of the research focus. The current thesis develops a theoretical framework which examines the role of consumption in resolving key paradoxes. Moreover, it extends the netnography framework from mainly text based research to the visual characteristics of virtual worlds so that it can be useful for the study of complex online environments and as a result, how the role of the researcher goes beyond netnography to virtualography is discussed.

<p>This thesis presents an approach for automatically generating signing animations from a sign language notation. An avatar endowed with expressive gestures, as subtle as changes in facial expression, is used to render the sign language animations. SWML, an XML format of SignWriting is provided as input. It transcribes sign language gestures in a format compatible to virtual signing. Relevant features of sign language gestures are extracted from the SWML. These features are then converted to body animation pa- rameters, which are used to animate the avatar. Using key-frame animation techniques, intermediate key-frames approximate the expected sign language gestures. The avatar then renders the corresponding sign language gestures. These gestures are realistic and aesthetically acceptable and can be recognized and understood by Deaf people.</p>

This paper describes the process of producing avatars for Jiddr. Today, social media surrounds the lives of youth, and communication via the Internet is an everyday occurrence for most people. Jiddr is an app that was created to give youths a platform for engaging with adults and other youths. It is a moderated forum which gives young people the opportunity to come i contact with adult role models who are knowledgable within various subject areas. On Jiddr, users are anonymous, and this is something I have placed much importance on during the design process. The relationship between critical design and interaction design has been central to this work, and I have created a concept as well as several avatars, which are presented in a prototype, based on previous studies that focus on anonymity. / Denna rapport redogör för processen att ta fram avatarer till föreningen Jiddr. Sociala medier omger ungdomar idag och kommunikation över internet hör till vardagen för de flesta. Jiddr är en samtalsapp som har skapats för att ge ungdomar en plattform där de kan prata med vuxna och andra ungdomar. Ett modererat forum som ger ungdomar möjligheten att möta vuxna förebilder som är kunniga inom olika ämnesområden. På Jiddrär användarna anonyma och detta är något jag lagt vikt vid under skapandeprocessen. Relationen mellan kritisk design och interaktionsdesign har varit central under arbetet. Och utifrån förstudier med fokus på anonymitet har jag skapat ett koncept och några avatarer som presenteras i en prototyp.

The recent leaps in virtual reality (VR) technology have unleashed revolutionary potential for applications in a wide variety of areas, including education, training, psychological-therapy, etc. As part of the effort on understanding how users interact with VR, I focused on studying full-body joint action using a road crossing task which involves perception, decision-making, action, and joint action. I have been heavily involved in the design, implementation, and construction of two large-screen, room-like stereoscopic virtual environment (VE) simulators. Using this system, I developed a three-part research plan with a series of studies to examine how people engage in full-body joint-action with a partner under three scenarios: 1. two people who are physically present in a co-occupied virtual environment; 2. one person who shares a virtual environment with a computer-generated agent (CG agent); 3. two people who share the same virtual environment remotely in physically separate places where each person is motion tracked and presented in the environment as a graphic avatar. The behaviors of participants were recorded and processed through a customized pipeline that captures important performance metrics, such as how participants pick crossable gaps and time their movements. The VE system, user study designs, and findings are introduced in this dissertation.

Active listening is a communication technique where the listener listens to the speaker carefully and attentively by confirming or asking for more details about what they heard during a conversation. Active listening has proven to be a very useful technique in learning and extracting insightful information from the conversation and poses many challenges when emulating such behaviours in a virtual character. In this research, various active listening behaviours are explored according to context, so that these behaviours can be adopted by Embodied Conversational Agents (ECA’s) to be perceived as more believable humanistic characters. This publication describes those behaviours in-depth by gathering insights via expert Interviews. An experiment was carried out with 20 participants (12 men and 8 women) to identify the implemented behaviours and their overall effect during a generic conversation with the ECA. The results showed that the participants were able to identify most of the implemented active-listening behaviours like Interested/inquisitive, agreement/approval (head nod and corresponding sounds), and provided insights for further improvement of the interlocution. / Aktivt lyssnande är en kommunikationsteknik där lyssnaren lyssnar på talaren noggrant och uppmärksamt genom att bekräfta eller be om mer information om vad de hörde under ett samtal. Aktivt lyssnande har visat sig vara en mycket användbar teknik för att lära sig och extrahera insiktsfull information från konversationen och ställer till många utmaningar när man emulerar sådana beteenden i en virtuell karaktär. I denna forskning utforskas olika aktiva lyssnandebeteenden i enlighet med sammanhang, så att dessa beteenden kan antas av Embodied Conversational Agents (ECA) för att uppfattas som mer trovärdiga humanistiska karaktärer. Den här publikationen beskriver dessa beteenden på djupet genom att samla in insikter via expertintervjuer. Ett experiment genomfördes med 20 deltagare (12 män och 8 kvinnor) för att identifiera de implementerade beteendena och deras totala effekt under ett allmänt samtal med ECA. Resultaten visade att deltagarna kunde identifiera de flesta av de implementerade aktivt lyssnande beteendena som Intresserad/nyfiken, överenskommelse/godkännande (huvudnicka och motsvarande ljud), och gav insikter för ytterligare förbättring av samtalet.

