This post contains a research proposal I wrote in 2008 for a project to involve virtual worlds that I never completed. This proposal would be changed to become the main experiment I did with the Roskilde University Virtual Worlds Research Group. If anyone is interested in building off this proposal, then please contact me.

Sense-making and users’ experiences with avatars: How knowledge is constructed of virtual worlds through virtual embodiment

Virtual worlds are moving beyond the realm of entertainment and leisure, such as online gaming and social interaction, and into areas that are traditionally information oriented[1]. Avatars are as their etymological origin proclaims; the embodiment of an individual in a virtual world that grounds users visually to these worlds (Bailenson & Blascovich, 2004). Avatars can serve as a replacement for complete virtual reality immersion that requires bulky interfaces to completely submerge users’ senses in virtual worlds. The literature assumes avatars are figures users identify as being like themselves or like the selves they’d like to be. Either way, it is believed this identification heightens users’ feeling of immersion in virtual worlds, thus facilitating their performance in the virtual world (Reinhard, 2005).

As of now, most research on avatars has focused on one of two things. Designers have sought to perfect links between users and avatars by manipulating avatar appearance and user control of avatars (ex. Kang & Yang, 2006; Nowak & Rauh, 2005). Media effects researchers have studied how the presence, appearance and actions of avatars impact users’ involvement in virtual worlds (ex. Yee & Bailenson, 2007; Yee, Bailenson, Urbanek, Chang & Merget, 2007). These studies have aimed at unearthing causal relationships between uses of avatars and performance in virtual worlds. Most studies have used experimental procedures to determine and control causal factors, and have not focused on how users interpret the processes of being embodied in virtual worlds. Indubitably, the focus on causal connections is a necessary part of the technology development. However, as virtual worlds and avatars continue to promulgate, impacting how users can interact with one another and with societal structures, it is increasingly important to take the interpretive turn to understand the user-avatar connection.

I would argue that we have entered a phase where understanding the reception of these new media forms has become crucial. What follows are two research questions that remain unanswered despite the interest in researching avatars and virtual worlds.

* How does engaging with avatars impact users’ seeking and use of information, including use in task completion, communication with others, etc.?

* Does the extent to which avatars are seen by users as “entertaining” interfere with and/or facilitate these different aspects of information seeking and use?

Understanding these questions will help us begin to understand how avatars can be used outside of design laboratories and online gaming communities.

Research Design Proposal

While there are various means to study people’s reception of media products, from television to movies to newspapers, the approaches to user studies in the communication, library and information science and human computer interaction fields has been long bifurcated into quantitative or qualitative methodologies (Dervin & Foreman-Wernet, 2003; Dervin & Reinhard, 2006; Naumer, Fisher & Dervin, 2008). Rarely have the strengths of the one been used to bolster the weaknesses of the other. However, an interest in constructing methodologies for studying reception that utilize both approaches is receiving increasing support in good part because the rise of new media, where user-centric designs promote and require user control over final outcomes. A need to focus on the user calls for methods that can both understand users’ perspectives in interpretive ways and generate results useful for the purposes of policy and design. It is with this stance towards research design that I support a merged methodology structure to study the reception of avatars, and why I believe the utilization of Dervin’s Sense-Making Methodology would be a useful methodological application in this study.

The application of Dervin’s Sense-Making would allow us to interrogate the user-centric process of engaging with virtual worlds via avatars through an in-depth interviewing protocol that could be standardized to compare people in qualitative and quantitative analytical modes. I am currently using this methodology for my dissertation, interviewing people on their experiences with media they deem to be intended for men or intended for women (Reinhard, 2008). The interviews so far yield results that highlight the complex, sometimes contradictory, ways people engage with the media, as well as providing the basis to argue that while there are differences in how men and women use the media, there are also similarities if one looks at their media use in a conceptually different way, beyond the sociodemographic category of gender (Dervin, 2003).

