I’ve been digging into this for a while and scanned through many forum posts, blogs and podcasts (including OpenBCI forums) and all I can find is either a $5,500 Headset and software (NuerOptimal), in which case i rather go to my local NF clinic and pay $4000 to someone who knows what they are doing, or find some hacked together software and hardware that are extreme un-user friendly, and are not even clear if they are effective or what they can achieve (the closest one is OpenBCI + BrainBay).
Is this stuff just too complex still? Are we not quite there yet?
I don’t mean to come off negative, reading back what I wrote it might sound that way. Juts a little exhausted chasing this thing. It is pretty obvious how useful and effective Neurofeedback is and it is still so inaccessible, such a bummer.
@Jay What type of neurofeedback are you interested in?
I’m not experienced with neurofeedback, but It’s reasonably easy to get an EEG system up and running with OpenBCI and BrainBay or OpenViBE. If you have questions about that I can probably help.
The good thing about the hacked together software and hardware is that the folks who are familiar with it are often quite helpful. I’d recommend getting in touch with William Croft. He’s experienced with neurofeedback, and uses the OpenBCI.
(I recently ran across OpenNFB, which is a new player on the block. It look like it’s aimed at technical users. I’ll report back once I’ve tested it.)
Good question Jay. Unfortunately, I think the answer is probably yes: there is still no truly affordable, user-friendly, and powerful neurofeedback system available. There are various proprietary, low-channel-count “brain training” devices (MindWave and Muse, for example) which are fairly inexpensive and reportedly user-friendly, but they are far from a general-purpose neurofeedback system.
One option would be to go with OpenBCI hardware and a (relatively) inexpensive proprietary neurofeedback software option like BioEra. I have not used BioEra, but it is compatible with OpenBCI and is much more geared towards neurofeedback than BrainBay or OpenViBE. A somewhat more expensive hardware option is Pocket-Neurobics, which works with BioEra or BioExplorer, and is apparently popular among home neurofeedback trainers. As @curiositry mentioned, William Croft is a great resource on this topic.
If you’re thinking of try out the OpenBCI, here are a couple of threads on the OpenBCI forum that you might want to check out:
@omphalosskeptic, i’ve been listening to andrew hill on both joe rogan’s show and dave aspery’s show and the main things I’ve heard that I am really interested in is mostly improving sleep (by training up sleep spindles according to him) as well as training focus and attention.
@AdamM, I actually own both the Muse and the OpenBCI. The Muse it seems is somewhat superficial for these purposes, and my experience with it has been a bit off. I do not feel the connection between anything I’m doing internally and the external effect of the sounds feeding back to me. Also, and I don’t know if this is just something he says or is legitimate, but according to Andrew Hill, these frontal area only systems are insufficient for having real effect, they don’t have the top of the brain for SMR training and don’t have enough electrodes anyway. Also they run a QEEG assessment to establish a baseline. But I don’t know enough about this to know how relevant this part is.
In any case, I am starting to doubt whether I am as technical as I thought I was, or I haven’t developed the thick skull to bang my head against the wall enough to get these tinkerer gadgets to work, including the OpenBCI (which is at least much better than the OpenEEG which I tried building a few years ago).
I’ll take a look at BioEra, but you might be right and there is a level of complexity here that hasn’t reached consumer fit and finish yet. But I do believe that the interest is big enough now that we will start seeing some more of these soon.
There’s a new free VPL visual programming language app available, called ‘neuromore’. This is similar to what BrainBay, BioEra, Bioexplorer do. Supports a range of devices. My small post on the OpenBCI forum is here,
Thanks for the link William. Looks interesting for sure, and if it has good built in NFB features it could fill a market slot that’s currently empty (especially if a free version remains available).
Have you tested Neuromore yet?
@Jay, what kind of technical problems did you run into with the OpenBCI? It’s not consumer-grade user friendly, but it’s likely the folks here could help you through some of the obstacles to getting your system running.
@wjcroft neuromore looks very interesting, thanks for the link.
@AdamM well, without even getting to the technical stuff, to start off I was under the wrong impression that my kickstarter reward comes with the head mount thing (because how else would you set all the electrodes?) but it didn’t, nor is it being sold separately on the website, and unfortunately I do not have a 3D printer to print one. So I am not quite sure how I am supposed to both place all of them correctly, or hold them in place. Taping each one to my head sounds a little crazy, am I supposed to go through that process every time I want to work with it? I need something I can put on or off my head easily. Am I the only one? But again, my main concern is location, I feel like I am approximating to much trying to place these.