This paper presents the reasoning behind and the results from usability testing of the user- and administrative side of FIPIC (Flexible Interactive Platform for Interviews with Children), a digital questionnaire tool for children. The user side was tested with 24 elementary school children between grades 1-6. They first chose which avatar and voice they preferred and then answered six questions using FIPIC while being timed on each question. The administrative side was tested with five researchers from different universities in Sweden using a scripted list of user tasks and the think-aloud method. The researchers would then answer some follow up questions after their test session. The results showed that the user side did not have any major issues that significantly affected the usability of FIPIC. The biggest issues were the non-clickable emoji’s under the slider function and having to scroll up and down. The solution is to make the emoji’s clickable and replace the drop down with a text field to avoid the scrolling. The results of the administrative side showed that most of the usability issues were present in the newer functionalities such as question groups and group connections. Functionalities that are not present in other common questionnaire tools. The suggested improvement is to make the design more intuitive. For example, the create question group button should be placed at the top of the questions page instead of having to first click the create question button. This makes hierarchical sense because a group is needed in order to create a question.

The purpose of this research study was to explore the impact of virtual character design on user emotional experience and user behavior in a simulated environment. With simulation training increasing in popularity as a tool for teaching social skills, it is essential that social interactions in virtual environments provide authentic opportunities for practice (Swartout et al., 2006). This study used Interactive Performance Theory (Wirth, 2012) to examine the effect of designing a virtual buddy character with ineffective traits instead of effective or expert traits. The sample population for this study (n = 145) consisted of first year university students enrolled in courses in the fall of 2013 at the University of Central Florida. Data on participant emotional experience and behavior were collected through questionnaires, researcher observations, and physiological signal recording that included participant heart rate and galvanic skin response. Data were analyzed using multivariate analysis of variances (MANOVA), Kruskal-Wallis one-way analysis of variance, and qualitative thematic coding of participant verbal behavior and written responses. Results of the analysis revealed that participants who interacted with an ineffective virtual buddy character had statistically significant higher averages of verbal statements to the antagonist in the simulated environment and statistically significant lower perceptions of antagonist amiability than participants who interacted with an effective virtual buddy. Additionally, participants who interacted with a virtual buddy of the opposite gender gave statistically significant higher ecological validity scores to the simulated environment than participants who interacted with a virtual buddy of the same gender. Qualitative analysis also revealed that participants tended to describe the female buddy character with more ineffective traits than the male buddy character even though effective and ineffective design conditions were equally divided for both groups. Further research should be conducted on the effect of virtual buddy character design in different types of simulation environments and with different target audiences.

In teacher preparation, more effective pathways and practices are needed for preparing, placing, and supporting beginning teachers and principals (Darling-Hammond, 2010; U.S. Department of Education, 2009b). A common issue in the field of special education is the lack of skill transfer from one setting to another (Dieker, Hynes, Hughes, & Smith, 2008). It has been posited that “practicing up” is not ethical in that novice teachers must attempt to teach with a limited knowledge of appropriate pedagogy and skill (Dieker et al., 2008). The new challenge becomes finding an effective mechanism that provides essential learning experiences and opportunities to refine teaching techniques to the highest standards of fidelity in a safely controlled and coordinated environment (Odom, 2009). Perhaps because of the ethical concerns in honing teacher skills on actual children, and despite the strong demand for professionals who are trained in discrete trial teaching (DTT), few studies have been published on training methodologies and fidelity of implementation (Fazzio, Martin, Arnal, & Yu, 2009). The training of teachers to implement evidence-based interventions such as DTT with fidelity while they are working with students with ASD cannot be over-emphasized in a teacher preparation program (Scheuermann, Webber, Boutot, & Goodwin, 2003; Simpson, 2004; 2005). The researcher utilized the TLE TeachLivE simulation classroom laboratory at the University of Central Florida as a mechanism to infuse cutting-edge technology and learning activities within program/project coursework. The TLE TeachLivE virtual classroom serves as a venue for pre-service and practicing teachers to safely gain proficiency and iv enhance fidelity of implementation of evidence-based practices. Participants experienced an immersive, real-time environment that featured interactions with Austin, an avatar that portrayed a student with autism. A trained interactor remotely controlled the behavior and responses of the avatar with which teachers engaged for practice sessions consisting of ten discrete trials. Between sessions, participants received individualized clinical coaching (ICC) on their performance. Upon termination of the intervention, two generalization probes were conducted to measure retention of fidelity over time within actual classroom settings. Results indicated that all five participants strongly benefited from learning DTT with ICC in the TLE TeachLivE learning platform. Across participants, the overall mean gain in fidelity from baseline phase (14%) to intervention phase (80%) was 66%. The fidelity means of participants in the generalization phase held to 90%, thus supporting the use of virtual environments for teacher preparation. Mean time among participants to attain higher than mastery level performance of 90% in intervention was 1.25 hours. This resulted in shorter training times than previously researched training programs (Arnal et al., 2007; Fazzio et al., 2009; Leblanc, & Luiselli, 2005; Thiessen et al., 2009). This investigation endeavored to reduce the potential of diminished pupil learning gains as a necessary consequence when honing skills in the appropriate delivery of instruction of discrete trial teaching.