Sense-Making Methodology supports the individual’s ability to speak about their experiences. The interviewer’s role is to provide the space and tools in the interview to allow people to recollect and make connections amongst the various thoughts, feelings, and theories they use every day to make sense of their lives (Dervin & Foreman-Wernet, 2003; Naumer, Fisher & Dervin, 2008). This interviewing protocol can be used to understand, from their perspective, the sense people make of how the media fits into their lives. With avatars and virtual worlds promoted as a potential communication medium, this same interviewing protocol could help us understand how people see such new media forms as relevant to their lives. Understanding this relevancy would provide information as to the shortcomings and opportunities for designing and implementing these new media.

Sense-Making Methodology could be applied in two different research studies that are similar in their underlying goal of understanding the user’s construction of the utility of avatars for navigating virtual worlds. The first study would be a quasi-experimental design to understand the step-by-step process of engaging with avatars. The second study would be a focus group to engage people to explore the commonalities and differences of their experiences with embodiment in virtual worlds. Both studies are being proposed as possible applications of this interviewing methodology. The studies could be conducted separately or concurrently, as will be explained below. These studies being proposed constitute a moderate approach to answering the research questions above. Should time and resources permit expanding this proposal, more in-depth interviewing of real world experiences with virtual worlds would be included to advance the informative value of this research design.

Phase I: Quasi-Experimental Study

The strength of an experiment is the ability to control the stimulus and conditions of exposure; the weakness is in the belief that the human being can be as easily controlled. Realizing that all humans are sense-making beings attempting to understand the moments of their lives, we should record their interpretive processes of reacting to the experimental stimuli, and from there seek to uncover patterns of differences and commonalities, and the conditions under which these patterns emerge. Focusing on processes would offset the critique that experiments are hypothetical situations with no bearing to real life experiences; the Sense-Making Methodology would encourage understanding how the person makes connections from the experiment to their real lives.

For this study, people with varying degrees of experience with virtual worlds and avatars would be exposed to different environments and tasks – a social interaction task; an information-seeking task; a cooperative play task; a shopping task. The user would be told what the parameters of the task are and how much time they have to complete it. The actual step-by-step completion of the task would be under the user’s control, which is a common parameter for experimental studies of game playing and design. After the task has been completed, the user would be interviewed to reconstruct what they did step-by-step; that is, instead of the experimenter designing what maze the user would run, the user would tell us. For the recollection of each step, the Sense-Making interview would be used to surround that moment of the process and interrogate how the person understood what was happening, what they wanted to happen, and how what was happening connected to their real lives.

It would be preferable to construct the study as a within-subject design; that is, have all participants complete each task and be interviewed following the completion of each task before proceeding to the next one. Comparisons could then be made across one person’s experiences with different tasks, as well as across different people’s experiences with the same tasks. This design would permit a more complex picture to arise than relying on demographic differences between people explaining reception, which would be in keeping with Sense-Making’s belief in the important variability among responses to situations rather than just between people (Dervin, 2003). A within subjects design could be accomplished with fewer people, perhaps as few as 30 and as many as 100. Initial sampling would have to be higher to account for attrition due to research fatigue as the commitment level may be perceived as being intense.

The tasks could be structured into an online world controlled by the project, such that participants would not be required to come to a lab to complete the tasks. Making the tasks available through a central source would increase the potential for continual participation throughout the study. Once it has been recorded that a participant has completed a task, that person would be contacted for the interview. Should it be more efficient, the task and interview could be done in a laboratory setting, provided that the participant does not feel pressured to perform. I freely acknowledge I do not have experience with programming responsibilities, but I believe the flexibility of this design would work with any programming situation that arises.

The resulting database could be analyzed in two ways. Qualitative analysis of themes, using either grounded theory coding or Sense-Making metaphorical coding, would provide exemplars of knowledge construction. Quantitative analysis of the coding categories could provide statistical descriptions, comparisons and even predictions of the conditions under which certain interpretive processes arise. Indeed, a project report utilizing both is preferable, as it would provide a deeper picture of how the participants viewed their virtual embodiment experiences.

An estimated time span for this phase would be contingent upon the construction of the virtual worlds, either from original design or by integration into a pre-existing structure, as well as the determination of how participants will complete the tasks, be interviewed, and the training of coders for reliability checks on coding. Even if the focus groups in Phase II, detailed below, are added to the data collection, total time commitment should not exceed two years for basic data analyses to be completed.

