In 2021, I presented the ‘Buddy Biking’ study at the ICDVRAT conference in Serpa, Portugal. The study was conducted with Asc. Professor Belinda Lange in Adelaide, Australia in February 2020 (just before Covid-19) impacted the world.
ABSTRACT:
Virtual reality (VR)-based rehabilitation is a growing technological field, which gradually becomes integrated into existing programs. However, technology has to support human behavior and -needs, including social relatedness, to achieve health-related outcomes. Elderly people have high risk of loneliness, and VR has technological affinity for natural social interaction. Previous studies have relied on competitiveness rather than collaborative elements, but research shows that competitiveness can lead to (feelings of) stress and aggressive behavior in some individuals. This article presents a mixed methods study to gather end-user feedback on a social VR scenario that encourages inter-player collaboration on a virtual tandem bike. Outpatients (, 64% males, years) were invited to participate with a co-player (friend or family). Participants biked on average 10.7 (± 3) minutes with a mean speed of 14.8 kmph (± 5.8). The results indicate potential and feasibility for the collaborative social biking application. Participants reported excellent usability-scores (85 ± 5), high intrinsic motivation in all categories: enjoyment (6.5 ± 0.5), effort/importance (6.4 ± 0.3), relatedness (6.3 ± 0.7) and minimal increase in symptoms of nausea, oculomotor and disorientation. Furthermore, participants found the social aspect enjoyable, agreed that collaboration eased tasks and that they lost track of exercise duration. Interpersonal interaction between participants varied, but was mostly positively rated valence, even if the sense of copresence was limited by physical constraints and avatar representation. Most participants expressed that they would use the program again, but future studies should explore how to improve location and appearance of the virtual coactor, as well as implement additional tasks.
Høeg, E.R., Bruun-Pedersen, J.R., Cheary, S. et al. Buddy biking: a user study on social collaboration in a virtual reality exergame for rehabilitation. Virtual Reality 27, 245–262 (2023). https://doi.org/10.1007/s10055-021-00544-z
LINK TO ARTICLE:
https://link.springer.com/article/10.1007/s10055-021-00544-z
All right I’ll I’ll skip the introduction now I don’t have my notes readily available but um we’ll see if we’ll manage all right so um my name is Emil and I’m here to present our work on evaluation of a virtual reality tantm bike ex game for rehabilitation uh and it’s a joint
Effort between fenders University Alo University and F municipality and my lab multisensory experience lab in uh in Denmark and uh as a part of my PhD I’ve been exploring how to develop these um applications for physical therapy to uh to see how in what degree we can can use it to to motivate
Patients um so my first naive understanding of uh of of motivation was this uh Quantified measure that we could just apply and then we would have some profit in the end but uh after a little bit of uh immerse immersement into the real real world and uh a little bit of
Perusal of self-determination Theory uh we found out that about this me psychological view on on motivation that involves the different um psychological needs and that needs these needs need needs to be uh equally fulfilled in order to uh instill this motivation in individuals through that will then in
Turns be able to to get these Mo uh these individuals to experience a sense of of well-being intrinsic motivation and and um selfefficacy and these basic psychological needs that the SD research has talk about is a autonomy the sense of having a um a sense of evoltion in what you’re doing
The task that you’re performing a competence feeling a sense of competence in the task that you’re you’re doing and then in in turn relatedness which is the sort of the inter personal dimension that introduced what is the connection to other people and in in those activities and when you have those need
Supports uh that is what create the the the found foundation for well-being and motivation which is needed in order to succeed in a rehabilitation uh context um but when we look broadly on the field of of U of social VR is very limited there isn’t a lot to to go after
Especially not in in the immersive domain and that is of course a problem when we because there’s a lot of things we don’t know about how to apply this in this new domain for for elderly participants what are the design criterias we have to abide by in order
To to make a a good game that that can motivate activity uh for the ones that have been used there has been a tendency to have a focus on competition uh that makes sense because the uh the non-immersive prototypes that have been used the Nintendo wiii these
These types of game styles are often embedded into the to into the into this into the framework that’s the the golfing the the boxing prototypes except for a few few articles that I found probably more as well that uh try to explore these different game modes so in terms of competition can
That be perceived as a in Need for The Elder elderly patient um what we heard again repeatedly from uh from patients in in Denmark and also in Australia was that they simply weren’t interested in participating in in anything that had involved competition they weren’t interested they were more interested in
Having that social um social experience just socializing in a non-competitive way uh and that’s sort of backed up by a study from 2012 uh illustrating that competitiveness tends to increase during the lifespan and then when you reach the age of 50 it tends to decrease again and
That has ramifications for the type of game that we design uh another problem with competition is that individual differences are Amplified so if we want to uh succeed with a game design we have to consider uh to what degree do we actually um cheat the system I mean
Because uh the the the patients will come from from different functional backgrounds they will have uh different levels of of capacity so we have to in a non uh in know non nonovert way we have to to to to to design this game so it doesn’t manipulate or we can try some
Other design methods another problem with competition is that it has a tendency to lead to stress and aggressiveness in two articles here um in in the first study they found that um the um competition had had a tendency to to increase in-game violence and U and foul language among participants
Um as opposed to collaborative or Co Cooperative game styles and in other style uh other paper the effect of a Cooperative game sort of flooded outside the study because they had a little uh post study experiment where member the other member Dr a pencil U and then you
Had you had a tenden to see that uh if the co access they PID together they were more inclined to to actually help each other subsequently as well that means that we have both in-game uh Behavior changes we also have Behavior changes after the game has been played