@Jay getting user-friendly headgear for the OpenBCI is a challenge. I believe Conor @russomanno15 is actively working on the Ultracortex 3D-printable headset system, and once this design is stabilized I’m sure commercial vendors will emerge (there was an offer to print OBCI headsets for $50 on this OpenBCI forum thread from last November).
As of this tweet yesterday, it looks like there might be some Ultracortex headsets available commercially in the very near future.
In the mean time, have you tried just sticking the electrodes to your head with 10-20 paste? You will not get super accurate placement, but for limited number of electrodes this can be a surprisingly effective solution. Other NeuroBB members will be more knowledgeable about this than me, but AFAIK in many cases the precision of electrode placement is not super critical for neurofeedback purposes.
Disagree. There is a lot of affordable Neurofeedback hardware/software out there. It is BCI hardware that is lacking. With BCI applications/research you want as many channels as possible, and that makes it expensive. In fact, just a decent set of cables and electrodes will cost a few hundred $US, which is more than most people will want to pay. Then we haven’t even spent anything on the amplifier.
For Neurofeedback you need two channels only and the market for these devices is so (over) crowded that I stopped further development of my own 2-channel systems a few years ago.
NFB and BCI are 1000 miles apart. I don’t do either and have spend my time more on the hardware development side. However, I must have spent 200+ hours following Pete van Deusen’s BrainTrainer yahoo group that only deals with NFB, so although no hands-on experience, I know a thing or two about NFB applications. I have discussed this subject in some detail on the OpenVibe forums.
In a nut-shell, if you want to do NFB, you get a decent 2 channel amplifier and BioExplorer. Doing anything else will be wasting your time and money. Of course you could use openBCI hardware, or some other BCI hardware, but it would be silly. If I want to plant a tree, I would use a spade to dig the hole. Using a back-axle digger or a spoon would give less favorable outcomes, although in theory, you could use said devices as well.
To learn NFB protocols, THAT is the hard part. Yes, you could use BrainBay, OpenVibe or any other software, but then you would need to implement training protocols from scratch on these software systems. With BE you have a huge community and you have access to training modules, etc. The only other practical NFB software is BioEra, technically more powerful and advanced, but it is much more challenging to use and learn. More problematic is the fact that there is not a big community to draw on.
I would say that 99% of NFB is done with 2-channel bipolar systems. I have seen here or there a person having reported that he has created a 4-channel protocol.
Thanks for joining the discussion @Stefan. I’ve been experimenting with OpenBCI, but haven’t dug into neurofeedback yet, and I have a few questions for my own edification.
I’m assuming you’re referring to OpenEXG, am I correct? (Also, I wasn’t sure whether OpenHardwareEXG is related to your project or just has a similar name…)
What two-channel EEG system would you recommend for neurofeedback?
Do you know of any kind of Wiki or guide to neurofeedback protocols, so I could learn more about how to implement a particular protocol and how it works? I’m thinking of a software/hardware agnostic directory with info like:
description of the protocol (what frequency/frequencies are being uptrained/downtrained)
the protocol’s intended effect, and why is produces it (if it’s not self evident)
recommended electrode montage
example algorithm to detect the if feedback conditions are met
recommended feedback modality
If there was such a resource, it’d sure be very handy for folks new to neurofeedback. If there isn’t, somebody should make it. (I don’t know enough about the protocols to be able to generate the content, but I could help with other areas).
@AdamM Thanks. I’ll wait for the UltraCortex sale i think. Sticking those to my head is elaborate and messy, and I wouldn’t want to do it for no specific reason, or just to see my brainwaves. My main interesting in any of this is Nuerofeedback for both personal use and research, the second is sleep research and sleep applications.
@Stefan I am not sure what kind of NF you are referring to but from what I understand 2 channel gives you very little to play with. Even in training for things like attention, let alone things like sleep spindles. Most if not all of the comprehensive scientific studies are conducted with a full head QEEG, not 2 channels. I have the Muse, and I am not looking to simply gain a little more “calm” and “focus”, which I feel I gain just as much from meditation itself than from the feedback provided by the system.
So unless we are using the same terms to mean different things, I don’t think we are on the same page. And if you are absolutely certain there are, as the title of my post asks, “user-friendly” neurofeedback systems, I would love links to some examples.
This line seem to miss completely what I am asking about.
@curiositry I somehow got my password screwed up and couldn’t reset it (system always tells me the link is too old and I should click on forgot password again). So now I am Stefanj
Well not really, OpenEXG was the first design, but that is rather “old”. I did some further work after that, which was never published. Some people will know that I even developed my own, low-cost 8/16 channel unipolar amplifier. The prototype is still lying on my desk… but I never got to write the embedded software for it. What actually happened is that due to the $300 Emotiv’s lack of raw EEG data, I wanted to rewrite the firmware of that device’s dsPIC. But my comments regarding the low input-Z of that design are well know… I came to the conclusion that spending my time re-writing code would be a waste. So then I thought OK, let’s spin a new hardware design, based on some ideas of Emotiv. At the end, my design diverted quite a bit, I used a different 16-bit ADC and different microprocessor, it was also not a wireless solution.
If I where to buy a 2-channel amplifier I would try the Neurobit Optima-2. Pocket-Neurobics seems to have some nice stuff as well, but I am hesitant recommending it. The old Pendants where very popular but the build quality was plain crap. Even their frontend design was lousy, not having ANY static protection or EMI filtering. At least on the one model that I got to inspect. I hope and think that Neurobics improved, but you know what they say about a leopards spots… can it change them?
I have been out of touch with NFB and EEG hardware dev. for the last 2 years, so I lost touch a bit. But I’d say that anyone interested in NFB should subscribe to the yahoo braintrainer group and follow their discussions. That is my recommendation.
@Jay If I speak of neurofeedback (NFB), I refer to the things the Othmers do and that Peter van Deusen (TLC) does. Protocols for ADHD, OCD, Autism, PTSD, depression, dyslexia, sleep insomnia, brain balancing, etc.
You are right that the scientific studies on things like sleep, etc. are mostly done on 24/32/64, etc. channel equipment. But clinical applications are more concerned with mapping the brain, rather than “closing the loop”. The medical profession still largely regards NFB as some sort of quackery. This view comes courtesy of big Pharma who will attack and black-smear any treatment regime that might reduce their profit takings on the little useless pills.
@Stefanj I have not seen any studies or protocols for the above done with 2 channels or anything less than 24. Do you have any links?
I do agree about the medical profession still being problematic with all of this but there are legit people who are well regarded in scientific circles (like Adam Gazeley) who are doing with that’s being taken seriously. But again, none of what they do either is with minimal brain wave reading, it’s all full head.
Jay, I assume you’ve seen Pete Van Deusen’s Brain-Trainer site. An abbreviated or full head assessment is done. Then protocols recommended, typically 2 channel; but also 4 channel. Most home users of this system are using the Pocket-Neurobics 2 or 4 channel amps. Stefan Jung has already mentioned it. I’ve used Pete’s protocols since 2008, they are effective. Bioexplorer and Pete’s protocols are also a great jumping off point for writing your own protocols.
Many professional neurofeedback trainers do 2 and 4 channel training: Brainmaster, Thought Technology, EEGer, Cygnet (Othmer), LENS, NeurOptimal, NewMind (Richard Soutar), etc. Yes, a small minority are using full cap sLORETA, but that’s expensive with commercial systems, $6000 and up. If you want to do Z-score training, add in the normative database. Obviously consumer / home users are going to bring all of this down from the stratosphere. But don’t dismiss 2 and 4 channel training protocols.
Thanks William. So please help me understand here, as I seem to be missing something or have some wrong information or something. I also definitely have a bias against websites that have not been updated since the 90’s (seriously, these are killing me).
If you agree that one needs a full head assessment, a 2 channel system isn’t sufficient. No?
There is no one single site that I’ve landed on, including all the ones that you mention, that have only a few or a single thing that you buy that is everything you need, pre built, easy to use and affordable. So again, and feel free to call me stupid here (it just might be the case), but it is not clear at all what I can buy, how do I use it and what does it do.
If you are willing to indulge me (any of you), this might clarify my issue: imagine I am a non technical person who just heard about neurofeedback for the first time today, I come to you and say, “that sounds amazing, I can use a device and improve my brain or cognitive performance or mental states, or my sleep etc’? sweet, where do I buy one of these? and how do I use it?” What would you say?