Phase II: Focus Group Study

While traditional focus groups have problems of silencing the expression of minority opinions and moderators overzealously exerting their control, a focus group structured on the Sense-Making Methodology seeks to reduce such power struggles by providing all participants with equal time to voice their thoughts without concern of negative reprisals from other participants (Dervin, 2007). Integrating a focus group into this project would allow for the participants to provide insight into the commonalities and differences between their experiences with the virtual worlds and the experiences of others.

The preferred integration of the focus group would be to recruit participants from the quasi-experiment phase to meet in groups of 6-10 to discuss these experiences. The focus group would use a Sense-Making interviewing protocol to surround their experiences with the tasks by asking users to reflect on the overall experience of each task. Questions on how they saw these tasks would be used as the foundation to allow all users to reflect upon their experiences as relating to each other’s experience. In this way, the results of the focus group serve as a member check – verifying that the users’ connections to each other’s experiences coincide with the patterns seen by the researchers in the analysis of the Phase I interviews.

However, the focus group could also be a separate study. Instead of recruiting users from the pool of people from Phase I, a sample could be drawn from people who are or have been active users of avatars in established virtual worlds, such as Second Life, and massive multiplayer online role-playing games, such as WarCraft and EverQuest. Such a focus group would allow users to make connections between each other’s experiences with these new media. Sampling from a variety of virtual worlds and ensuring equal distribution of different types of experiences into the same focus group would provide a favorable distribution of situations to compare across. This structure would require the same set of questions to serve as the basis for the discussion, with a focus on their real experiences rather than Phase I tasks.

Conclusion

Whether or not avatars and virtual worlds become more than just a passing fad will depend upon how they are received by the general public for activities beyond gaming and social interaction. Can virtual embodiment be useful in other contexts that matter to people in their everyday lives? Could interacting with avatars improve people’s information seeking for many topics, from medical diagnosis and treatment to finding a book in an online bookstore?

To answer these questions we need to do more than conduct studies of the sort that now dominate the literature, utilizing various demographic and technical factors as predictors of the uses and impacts of these new media forms. We need to also understand how users of these new media utilize them in constructing their information worlds, their lived experiences, and their identities. To do that, an interpretive turn in this research is needed.

What I have outlined above are only two possible ways to take this turn using a methodology that has been used widely and successfully in user studies in the communication, library and information science fields. If avatars are to be useful to us in the future, then we need to understand their potential from the perspective of those who use them. Understanding the processes of engaging with these new media forms could provide communication systems seeking to employ them information as to how this can be best accomplished.

References

Arita, D. & Taniguchi, R. (2003). Non-verbal human communication using avatars in a virtual space. Knowledge-based Intelligent Information and Engineering Systems, Part 2, Proceedings, vol. 2774, p. 1077-1084.

Bailenson, J.N., & Blascovich, J. (2004). Avatars. In W. S. Bainbridge (Ed.), Encyclopedia of Human-Computer Interaction: Berkshire Publishing Group.

Bente, G., Ruggenberg, S., Kramer, N.C. & Eschenburg, F. (2008). Avatar-mediated networking: Increasing social presence and interpersonal trust in net-based collaborations. Human Communication Research, 34(2), p. 287-318.

Bideau, B., Kulpa, R., Menardais, S., Fradet, L., Multon, F., Delamarche, P. & Arnaldi, B. (2003). Real handball goalkeeper vs. virtual handball thrower. Presence – Teleoperators and Virtual Environments, 12(4), p. 411-421.

Dechau, J., Finke, M., Gerfelder, N., Ide, R., Kirste T. & Spierling, U. (2001). The Telebuddy®: Collective tele-presence and tele-conversation through physical avatars. Computers and Graphics, UK, 25(4), p. 601-608.

Dervin, B. (2003). “Users as research inventions: How research categories perpetuate inequalities.” In B. Dervin & L. Foreman-Wernet (Eds.) Sense-Making Methodology: Selected writings of Brenda Dervin (pp. 47-60): Hampton Press Inc.

