Researchers Say Head-mounted Haptics Can Combat Smooth Locomotion Discomfort in VR


Researchers from the National Taiwan University, National Chengchi University, and Texas A&M University say that haptic feedback delivered to the head right from a VR headset can significantly reduce discomfort related to smooth locomotion in VR.

Moving players artificially through large virtual environments isn’t a trivial task. While there’s many different ways to move around in VR, smooth locomotion—the kind you’d find in most first-person non-VR games—is a popular method because it maps easily to existing game design paradigms. Unfortunately this method of virtual locomotion isn’t comfortable for everyone.

Image courtesy Yi-Hao Peng

In a paper published as part of CHI 2020—a conference focused on human-computer interaction—researchers describe their WalkingVibe system which uses simple head-mounted haptics to provide sensations that synchronize with the movement of the user in virtual reality. After conducting an initial study with 240 participants, the researchers say the system can significantly reduce discomfort associated with smooth locomotion and even improve immersion.

Using the Vive Pro Eye headset as the foundation for their work, the researchers tested two different types of head-mounted haptics: vibrating motors and actuated tappers (literally little arms that can gently tap the side of the user’s head). The haptics were synchronized to virtual footsteps to offer a stand-in stimulus for the sensations associated with real walking.

Users were walked through three different virtual environments | Image courtesy Yi-Hao Peng

The researchers built a test VR application in Unity, which was linked to the haptics, in which users were walked through three different VR environments and asked to rate their level of comfort.

To check their work, the researchers also ran the same tests with both visual and auditory stimulation (artificial head bobbing and footstep sounds) without any haptics to isolate any effect. They also ran tests with randomized haptic stimulation to tell if the synchronization of the stimulation mattered to the outcome.

The results from the 240 participant study show a significant improvement in comfort and an improvement in realism from the haptics compared to the other methods tested.

[…] all 2-sided tactile designs significantly reduced VR sickness compared to the conditions with no haptic feedback. In addition, WalkingVibe with the 2-sided, footstep-synchronized vibrotactile cues significantly reduced discomfort compared to all other conditions and significantly improved realism compared to all tactile conditions, including tapping-based feedback.

Image courtesy Yi-Hao Peng

The researchers also discussed the limitations of their experiments. Notably, users in this study were physically seated; the same tests were not conducted with physically standing users. Many VR games with artificial locomotion accommodate seated and standing players. Furthermore the researchers said that the artificial movement during the tests was not under the control of the test subjects; they were essentially taken along a guided path without active control over their movement.

Facebook Publishes New Research on Hyper-realistic Virtual Avatars

The full paper is titled WalkingVibe: Reducing Virtual Reality Sickness and Improving Realism while Walking in VR using Unobtrusive Head-mounted Vibrotactile Feedback, and credits researchers Yi-Hao Peng, Carolyn Yu, Shi-Hong Liu, Chung-Wei Wang, Paul Taele, Neng-Hao Yu, and Mike Y. Chen.

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. See here for more information.

Ben is the world's most senior professional analyst solely dedicated to the XR industry, having founded Road to VR in 2011—a year before the Oculus Kickstarter sparked a resurgence that led to the modern XR landscape. He has authored more than 3,000 articles chronicling the evolution of the XR industry over more than a decade. With that unique perspective, Ben has been consistently recognized as one of the most influential voices in XR, giving keynotes and joining panel and podcast discussions at key industry events. He is a self-described "journalist and analyst, not evangelist."
  • “Notably, users in this study were physically seated; the same tests were not conducted with physically standing users”.

    “Furthermore the researchers said that the artificial movement during the tests was not under the control of the test subjects; they were essentially taken along a guided path without active control over their movement.”


    If the researchers are reading this, please rerun your study using standing players and user determined movements?

    These are the real world conditions we play in, and the haptic feedback should prove useful

    personally I’ve always found VR walking simulators more comfortable and realistic when walking on the spot to generate proprioceptive feedback, it doesn’t need to match as a mixed signal is better than no signal. And Beat Saber is a proprioceptive feast with all the dancing about…

    • Gerald Terveen

      admitedly this is a wild guess! I don’t think that testing seated was a choice per se, but more likely a necessity. messing with your vestibular system while your body depends on it to balance itself out seems like a bad idea.

      but that doesn’t mean at all this is pointless – quite the oposite actually. seated gameplay will stay dominant as people want to play seated for many genres and longer sessions. being able to move seated with reduced levels of discomfort seems like a great move forward for VR!

      • Ad

        No? Seated is not more popular in most genres.

        • August R Graham

          Outside of racing, standing is by and large more popular. It’s why controller tracked room-scale headsets have exploded in popularity. Steam user statistics back this up.

        • Gerald Terveen

          most genres have not even made it to VR … I am not limiting myself to VR, but to games here. Most gamers still play seated (on a screen) and I dare to say that if VR wants to turn into a real force then it will have to compete with the seated games.

          I for one want a civilization style game … and I don’t think I want to play it standing.

          • Ad

            I don’t necessarily disagree with that, as long as they fully take advantage of sitting. I play Tabletop Simulator sitting and I think if I didn’t have an index it would be less comfortable because headset comfort is really important obviously but also because looking down the way you do in seated play can be an issue where you have more of a range of movement when standing and crouching.
            Except what locomotion is there in a civilization game? That kind of resets your argument.

          • Gerald Terveen

            ohh – actually there can be a ton of locomotion. in real civilization the camera zooms in and out of the landscape and flies over it. in VR that is you doing the same and it can be quite uncomfortable.

          • Ad

            Uhh, not really. Games with a similar concept don’t have a lot of locomotion and I think the ideal situation for Civ in VR would be the map folding up into a globe, then you turn the globe and straighten it out again. A little like google earth. It’s not uncomfortable enough for a haptic solution to be introduced, not to mention that this haptic method is for simulated movement through space like walking.

          • Gerald Terveen

            you are looking to circumvent a solution … that is just a bad way to think about things to begin with mate!

          • Ad

            And yours just looking for a problem.

          • wheeler

            Totally agree here. It’s just not realistic to expect a good percentage of gamers to stand up every day to play games. Anecdotally, when I’ve played VR games with my “normal” gaming buddies, this is something they would always complain about. And after several years of standing VR, I’ve simply gotten tired of this and play VR games seated unless the actual game is necessarily played standing (Beat Saber). I’ve found I can play much longer this way and I’m less conscious of my physical body. I think this is one (of many) issues with VR that have lead to infrequent usage. Ultimately VR has to be designed for gamers and not the other way around.

            With a swiveling chair seated play is pretty viable assuming the game supports virtual crouch, but the design of the chair has some unique aspects to it for VR (e.g. narrow back, narrow seat, no arms). The main issue right now is that the cable wraps around the base so wireless will really help here.

          • Gerald Terveen

            and that is why I think these innovation could make a difference. I don’t have a nice swiveling chair where I game, but it makes one hell of a difference when I use one!

            with comfort issues being solved in seated mode we might be able to use stick turn instead without getting sick!

          • wheeler

            Absolutely. Seated VR with a chair that doesn’t swivel or that has a high and wide back or arms that get in the way is very clunky. The swivel function is absolutely necessary unless the game is purely forward facing (unless one is in the minority that can tolerate smooth rotation). Stationary casters are like $10 a pack and are necessary for stability. Arms or a high back (forget head rests) impede turning your upper body and frustrate a ton of other motion controller movements. Once you address all of these things (which basically consists of removing parts) it becomes much more viable.

            Wireless or a slip ring adapter will solve the cord wrapping issue. Another thing you could do is attach a tracker to the chair and use that as your movement direction instead of using controller relative or head relative movement where you have inputs interfering with each other.

            VR is still early and I don’t think we should assume we’ve figured out the best mechanics, input, modes of interaction, etc etc. Things like physical exertion and play space size are glaring issues for VR. It was only a few years ago that the industry had basically assumed that smooth movement in VR wasn’t going to be a thing outside of cockpit games but now it’s the dominant form of play.

          • Gerald Terveen

            “VR is still early and I don’t think we should assume we’ve figured out
            the best mechanics, input, modes of interaction, etc etc.”

            Amen to that mate!

            I use an old massive armchair that I have put on wheels. Love that chair, but it does not rotate – allows me to move very freely though and using a cushion the arms are low enough to not hinder me at all.
            I also have nice swivel chair – but that one I need to make some space for …

          • I’ve been using a wheeled mechanics stool for seated VR over past couple of years, it’s height adjustable with gas piston and has 360 degree unrestricted rotation.

            I only use for Aircar and Assetto Corsa (only seated games I regularly play) but its ideal, especially with Xbox wireless dongle plugged into index Frunk. As long as the tether is routed carefully before starting, it’s very usable. It’s also great for cleaning my headset while getting ready as my VR room has no furniture!


          • This has been my favourite for seated VR for past 4 years. Getting the balance point right (lean back a bit) for Aircar feels amazing, height adjustable gas piston is really good for getting to your seated depth just right.


          • wheeler

            Looks like it will get the job done. Can you share the model? Also, have you considered replacing those rolling casters with stationary casters? That would give you much more stability. I’ve been using these (without the foam pads)

            I first messed with an old stool I had laying around but it was too squeaky and unstable. Then I tried an Amazon Basics Drafting Stool but I ended up returning it due to instability. Now I’m using a Steelcase Jack Drafting stool with the arms and back removed–what I like about it is that it’s really sturdy and can extend high enough that your thighs are at an angle which mitigates some of the motion controller obstructions caused by one’s legs. The seat is a bit too wide though. Along with this, I’ve been hacking around with a modified version of openvr input emulator that allows me to smoothly translate up and down while seated (wish SteamVR allowed this natively) and I find the combination to work really well.

            I’m curious about the model stool you’re using though because I’d like to put together a solution for your average person (that is, at a price your average person interested in seated play can stomach). In particular, does it rotate smoothly and is it at all noisy? Also, how is the stability of the seat? That’s one major issue I’ve noticed with a lot of swiveling stools–there’s often a bit of play or shifting in the seat which results in unexpected movements in VR that don’t feel good. I think most people try seated VR in their average high-backed gaming/office chair with arm rests but that kind of setup interferes with too many VR movements.

          • It’s made by “Park Tools” for bicycle workshops. It’s really solidly built and has no wobble. It’s always used on anti fatigue workshop matting (can see in image) so sits stably and doesn’t move unless unweighted.

          • wheeler

            I ended up getting this Park Tool STL2 stool. So far I’m really liking it. It’s way more stable than my other stool. Still has a slight bit of wobble but it doesn’t do so unless I intentionally attempt rock it (and it may just be inherent in the design of stools used gas lifts). The seat is really small so there’s basically no chance of controllers hitting it. It’s omnidirectional so the gas lift lever can be moved to the back of the seat and out of the way of motion controllers. It’s also extremely low mass which minimizes the force necessary to rotate or stop rotating. There’s also an unused mounting point on it that looks like I can use to attach a tracker to.

            Only issue was that I could not for the life of me find stationary casters that would fit the chair. The casters’ stem diameters are an usual size (5/16″ or 8mm). After a while I gave up looking for matching ones and just dremmeled the wheels out of the casters and put some rubber pads on their place.

          • That’s the same stool I’ve been using since 2016, it’s a great choice. I really like using it as a base for setting up / cleaning my headset too.

            It works brilliantly for seated VR, in Aircar I lean back until I find balance point with my legs in front, feels like flying (be careful you don’t fall off!)