If we just take the example you linked to, which is actually the only place I found anything remotely close to a “everything you need” offer, that starts at $3500. So again, maybe the answer is, there isn’t anything out there like what you are asking and the only thing is too expensive and everything is not user friendly or incomplete. Or i’m still missing something and your answer would be something else. I do hope to understand why most here seem to think I am way off.
Pete has a variety of packages on his site. The entry point is about $1000 which includes a Pocket-Neurobics 2 channel amp and Bioexplorer. Another hundred or so dollars for a subscription to Pete’s protocol package.
The whole head assessments are done with 2 or 4 channel amps, 2 or 4 channels at a time. Brainmaster has a similar concept called Mini-Q. Yes, it’s not quite as accurate as a full 19 channel parallel QEEG. But many clinicians use these Mini-Q type of assessments. Another popular one is the Swingle Clinical-Q, similar idea.
After doing your home assessment, you can send the data in to Pete and he’ll look it over and send you a protocol list that you can run. You could also devise this using your own skills, if you take one of his tutorial courses. His Excel spreadsheet can process your data and do some summary stats.
Granted, this might not be as turnkey as you are looking for. But in a sense you are learning some basic “auto-mechanic” skills. So you definitely need some familiarity with this level to perform and monitor your own home training.
I think you said you were familiar with the NeurOptimal system. This is the only one I know that is completely automatic. It’s working on different concepts than traditional band training. So does not have the need for assessment. Although those stats are available with the full version (at the C3 C4 sites.)
In my experience, there are no neurofeedback solutions that are both completely automated (require no technical skill whatsoever) and work for everybody. Neurofeedback depends highly on the individual, which makes it really hard to automate the protocols.
The products that promise a one-click solution always rely on very simple protocols that work on a broad range of individuals, to some extent.
This explains the lack of easy-to-use products, but not the high price. I’m not sure why even non-certified devices and software is sold at such high prices. The only explanation I have: Because they can. Most EEG devices and software is bought for medical and research applications with money from large budgets. These customers also almost always require that the software and devices are medically certified, which is highly expensive for the manufacturers.
I cannot recommend any of the medical grade products for private use - they are just not worth it and require a lot of knowledge to operate. I could, however, recommend consumer devices like the InteraXon Muse or devices from Emotiv. They are truly the future of personal EEG, and there are a lot more devices to come in the future. Their signal quality is surprisingly good, most of the limitation is caused by the dry electrodes (which is the core problem the EEG device makers are working on). Wet electrodes don’t sell.
I suggest you get an Insight (better spatial resolution) or a Muse (only frontal lobe, ear electrodes are useless) and check out our neuromore Studio v1.0, which we will release next week. It supports the Insight ;).
It’s not a one-click solution but it allows you to experiment, build own protocols and do everything you can think of with the measured data. You can even create the neuroptimal protocol, if you know how it works.
Oh I just read that you already have an OpenBCI; we currently support that on windows, too. And BTW: precise electrode placement does only matter if you got many electrodes and want high spatial resolution! If you measure only frontal lobe or the left/right hemisphere the location does not matter very much.
You can also buy dry electrodes and build a simple headset from a cap or elastic straps yourself.
Manu, thanks for that reply. Can you clarify the business model for neuromore? Will the Studio always be free, or are you looking to transition to some kind of Freemium model. Your website mentions that the ‘engine’ and the Studio are separate entities. Are you hoping to license the engine to neurofeedback app builders, etc.
Your statement that NeurOptimal “relies on very simple protocols” is not true. Do some investigation into Val Brown’s research.
@Manuel_Jerger: Welcome to the forum, and thanks for joining the discussion! Neuromore looks quite interesting, and I look forward to testing it with the OpenBCI.
I second @wjcroft’s question about the long-term business model — it looks like you’ve got plenty of people with business background on your team. I assume they are there for something other than releasing free software.
Will neuromore come with sample NFB protocols, or do you have plans to develop a way for your users to share and collaborate on protocols, as @Curiositry suggested?
Perhaps it is a result of my general ignorance about how most neurofeedback protocols work, but wouldn’t it be possible to establish a software-independent standard format (perhaps XML-based) that could then be shared and used in various systems (with conversion/translation as necessary)? I’m curious what those experienced with NFB think about this idea.