Electroencephalography is by no means a new technology; first used to record the electrical activity of a human brain by Hans Berger in 1924, the phenomena of seemingly spontaneous electrical activity in the brains of mammals has been studied since as early as 1875. To put this in perspective, this was before Greece began using the Gregorian calendar. As a nearly century old technology based in a relatively simplistic design, one who has never looked into purchasing one before might easily assume that EEG devices have become affordable and user-friendly, like so many other technologies available to us today; perhaps this dire misconception is a side effect of being spoiled by the rapid rise of the personal computer as a consumer technology. Unfortunately, while researching the options on the market for various projects, I’ve found that these two elusive factors, affordability and ease of use, are not available in any single product.
EEG equipment affordable enough for small scale research and development carries the potential to usher in a rapid expansion in our ability to interface with technology and acquire data regarding the impact of substances and stimuli on the electrical activity of the brain. Many exploratory souls self-experiment with everything from purported “mind-enhancing” drugs to tDCS in the hopes of achieving elevated cognitive abilities; yet few, if any, collect any objective data for comparison. The technology is out there, sure. We have incredibly powerful EEG units that can monitor 100 leads, automatically correct for artifacts, and interpret the signal received. These machines can literally analyze their own findings as they are recorded and correctly diagnose whether a person is comatose or merely high on benzodiazepines. They also cost upwards of $30,000.
My first impulse was to buy an older, used EEG. Companies such as PEMED and Admar Neuro sell a plethora of used EEG devices and peripherals; surprisingly, I even found a few units for sale on Ebay. Seeing units with price tags as low as $500 made me hopeful. Email correspondence with representatives of these companies revealed that although most medical equipment can be sold only under order of a physician, EEG devices are not similarly restricted due to their non-invasive nature. My excitement grew, but I noticed a strange pattern. There was a price gulf. One can find a slew of machines around $500-$600… then a large void… and then another cluster of machines starting at around $4000. I had to really dig to make sense out of this and it turns out that the $500-$600 cluster represents the formerly dominant analog model, which has since been largely abandoned in favor of the more sophisticated and user-friendly digital versions. Digital EEG devices begin at around $4000 even when well used. I figured, no problem, right? After printing out the data it would be a breeze to scan the documents. Further research revealed that the EEG manufacturers each made their own priority size paper and print cartridges that are no longer in production. Used analog EEGs really aren’t an option and digital EEG units are still out of the price range for the majority of small scale researchers. Strangely enough, during my search I found that modern electroconvulsive shock therapy units can be obtained for as little as $500.
This unmet need for a less expensive EEG has enticed a number of smaller companies to offer products that supposedly fill this market demand. Take, for example, the NeuroSky Mindwave. I absolutely had to try it. I was excited. I really wanted to believe. It completely failed all of my expectations. The most glaring issue was that there was simply no way to print or save raw signal data. No, really — I downloaded every single free software package I could at the time and wrote exasperated emails to NeuroSky until my fingers bled. They advised me to purchase a $500 software bundle which “opened up” this capability. I chose not to do so. The reason being that it only uses one channel. You read that right: one channel. Furthermore, this one channel of information is collected with a dry metal contact electrode. NeuroSky doesn’t even advertise in terms of EEG bands, and has instead made up an entirely new set of terminology, claiming that it measures “mental states.” A moderately better product is the Emotiv Epoc line. The Emotiv Epoc uses 14 different electrodes and a gyroscope. It’s pricier, but actually does something similar to what it claims. Marketing shows people controlling cars and interfacing with computers and actually being able to do things consciously. If one watches the “mind-controlled car” videos, however, the electrode placement is obviously for electromyography, not encephalography. The car is being controlled by the electrical signals of the muscles of the face being contracted consciously by the wired up fellow in the seat. In a round-a-bout way this is a car being controlled by the mind, albeit with this reasoning turning the wheel with your hand is as well. One positive is that the Emotiv uses saline electrodes rather than a dry metal system. The saline system is far superior in terms of data collection capabilities. The primary reason why I was unsatisfied with Emotiv comes down to raw EEG data. As in…you do not have access to the signal. This doesn’t even make sense to me. Neither the NeuroSky or Emotiv headset do what an EEG is supposed to do: give you a recording of your brainwaves. There are now “advanced” models, such as the $750 “super” Emotiv unit which does provide raw data, but why would I have to pay this much more for the device to do what it’s intended to do? After researching these products, I concluded that purchasing an EEG wasn’t the answer. I would have to look into making one. In 1924, the big debate was whether or not retiring the Julian Calender system was actually a diabolical satanic scheme to make Christians celebrate Easter on the wrong day. I felt confident that if Hans Berger could make an EEG that recorded data, I could too.
A great resource for those interested in fabricating an EEG device is the OpenEEG project. Although the origins of this project are obscure, a visit to the wayback machine shows that the project was pretty well fleshed out by December of 2002, and a full set of schematics with open source software was available by late 2004. This is impressive considering that commercial digital EEG devices weren’t prevalent until the mid 90’s. A mere ten years later the technology was made available for nearly anyone with a bit of soldering knowledge and a couple of hundred dollars to spend. By 2005, even the soldering prerequisite was made obsolete by the European company Olimex, which offers a completely assembled two channel EEG for around $150 using the OpenEEG product design. I love this product, really… but while affordable, it has significant room for improvement in terms of ease of use. The hardware itself is simple. The documentation and driver installation process on the other hand is like some kind of sick puzzle. No one is to blame for this really. Keep in mind the OpenEEG project software was written entirely by unpaid folk with a passion, and it was intended to be installed in an operating system that is now five iterations old. It seems that each installation, even on PCs running the same OS, requires unique solutions. Interestingly, Olimex is even willing to provide a degree of customer support by email. Once installed, the EEG unit is rather easy to use — but the installation process itself is a real struggle.
Overall, the EEG-SMT unit from Olimex is best value option. As long as one is prepared to fiddle and tinker a bit during the initial set up, it certainly fulfills the inexpensive EEG at home niche. Our group, Science for the Masses, has purchased, installed, and performed initial testing of four of these units. While this purchase certainly cut into our project funding, it’s exciting as it opens up a range of future projects. Indeed, the fact that this device can be used to measure biofeedback from a variety of sources other than the brain (eg ERG in our current project, measuring the electrophysiological response of the retina to stimulation) is one of its most exciting features. Future endeavors will likely include determining the changes in patterns of electrical activity of the brain secondary to tDCS and nootropic use. Other potential projects include research in biofeedback and transcranial magnetic stimulation. I proposed purchasing one of the electroshock therapy units to somehow incorporate in a project, but was unfortunately vetoed by my team members. Finding and learning to use the OpenEEG design was an investment of time, effort, and funding, but it was an investment that we fully expect to pay off in allowing us to contribute to the understanding of how human abilities can be augmented.