Last week, I attended the CHI 2009 conference here in Boston. I have been involved with the computer-human interaction (CHI) community since the early 1980’s, and I host the monthly meetings of the local BostonCHI chapter, the first ever local CHI chapter. In addition to attending sessions, I was helping to staff the BostonCHI booth at the conference. Here I am at the booth with BostonCHI co-founder Kate Ehrlich and current chair, Doug Gibson.
There was not too much at the conference related to virtual worlds, but I did come across a few relevant tidbits to share with you.
Judy Olson on "Social Ergonomics"
One of my favorite social science researchers, Judy Olson from the University of California at Irvine, gave the keynote talk. Although the talk didn’t mention virtual worlds, the theme of the talk was highly relevant. Judy talked about a concept she calls "social ergonomics," which she describes as the "design of workplaces and systems that fit the natural social capabilities and inclinations of users." She discussed her large body of research on the impact of physical proximity on work outcomes. People who are "radically co-located" – working together in the same space – are almost twice as productive as those who are distant. This is due to awareness of others actions, gestures, and gaze, as well as the ability to have impromptu conversations. In real life, we judge how to behave towards others based on the distance they stand from us when talking coupled with other subtle social cues, such as eye gaze.
Her analysis of some of the newer video conference systems was fascinating. In these systems, there’s often a large monitor and people on both sides of the conversation are facing the monitor, staring directly at one another. In real life, this is a confrontational stance. Most people in natural conversations stand or sit at an angle to one another that can be as much as 90 degrees.
Another relevant research area she talked about had to do with eye gaze as a social cue. When someone is speaking, they tend to move their gaze away from the listeners until they are ready to end their turn, at which point they look directly at the next person and pause. The listeners, however, typically continue to look at the speaker. Also, humans are apparently much more sensitive to right-left eye motion than to up-down motion. We perceive that someone is looking away from us when they move their eyes only slightly right or left. They have to make much bigger movements up or down before we perceive them to be looking away. So if someone is looking at our forehead or chin, we will perceive that they are still looking at us.
This research has many implications for virtual worlds, especially if we want to achieve the next level of immersion and start to see some of the benefits of "radical co-location" in world. I think the most significant takeaway is that we must find a way to map natural, unconscious head gestures and eye gaze onto our avatars to replicate the social cues from the real world.
Nick Ducheneaut from PARC presented an interesting study about avatar personalization called "Body and mind: a study of avatar personalization in three virtual worlds". One of his main conclusions was quite surprising. It turns out that hair style and hair color are the most significant avatar features. They are the features that people customize most often and that users seem to care about the most. Nick had several theories for why this might be the case. He argued that in real life, hair is our most malleable physical characteristic so people are used to "configuring" and changing their real-life hair. Hair is also highly visible to users of the virtual world because people spend a lot of time in third-person mode staring at the back of their avatar’s head. Finally, hair can be seen from far away and is the feature that most helps to identify other avatars at a distance.
Another finding from the study is that most people create avatars that have many characteristics similar to themselves. The most common type of avatar is one that can be considered an "idealized self" – similar characteristics to the real person, but improved in terms of weight, height, athleticism, and so forth. Those people who were most successful in creating an idealized self were also the people who were most attached to their avatars.
Challenges for Virtual World Users
The paper "Acquiring a Professional ‘Second Life:’ Problems and Prospects for the Use of Virtual Worlds in Business" (PDF) was presented by a CMU student who was an intern at IBM. It covers 5 challenges for virtual world users in a business context. One of the big challenges is motivating business users to try virtual world technology in the first place. Some people felt that no technology would ever replace face-to-face interaction, others thought the virtual world was too much like a game, while others worried that their management would not approve.
As we all know, another challenge is getting the technology to work and then learning how to use it. The paper discusses the process of "becoming a competent virtual person," which I thought was a great turn of phrase.
The other major challenges involved learning to control an avatar, interacting with others, and finding compelling activities that take advantage of the virtual environment.
Some folks at Kodak Research did a study called "Capturing and sharing memories in a virtual world." Not too surprisingly, they found that people in the virtual world, particularly heavy users, liked to capture and share "photographic" memories – in this case screen shots – in much the same way as people in the real world. And like people in the real world, virtual world users have inadequate tools for organizing snapshots for easy viewing and retrieval.
Improvisation for Brainstorming
I’m convinced that if we can do it right, virtual worlds will prove to be an excellent medium for remote brainstorming sessions. That’s why I was interested in the talk "Using improvisation to enhance the effectiveness of brainstorming" by Elizabeth Gerber from Northwestern University. She talks about using theatrical techniques to get the creative juices flowing. Most of the techniques she suggested required face-to-face interaction, but there was one that I thought might work well in a virtual world brainstorming session. She had everyone in the audience take an every day object out of their pocket or bag. Then, working with a partner, you pass the object back and forth. The person holding the object has to think of a possible alternate use for it. For example, if the object is a pen, you might be able to use it as a "back scratcher," "hole puncher," "game spinner," or "drum stick." The idea is to come up with a
s many alternate uses as possible in a short amount of time. The exercise is intended to help people generate a lot of ideas quickly.
I was imagining doing this fun exercise in the virtual world by having a person drop an object or a photo of an object into the world. Each team could record their ideas on a shared whiteboard or just take turns speaking the ideas aloud.
User Experience in Open Source Projects
I attended a special interest group meeting on the topic of integrating user experience into open source software. It was interesting to hear about various efforts in this space and find out about some of the resources available. First, check out the OpenUsability web site. This site attempts to match people working on open source projects with usability professionals. They also sponsor the Season of Usability mentoring program for students.
I was particularly interested in a group called Aspiration which tries to help non-profit organizations improve their software. One way they do this is by running "usability sprints" – 3-day workshops where each participating non-profit identifies their most significant usability problem and a team of experts works with the developers to solve the problem, often writing code on the spot. I was thinking that a number of the non-profits working with Wonderland might like to apply for this program. They said they’re looking for new projects to support for a 2009 sprint.
Another potentially interesting group that might be relevant to Wonderland is Teaching Open Source. This group tries to match up people in open source projects who are willing to be mentors with students interested in working on a open source project.
3D User Interfaces
I attended part of a full-day course on 3D user interface design. The course notes will soon be available in the Resources section of the instructors’ 3D UI web site. This course covered a lot of materials, but much of it was focused on virtual reality, caves, Wii-motes, and the use of head-mounted displays.
And finally, I leave you with a pointer to a fun video that a former colleague of mine recommended for anyone interesting in world building: