We recently had a meeting at the Astron radio telescope for the COGITO project with Daniela de Paulis, Stephen Whitmarsh, Guillaume Dumas and others. One of the goals of that meeting was to try out the combination of the EEG system with the Oculus Rift VR system.
For the COGITO project we are using the GTec Nautilus EEG system. Our specific system comprises of a 32-channel wireless amplifier that mounts on the back of the EEG cap, in combination with EEG caps in three different sizes. The caps have 64 holes at a subset of the locations of the 5% electrode placement standard. We are not using the “Sahara” dry electrode option, but rather the regular wet electrodes.
We started by removing all electrodes and cups from the cap, to get a clear view on which electrode sites are accessible. The central electrode locations (i.e. the z-line), temporal electrode locations and occipital electrode locations are occluded by the VR head mount. But there are still plenty of electrode locations accessible.
We are using the Nautilus in combination with wet electrodes. These consist of a small cup that is mounted in the holes of the cap. Each cup comes with a label. It is a bit fiddly to mount all the cups on the cap; not something to do every day.
The electrodes themselves are fixed to the wireless amplifier and “click” smoothly onto the cup. We made some small adjustments to the selected electrode sites to have them all fit nicely in the spaces of the VR head mount.
The Nautilus EEG system has two sets of wires, going to the left and right. That is convenient with the head mount.
Putting on the VR headset on top of the EEG cap and removing it again requires care, it is easy for electrode wires to get stuck. But once the VR headset is mounted over the EEG cap it all fits nicely and is comfortable for the subject.
Below you can see Guillaume, wearing the EEG and VR system while he is seeing the 3D movie and while EEG is recorded.
Hi, how does this nautilus helmet and electrode costed you? Or have you been able to rent it?
The list price for the 32-channel system is somewhere around 15,000 euro. We bought it for the Cogito project, which G.tec generously supported by offering the system to us with a significant discount.
Is this research published? If it is, can you link it to me? I am doing research close related to this, for my master thesis, so I’m very interested.
This is not research, we are just a bunch of enthousiasts messing around with fancy equipment 😉
See the COGITO page on FaceBook if you want to follow this.
Hello Sir, I want to ask you, Could I use these pictures for my lecture ppt materials?
I’m just surfing about the Collaboration between VR and EEG system for neuroarchitecture.
Dear Plaffordance, thanks for asking. Hereby I grant you permission to use the photo’s from the blog post “EEG combined with VR” for your lecture material. Best regards, Robert
Sir, I sincerely thank you for permission to use related photos. Thank you.
Hi Robert! I am a PhD student at Leiden currently working on a project with a combined EEG-VR design. I was wondering if you had any tips for minimizing EM or any other artefacts in the EEG signal. We use a Biosemi ActiveTwo system in Leiden. Please let me know if you require any further info. Thanks!
Hi Huseyin,
The frequency of the electronics in the VR headset (and potentially the digital radio) is much higher than that of the EEG, and the EEG is low-pass filtered prior to digitization to avoid aliasing. That means that electronic noise of the headset, even though large in amplitude, might be sufficiently suppressed and not visible in the EEG. Furthermore, EEG in general is noisy anyway, also if you record it without a VR headset. So the artifacts you see might be more general in nature and relate to your task, your subjects behaviour, your EEG system, or the environment, rather than to the VR.
Hence the first thing to do is to check whether you have any electromagnetic interference and whether you have other artifacts that do relate to the VR. If not, there is no problem – or at least no problem other than you would have recording EEG anyway. If you do see interference or additional artifacts, the nature of those can give you information on how to deal with them: either take action to avoid them, or to clean the data afterwards.
I suggest that you do an EEG recording, alternating with the VR headset switched on and off (say for a minute each, and repeat that multiple times). That allows you to look for differences in the spectrum of the continuous EEG signal. In the same way you can experimentally evaluate other artifacts, for example movement artifacts (have the participant move with the VR headset on and without the headset and check for differences in the artifacts); it might well be that the artifacts are not related to the VR headset but merely to the movement.
Besides trying to avoid artifacts and cleaning the data, it is also important to consider the confounding effects on your analysis. Eventually you might want to do a statistical comparison; for that you should ideally design your experimental contrast such that there is no confounding effect (i.e., the artifacts – if present – are not different between conditions).
best regards,
Robert
Thanks, Robert!
Hi Robert,
Thanks for putting this up, it looks interesting!
Did you synchronise the start of the start of the stimuli to the EEG data using markers in e.g. LSL, or did you use timestamps within the data to link them? Or was this timing it not an issue for your setup?
Thanks for the help,
Phil
Hi Phil,
Timing was not an issue as we did not use it in an experimental situation with tight control of the stimuli, but only in a very loosely coupled neurofeedback setup. Actually, perhaps it would be better to describe the setup as an open-loop system, as there was no intentional feedback to the participant. Furthermore, data was only directly processed and transmitted and not stored for later offline analysis. See https://doi.org/10.1002/essoar.10503880.3 and https://cogitoinspace.org for further details.
best regard,
Robert
Thanks Robert, very interesting!
Dear Robert. I was thinking of running some studies combining EEG (e.g. brainproducts liveamp 64 ch – wireless https://www.brainproducts.com/solutions/liveamp/) with VR (Quest 3). This is for an action learning task, where however I only care for data during stationary foreperiods preceding participant actions (where they are stationary). My main concern relates to number of channels I might have to drop in order to strap the HMD on participants’ heads. I see there are various options for straps, but all configurations inevitably occupy some real estate on the scalp…. is there an ideal strap option for EEG? Do you think electrodes can be placed under a strap or you just have to drop channels no matter the configuration? I assume electronic artefacts are also an issue, but as I read above these can be identified and removed from the signal. Are there other challenges you think I might face?
Thank you!
Hi Nicola,
The straps for the VR goggles are indeed a challenge w.r.t. flexible and uniformly spaced electrode placement. If you have relatively flat electrodes, I can imagine that you could try to add some padding to allow electrodes below the strap. The GTec electrodes that we used here are quite high (probably more than 10 mm) and consist of a cone-shaped cup that holds the gel, with the actual electrode on top of it. The LiveAmp electrodes might be flatter.
Imagine a small donut-shaped piece of foam, for example a circle cut out from a thin camping mattress or yoga mat, with a hole in the middle. If the electrode fits in the middle of the donut and does not stick out, the foam would provide padding and the VR goggle strap could go over it without directly pushing on the electrode. I would imagine that I would prepare a bunch of these donut-shaped rings with double-sided adhesive tape, then do the EEG preparation as usual, then apply some of the rings over central and temporal areas, and then place the VR strap on top of those.
Whether this would work with regards to movement artifacts and participant comfort I don’t know; we simply avoided the straps with the electrodes as we only had 32 channels and a 64 channel cap, and did not need uniform coverage for the art project. Perhaps there are others that have published ideas like this in research papers.
Good luck, and please keep me and others that come across this page posted on your experiences!
best regards, Robert
Thanks Robert for this advice. I see what you mean, and could be a viable option. I’ll look into electrode dimensions & cap characteristics to see whether your solution might be helpful. Let’s see where this goes. I’ll keep you posted!