Which I think is pretty important and with those things in mind we set out to develop this body biking system um and what we wanted to know specifically was how does the user experience a collaborative environment get the feedback on it um is the joint effort perceived as hampering or eating
Performance does the user experience any side effects uh and can the appearance of a virtual Hill increase Cadence since we could not manipulate with the um with the um with the bike the exercise bike but we could manipulate with the perceived speed so what we did when we
Had a hill we turned down the the the speed to see if they would actually compensate and increase their Cadence and um it was developed in unity and we made a custom built a sensor but uh as with all prototypes the real problem started with the networking part um I really
Wish I could play this probably I could um like this um because that’s sort of a Pandora box when you start to interfacing two computers it becomes really a hassle as a one person online set what what you’re trying to uh to achieve is one of the most complex things in networking so
Great um with that in mind we set out to solve it with the interpolation and it got a little a little bit better uh just predicting based on uh on on on the current Cadence where would the next position be but we still weren’t quite satisfied um because
There was still a little bit of jittery um but what we noticed was that this uh this jittery or this lack was only based around the uh the host so what we did was just put the host in the back and the one in front could not see any
Uh any lag so that was a cheap solution that works on a tandem bike not on any other conditions um so just a quick um demonstration of what the final um scene look like this is from the first the person seated in the front and then you had both auditory and
Visual feedback in the in the slow down at the streets and the and the rich environment the Eagles all right so methods we did a We performed a mixed methods user study in in a outpatient W in in Australia we screened 37 uh potential patients um and finally eight patients agreed to to
Participate uh one dropped out three members joined their respective Partners if they didn’t have a partner they were set up with a member of the research team we ended up with 11 participants seven males and uh 60 years plus minus so maybe not um exactly elderly adults
Other some some of you might take offense um so before uh the start uh we had the consent form the interview demographic questionnaire we did a baseline VR sickness questionnaire to measure the uh the change scores uh then during the test we measured the interpersonal dialogue interaction that go went on between them
In the in the system itself Unity locked the Cadence and the virtual slope of the hills and the post test we measured the ssq again then they we used the virtual embodiment questionnaire to tap into the experience of the sense of embodiment the intrinsic motivation inventory usability and semi structed interview so
The test was closely managed before Co hit so that was quite lucky when I managed to to wrap it up um so in terms of motivation both of them were high but what we noticed was that there was a difference between participants located in the front versus in the back where
Was higher in the front where they can actually see the participant in front of them which is quite interesting um in terms of embodiment there was a high sense embodiment and agency low sense of altered body schema which was quite interesting because the only like really virtual Avatar was a floating bicycle
Helmet um so the social interaction was um Vivid talking it was most balanced it was um positive veillance they talked about um good things things they explored or noticed in in the environment and the mood was overall good except for the one participant that dropped out tended to be more um
Negative so in terms of VR sickness it um it was okay along the uh the subscales and the total score was also fine um usually has to be under 20 but that’s sort of for military um simulators and not specifically for immersive uh virtual reties so I think we can get
Away with a little bit more uh but one thing that was uh proved could be a problem with this excessive sweating because one of the item of the ssq measures the uh uh sweating at an early onset of noria um so that’s a problem that can confound the measure so what we
Did was to to sort of just retain that score at the Baseline level uh to to to to detect how that would look like and you can see the modified post adjust score over here okay so uh in terms of the OS regression we ran with the Cadence vers versus
Slope there was a um the um The Intercept Y axis is uh just the slope when there’s no slope on the on the hill just going straight ahead there was a significant effect on the virtual slope uh 0.43 uh RPM increase per one degree increase and U in terms of predictors we
Could say that the H and sex explain some of the V variations in the data um participants statements uh one uh participant remarked that you’re more or less you know other people are around but you don’t interact with them whereas this here the body biking system you’re
Working together as as a team sort of bringing in the context of the uh of the existing therapy to explain how this could be used in a in in a novel way um in terms of the competitiveness um another participant remarked that I don’t have a competitive bone in my body
So it wouldn’t interest me when asked would you rather engage in a competitive play style um so one of the other remarked that so I did 6 kilm and I was surprised because I can only do four so that sort of uh speaks about the the sense of competence that you can go
Beyond what you thought you was were capable of which I think is very important that you can actually show this to the patients and that they’re capable of much more than they think they are uh to sum up uh there was a high motivation uh low VR signal score but
Sweat was a confounder uh front seat placement led to more side effects uh there was a pos positive interpersonal interaction going on between the participants when we asked them about their sense of how much time they’ve spent at it they had a tendency to to underestimate the length of the St
Training so this a time compression effect that they actually thought they’ve been biking for for much longer than they actually had wice Vera so there was a need for for better Avatar represent ation that they remarked that they sort of felt disconnected from the bike likely because they only had this
Uh floating bicycle hem to to to facilitate their their virtual presence um so if you want to know more we just published this journal article in in the special issue so um feel free to to PR that of course Shameless self-promotion as well uh thank you for your time how was that perfect
And we’ll take questions in the end thank you very much am