Madis Vasser is a psychology student at University of Tartu Virtual Neuroscience Lab, and he collaborated with the computer science department to create a VR toolbox for doing experimental psychology research. He was showing off a demo of a change blindness experiment that he created within Unity at the IEEE VR conference.
Change blindness is the effect where it is difficult to notice changes in your virtual environment. It’s not easy to make objects disappear in our physical reality, and so we’ve evolved to not really notice subtle changes in our physical environment. But in VR, it’s very easy to make small changes to your environment that are really difficult to perceive.
The VR experience called Sightline really exploits this psychological phenomena to great effect to be able to cut between scenes in VR. So even though I’m intellectually familiar with the change blindness concept, when I tried out my own perceptual acuity in a controlled experiment I still found it to be an effect that was way more difficult to perceive than I would have expected. Here’s more results from Madis’ change blindness research.
Maris talks about creating a generalized VR toolkit for non-programming psychology students to do other research projects, as well as some of the ethical implications of replicating controversial psychological experiments like the Milgram obedience experiment. While no ethics review board would allow psychological students to replicate this study, there’s nothing preventing people from recreating the experience in a VR experience. There are many open ethical questions around this that will be interesting to see how they play out both in more controlled research environments, but all the free market of consumer VR experiences.
Become a Patron! Support The Voices of VR Podcast Patreon
Theme music: “Fatality” by Tigoolio
Subscribe to the Voices of VR podcast.
Rough Transcript
[00:00:05.452] Kent Bye: The Voices of VR Podcast.
[00:00:12.039] Madis Vasser: My name is Marit Vaser, I'm from Estonia, Tartu University, and at IEEE VR we are showing a little toolbox that we are developing for Unity to make the life of people like me easier who are maybe not computer science students but still want to do virtual reality experiments, let's say in the field of psychology. So we're developing a little add-on that will greatly simplify the making of different well-known experiments in VR that are currently more difficult to do. So let's just say here we're demoing, for instance, the effect of change blindness, the phenomena of you not noticing really big changes in your environment. So it's been previously done in two dimensions, but we're trying to bring it to 3D and really see if it's easier, more difficult, and what does it say about our brains.
[00:01:10.342] Kent Bye: Right, so maybe you could talk a bit about your background in terms of what you're studying in terms of psychology and then what's involved with this VR toolkit for experimental psychology.
[00:01:20.466] Madis Vasser: So my background is quite colorful. I started out researching sleep and dreaming and now after five years I'm transitioning into intention and VR and stuff like that. and with this toolbox the psychology department collaborates with the computer science students and so together we from the psychology department place the order of what is needed and CES guys then figure it out how to make it and it's basically a win-win situation for both because we all get our thesis done like that.
[00:01:55.013] Kent Bye: Yeah, in terms of the change blindness, you have a little experiment here where you have one minute to look around and try to identify the object that is missing. And so, in the process of that, have you actually put this out into the field and started to research this more formally?
[00:02:11.922] Madis Vasser: This is the plan. Currently we're really just working out the kinks and documenting and making our code more readable. So it's still a really fresh demo. But from the conference floor we've seen that it works really well. People are amazed at how bad their attention sometimes is. So we have three rooms and three different things changing. in general, in overall. So some people get 0 out of 3, some get 3 out of 3, and it's all normal. So because it's a devilishly difficult task to perform, you should know.
[00:02:46.412] Kent Bye: Yeah, I actually had a chance to do it, and I've seen the experience of Sightline and experienced it, but when it's in a controlled research environment, and I'm looking around, it took me at least 30 seconds to find the first one, and I think even the second one, and the last one I couldn't find at all, so it was kind of surprising. I mean, the biggest one, right? Well, yeah, and I don't know if it was like because of the of the judder, you know I didn't want to move my head too quickly because I know that makes me motion sick and so being Familiar with VR enough. I know how to kind of move around but even with that and being familiar with VR experiences I think the effect of change blindness under controlled conditions like this is still a pretty mind-blowing effect that your mind could be tricked or fooled in that way and
[00:03:27.668] Madis Vasser: So, one of the questions I had was like, since this effect has been extensively studied in 2D previously, and I wanted to know if you put it in 3D in a controlled environment, like you said, will it be, for one, easier? Because, you know, you have this special information that you didn't have before, like, you know, the table is on the right and the chair is on the left and the room is this big and composed of these things, because humans are really good at special information. Or it's even more difficult than in the classical studies because you're immersed in this world and usually when you're in a room things just don't disappear and appear unless they're stolen by someone and then brought back again. We've seen both sides currently so we have to figure out what is causing it. In my demo here I was varying the distances between the objects. are things closer to you, easier to spot or things that are like five meters away and the results are in processing.
[00:04:27.557] Kent Bye: Yeah, I noticed just one of the comments that Mel Slater had said, which I found really interesting, was that the further away that your experimental subjects are from the computer science department, the better your results. And so I actually experienced that in your experiment, where I know what a 3D object kind of looks like and what the background texture is. And so I'm sort of more paying attention to the foreground objects of stuff that's like a 3D model rather than stuff in the background.
[00:04:51.920] Madis Vasser: The pro tip is looking at the lighting because all the other objects are static and are like baked in but the one that is changing has to be dynamic and different lighting applies to it so if you're a real graphics geek you can probably tell it. But about the graphical quality and rendering and such things For me, personally, it is really important that experience looks as good as possible, especially when you're looking at attention, because there are so many really subtle and different and tiny cues we use in our everyday life to assess distances and relationships between objects. So if you put a virtual world full of perfectly illuminated squares, you might be studying something completely different than what you thought. And one nice thing that came out of a discussion a few days back was that if you're doing depth perception studies, the most horrible shape you can use is a sphere, because it's so ambiguous.
