Maximilian Landeck is a student in the HCI Group at the University of Würzburg, and he presented a poster at IEEE VR conference investigating the impacts of avatar embodiment on virtual fitness training. He was initially interested in seeing whether or not the fidelity of the avatar from photorealistic to stylized and from human-like to not human-like had an impact on their performance in a virtual fitness scenario.
He used a Kinect to capture the movements of the participant’s bodies, and the projected that onto a screen so that it was mirroring their movements. He found that the human-like avatars invoked the virtual body ownership illusion more than the non-humanoid avatars such as having spheres for hands. He didn’t find any significant difference in performance based upon the fidelity of the avatars, but he did find that the non-realistic avatars that looked stronger and more fit avatars had participants report that they felt like they had more power. This is a surprising result that he intends on exploring more in future research.
Become a Patron! Support The Voices of VR Podcast Patreon
Theme music: “Fatality” by Tigoolio
Subscribe to the Voices of VR podcast.
Rough Transcript
[00:00:05.412] Kent Bye: The Voices of VR Podcast.
[00:00:12.039] Maximilian Landeck: My name is Maximilian Lendeck, I'm from the University of Würzburg, from the HCI group, and we were asking ourselves how to use a common metaphor game like a Kinect sports game, where people can do some sports-motivated movements at home, maybe against physical inactivity, or only to a contrast to only sitting on the table on the PC. and we redid this in an experiment with a virtual avatar design and wanted to know if the realism of the avatar or of your self-representation in the virtual mirror is important for doing maybe more motivation to sports and therefore we tested a set of avatars and the different illusions of body ownership with the different avatars and the different performances. We used a simple set of a workout. The workout lasts about 30 seconds where the participant has to touch as many blocks as they could. There were only three blocks presented in the virtual mirror and they were randomly activated to touch and so on. So out of the touch blocks we measured the performance and out of a questionnaire we called that illusion of body ownership. Depends on the few papers in this topic. Therefore we could say if the illusion of owning the body you see differs between our set. We use a set of realistic avatars such as a male human avatar and a female human avatar. And in addition to that, a set of non-realistic abstract looking avatars, like a warrior, a big strong warrior, or a robot, or a block man. Block man is only the general shape of a human. And for that kind of a control issue, we used no avatar at all, only visualizing the spheres from the hand positions. Therefore we could say out of the results that there was no significant difference between the avatars, between the realistic and the non-realistic avatars. So just maybe we can take advantage of the issue that we present the user a stronger looking avatar and therefore motivate him towards sports. And yes, out of the qualitative analysis or out of the open questions, we suggest also that a feeling of greater power existed with bigger, stronger looking avatars. There are some user comment examples like Yes, I felt stronger because the avatar looks more massive, more fit or because the avatar transfers a stronger feeling to me. So of course we have to redefine these results in focusing more on the feeling of power issue and of course maybe redesign it in a full immersive setup and also with a higher quality motion tracking system like the Kinect we used in this experiment.
[00:03:11.265] Kent Bye: Yeah, so it sounds like by embodying a character that perhaps is a vision of our future selves of maybe stronger, faster, whatever, you know, kind of like at least have a physical appearance that's different than our current avatar, that you could start to invoke some of those feelings within that. And so how would you plan on trying to measure that sense of power in the future as you move forward?
[00:03:34.682] Maximilian Landeck: That's an interesting question. Maybe we could measure the power of a longer time. Repeated studies that maybe the participants get more effective in doing the same workout and lower down their heart rate as the maybe a group with doing it with stronger looking avatars. And also out of questionnaires we're asking about motivation towards sports. Or maybe also the question how often they do themselves virtual training. If it is like a volunteered condition, like you decide your own how often do you go in sports or if you never go out doing some sports, that maybe is an important hint to, as you said it, your identity between how you see yourself and how you see yourself inside the virtual mirror. It's hard to say, but I think you can move how you want to move and not move like you think you're a normal guy. But we will redefine these questions on the stronger looking avatars and I think it's an interesting research topic. And we can extend this topic, maybe not only about motivation towards a sport addicted, maybe also about if people have physical disorder and they're seeing themselves, yes, they're not motivated because they had an accident and they can't move as they would move. And so if they see a stronger looking avatar, they are more motivated on moving themselves and on going forward. So medical and health is just physiotherapy of course and so on, yeah.
[00:05:21.780] Kent Bye: And so, did you find that there was no difference between the avatars that had a body and even between just the spheres of the hands moving?
[00:05:30.368] Maximilian Landeck: Yes, of course, that was important. What we suggested before already, there was a significant difference in illusion of body ownership and performance between the set of avatars and no avatar at all, yeah.
[00:05:43.059] Kent Bye: Okay, so what was the difference?
[00:05:45.435] Maximilian Landeck: The difference in the measured data, we had significantly less feeling of the illusion of body ownership when you use no avatar at all, only the spheres. And you had also, of course, less performance if you use only the spheres, yeah.
[00:06:00.319] Kent Bye: Okay, so it sounds like invoking the illusion of a virtual body ownership is important to performance, then, it sounds like. Yeah, so maybe you could talk about why you think that is or maybe sort of what that implies.
[00:06:13.150] Maximilian Landeck: I think that implies that you have to look at the body or at a human-shaped figure, but it also implies that it does not have to be a realistic human figure of yourself. Then you can go on saying how we want to look at the human shape and how can it differ, how we can maybe improve that. And even if it's about some technical issues or cost-effective issues, we don't have to use very high quality realistic human models.
[00:06:45.223] Kent Bye: And so what kind of experiences do you want to have within virtual reality?
[00:06:50.228] Maximilian Landeck: I have experience from the HDI group, full immersive tracking systems like the work of Johanna. I've also known some work of other guys with the Oculus Rift and Hydra setup. I myself worked a lot with the Kinect setup and will further go on about that. Yes, the work of Mel Slater from the University of Barcelona was a big starting point for us and a big basic experience. Yes, we really rely on his works and try to go further, or not even go further, to extend the topic in a wider range of other questions.
[00:07:28.420] Kent Bye: Great. And finally, what do you want to see come out of all this? I know there's a lot of implications about the sense of self and bigger philosophical issues, but I'm just curious from your own perspective as what you see the applications or the ultimate potential for the use of virtual reality.
[00:07:47.222] Maximilian Landeck: Yes, I see there's a big potential about being more effective with interacting with digital systems, of course, with giving the users a more natural feeling about interacting with digital systems. And I think also an important issue that the factor you have often learned to use a system will reduce a lot through virtual environment research, yes.
[00:08:14.597] Kent Bye: Okay, great. Thank you.
[00:08:16.720] Maximilian Landeck: Thank you very much. Goodbye.
[00:08:18.804] Kent Bye: And thank you for listening. If you'd like to support the Voices of VR podcast, then please consider becoming a patron at patreon.com slash voices of VR.