            Glad you managed to sort your caster issue. Happy VR! Cheers, Rob Cole


      • Norkiff

        ‘seated gameplay will stay dominant’

        Seated play has never been dominant and will not be in any time in the future, standing play is by far more popular and allows for greater freedom in VR games,

        I’ve only played a VR game seated twice, one of which was Aircar on Steam which is more of an experience than a game but you fly a flying car around a city, that’s it. It looks pretty, not a whole lot more. You can’t take your hands off the controls nor get out of your ship, you just fly around.

        The other was VTOL VR, again another flying simulator, you can’t get out of the plane and you never need to reach down below your waist to access any of the controls.

        Both of those games make sense to play seated, literally every other game I find it incredibly uncomfortable and inconvenient to play sitting down.

        • JB1968

          Have you played games like NMS or Skyrim in VR? This kind of games with hundreds of hours of gameplay is most popular among gamers and 99% people are playing seated(after the first wow effect).
          Of course simple action games with small play session in average are better to play standing…but absolute majority of gamers have very small play space so it usually ends up sitting if possible.

    • Keith

      Studies are designed to eliminate variables. If standing, then you have to consider that everyone has a different skill level of balance. If user-defined movement, then every move is a variable. Putting people on rails means that everyone has the same experience. The point is to compare the effect of the haptics across multiple people. Eliminate all other variables as much as you can.

      • Seated VR takes ankles out of generating proprioception, very important for balance.

        Cues from the ankles indicate the body’s movement or sway relative to both the standing surface (floor or ground) and the quality of that surface (for example, hard, soft, slippery, or unevenDuring quiet standing, human balance is achieved by constantly reconfiguring ground reaction forces under the feet to counteract the sway of the body. The point of application of the vertical ground reaction force vector is known as centre of pressure (COP)1. COP displacement is one of the most common techniques to measure human balance during quiet standing (a technique known as posturography or stabilometry)

      • Everyone has a different amplitude of proprioception, inherent to the body/brain connection of each individual; some people have “body intelligence” with an enhanced sense of proprioception – they may experience motion sickness more readily due to the mismatch being more obvious, for example, well trained physical athletes.

        Balance “skill” isn’t as variable (unless working incorrectly) as you may think, poor sleep will actually degrade standing balance more than natural variation between individuals. Balance is simply the ability to maintain the body’s centre of mass over its base of support.

        By sitting, you remove very ‘loud’ proprioception cues from the ankles, which are primary signal for detecting standing surface and texture of surface through body movement and sway. Understanding direction of head turning comes from proprioception generated during neck movement.

        I’d be very interested to see this run again with standing participants, and the outcome!

  • kontis

    This was quite a popular concept a year or so ago on Reddit, some people even tested DIY solutions. I think there was even an anti-motion-sickness device like that sold on Amazon.

    Palmer Luckey promised us an open source solution (not sure if vibration based, though), but it’s probably not happening.

    • Gerald Terveen

      I was wondering about the hints Palmer made back then as well, but I admit I don’t remember the details. Makes you wonder if it has been abandoned, or if some clause in his Facebook contracts shut it down or if it is still in the works.

    • Ad

      “an open source solution ”

      Now why would Monogon do that?

      • Andrew Jakobs

        He already has an opensource solution for AI assisted driving (not that it would be the best around, but it was opensource).

  • Ad

    This says nothing on its own, having them be seated is a massive problem and means they didn’t test the actual use case. Same with not having them control it, although that would have improved comfort for most people. You could always reach out and do a little experiment yourselves.

  • wheeler

    What about more advanced haptics like miraisens that can give you the actual sensation of a force? Is the nerve density in the face / head high enough for the same effect to work?

  • Do you remember the vibrating headband that created some hype in the VR communities because it could remove motionsickness of some years ago? Pepperidge farm remembers

  • sfmike

    Very intriguing research.

  • JB1968

    They should have put them into bunch of Wipeout tracks to see the real effects ;-)