Dervin, B. (2007). Focus groups for participatory research: Design using systematic dialogic principles drawn from Sense-Making Methodology. Presented at IAMCR 2007 Congress, Paris, France.

Dervin, B. & Foreman-Wernet, L. (2003). Sense-Making Methodology: Selected writings of Brenda Dervin (pp. 47-60): Hampton Press Inc.

Dervin, B., and Reinhard, C.D. (2006). Executive summary. In Dervin, B., Reinhard, C.D., Kerr, Z.Y., Song, M. & Shen, F.C. (Eds.) Sense-making the information confluence: The whys and hows of college and university user satisficing of information needs. Phase II: Sense-making online survey and phone interview study. Report on National Leadership Grant LG-02-03-0062-03 to Institute of Museum and Library Services, Washington, D.C. Columbus, Ohio: School of Communication, Ohio State University. Available at: http://imlsproject.comm.ohio-state.edu/imls_reports/imls_PH_II_report_list.html

Frery, A.C., Kelner, J., Moreira, J., & Teichrieb, V. (2002). User satisfaction through empathy and orientation in three-dimensional worlds. Cyberpsychology & Behavior, 5(5), p. 451-459.

Glantz, K. & Rizzo, A.S. (2003). Virtual reality for psychotherapy: Current reality and future possibilities. Psychotherapy, 40(1-2), p. 55-67.

Kang, H.S. & Yang, H.D. (2006). The visual characteristics of avatars in computer-mediated communication: Comparison of Internet Relay Chat and Instant Messenger as of 2003. International Journal of Human-Computer Interaction, 64(12), p. 1173-1183.

Ku, J., Han, K., Lee, H.R., Jang, H.J, Kim, K.U., Park, S.H., Kim, J.J., Kim, C.H., Kim, I.Y., & Kim, S.I. (2007). VR-based conversation training program for patients with schizophrenia: A preliminary clinical trial. Cyberpsychology & Behavior, 10(4), 567-574.

Naumer, C.M., Fisher, K.E., & Dervin, B. (2008). Sense-Making: A methodological perspective. Paper presented at the 26th CHI Conference, Florence, Italy.

Nowak, K.L. & Rauh, C. (2005). The influence of the avatar on online perceptions of anthropomorphism, androgyny, credibility, homophily, and attraction. Journal of Computer-Mediated Communication, 11(1), p. 153-178.

Qiu, L.Y. & Benbasat, I. (2005). Online consumer trust and live help interfaces: The effects of text-to-speech voice and three-dimensional avatars. International Journal of Human-Computer Interaction, 19(1), p. 75-94.

Reinhard, C.D. (2005). “Hypersexualism in digital games as a determinant or deterrent of gameplay: Do men want them and do women want to be them?” Master’s Thesis, Ohio State University.

Reinhard, C.D. (2008). “Cross-gender media engaging as user agency mediations with sociocultural and media structures; or, give her the ray gun, he’ll take the hankie.” Dissertation Proposal, Ohio State University.

Wang, L.C., Baker, J., Wagner, J.A. & Wakefield, K. (2007). Can a retail web site be social? Journal of Marketing, 71(3), p. 143-157.

Yee, N. & Bailenson, J.N. (2007). The Proteus Effect: The effect of transformed self-representation on behavior. Human Communication Research, 33(3), p. 271-290.

[1] Research has shown how avatars in virtual worlds can facilitate online shopping experiences (Qiu & Benbasat, 2005; Wang, Baker, Wagner & Wakefield, 2007), treating mental disorders (Glantz & Rizzo, 2003; Ku, Han, Lee, et al, 2007), online education and simulator training (Arita & Taniguchi, 2003; Bideau, Kulpa, Menardais, et al, 2003), navigational tools (Frery, Kelner, Moreira & Teichrieb, 2002), as well as various goal-oriented communication tasks (Bente, Ruggenberg, Kramer & Eschenburg, 2008; Dechau, Finke, Gerfelder, Ide, Kirste & Spierling, 2001).