[00:05:52.023] Kent Bye: Are there other psychological experiments that you think would be very well suited to exploring further in VR environments?
[00:05:58.949] Madis Vasser: I already got some requests to make the classical Milgram study in VR. That's the one with the obedience written all over it. When you're in a laboratory setting and there's an experimenter who tells you to shock other people. I'm not sure we're going to make an academic study out of it, but we're definitely going to replicate a fun experience for the whole family. In terms of actual research with the toolbox, we're also looking into memory, so how our spatial and temporal systems work. And this is something you also can't easily do in real life, like let's say you have to walk through a house and in the last room in the house there's a giant screen and on the screen you're being shown scenes from your own eyes that you have passed previously or some new scenes or images and you're being asked like do you remember being there or not. So that's another sub-team of our team doing that and hopefully in the final toolbox these all will be incorporated.
[00:06:59.790] Kent Bye: Yeah, and in terms of the Milgram study, one thing that comes up is that, you know, in that case they were shocking other characters that, you know, there was just an audio recording, but they didn't necessarily know, they thought that they were being told to shock other human beings. And so it brings up the question of some of the ethical and moral implications of putting people in these situations that could end up being traumatizing to them. And so, as a psychological researcher, what are some of the protocols that you have to follow in terms of, like, disclosing to somebody some of the things that they may be experiencing in your experiment and then how much of that should be translated over into VR?
[00:07:37.291] Madis Vasser: This is a difficult question because it depends on the experiment at hand but you definitely should warn people beforehand what is the experiment going to be or the experience going to be. But for some studies, you're right, it doesn't work that way because then it abolishes the effect. So I know doing a classical mirror-gram study in today's academic world is pretty much impossible because no ethics committee will allow it, but we have to figure it out. But if it's a free experiment on the internet, that you can do on your own free will and you're being informed extensively beforehand that this is going to, this might affect you, then if you press yes like five times then it's pretty obvious that you're okay with doing the experiment. So this is one approach. So because we don't see people presenting you with five pages of documentation every time you try an Oculus demo because there's this health and safety warning. It would be good if you just sign on five papers that you are doing it on your own will and you take all responsibility and if you fall over it's your fault and so on and so on. I'd like people to take more responsibility of their own actions as well, so we'll see. But yeah, we could say that it's really unexplored, at least. And people are going to do experiments at home with Oculus Rift and other HTMUs, there's no question about it. People are already doing some weird experiments. brain stimulation stuff at home with electrodes and currents running through their brain. So I think it's just, we should at least release some guidelines that they're doing it anyway, let them do them safe at least. So this should be true with Oculus as well. We know there are experiences out there, be safe, follow these guidelines, don't overdo it. If you do overdo it, contact these people.
[00:09:29.417] Kent Bye: Yeah, I think it's an interesting point to make that no ethics committee would allow the Milgram experiment to happen again in real life, but yet, would it then be okay to do in virtual reality?
[00:09:39.181] Madis Vasser: I think no ethics committee, at least in Tarzana University or close by, is actually ready even to discuss what is virtual reality, because with this commercial explosion, it is coming in so fast, the academic world is, I think it's slightly lagging behind.
[00:09:56.080] Kent Bye: Yeah, and it is a little bit of the Wild West in terms of what people are able to do. So there'd be nothing to stop somebody to recreate the Milgram experiment in a VR experience.
[00:10:07.454] Madis Vasser: So I guess that's then the moral imperative for us to make a good one and an informed one. So my first and so far only release to the Oculus Store has been this demo about being a lab rat. And this is supposed to be an educational demo of how might it feel to be a laboratory rat doing the classical experiment. And in the readme there's like a few pages about how you should approach this game before we even start. There's much work to be done.
[00:10:39.597] Kent Bye: Great. And finally, what do you see as kind of the ultimate potential for using virtual reality in psychological research?
[00:10:46.262] Madis Vasser: Well, I guess during this conference, it's emerged that body perception is really one way this could be used extensively, or it's a big, big part of everything, of this cyber sickness and immersion. So I think past psychological research can be really used to make the connection between your real body and virtual body and make it robust. Let's say it comes to standard that next year every game starts with you calibrating your body like according to some virtual body and brain makes this link between these two and you're off to a good game. And when you come back, this is what I've been going around and preaching that what Mel Slater said that it is possible to transform your body to something else. the virtual body but I see it applicable other way around as well so you come out of the virtual reality like you sat there like five hours let's say and you take the headset off and boom you're back in your real body it corresponds to your every movement and it's totally under your control so if the transition from real to virtual takes like a few minutes max this the other way around I cannot see it taking any longer especially since you have your own body for like 20 years or 30 or So these 20 years of your own experience in your body does not anchor you to your body, then how could this a day in a virtual reality anchor you to a virtual body? Since the long-term effects have not yet been established, I guess the community is going to have to go trial and error on that one. So the first people, like the early adapters, are really the guinea pigs now. So I plan with my own research as well to facilitate this massive experimentation. So everyone has a headset at home. This means they can basically teleport themselves to the lab. So you just set some guidelines that please be in a quiet room and not too bright room and then you put the headset on and like people over the world can have the same experimental experience and later the data can come to the lab and Well, there is a problem with the sample, right, because everybody owning a TK-2 now is a VR geek in a way, so it might not translate to the general public. But in the future where everyone has a headset, then you can get like a worldwide sample for your little experiment and it's gonna be awesome.
[00:13:07.456] Kent Bye: Okay, great. Well, thank you. Thank you. 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.
