For convenience we are going to put two electrodes side by side, about 2 inches apart, on the forehead--just below the hairline. One of these should be the neutral lead. The third electrode, which should NOT be the neutral one, goes on the back of the head for maximum EEG signal. There is a bump at the back of the skull where it sticks out farthest, just above the neck. Go an inch above that bump. Prepare the scalp, mix up some electrode paste, clean and prepare the electrodes as described in that section.
Homemade electrode paste may not be enough by itself to hold the electrodes on. Some skin-friendly tape should take care of the two on the forehead. For the third electrode I took a 26-inch piece of double-sided velcro, an inch wide, and wrapped it around the head from the posterior (occipital) electrode to just above the eyebrows. This should be just snug enough to keep the occipital electrode secure.
After connecting the Arduino amplifier, start the Processing program as described under "Installing the Software for Arduino." (Once it's installed, the Arduino will run the acquisition program indefinitely every time it is plugged in, until you do something to change it.)
Or, using the simple preamplifier and the Mac audio input, hook up the leads, plug the jack into the audio line in plug, start the appropriate CB program, and turn on the preamp power.
On PCs without an Arduino, the easiest combination is the Behringer U-Control and Zelscope, although the maximum signal height will be relatively small. Using the Behringer U-Control and MSII for EEG allows a larger display but is less direct, because MSII only displays 250 mSec of data in real time. However, it has the advantage of showing you very clearly whether you have excess 60-Hz noise (or muscle contraction artifact). I have found this approach to be more difficult for EEG. If you try it, first plug in the U-Control and let Windows recognize it. Then start MSII; it should also recognize the U-Control as input. Then plug in and turn on the simple preamplifier. You will have to play with this one.
The famous alpha rhythm comes out with the subject awake, relaxed, and eyes closed. It varies greatly in amplitude among normal subjects. Below are two examples of my EEG, which is not particularly high amplitude, using the Arduino. (Click on the figures to enlarge them.)
Five seconds of EEG sampled at 180 Hz, which is the fastest the processing program runs on my laptop, and the best result. Arrow 1 points to the "artifact" from the moving lids as I close my eyes. Arrow 2 points to the "artifact" as I open them. In between is a train of alpha waves around 10-11 Hz that is maximum for a few seconds after eye closure.
Ten seconds of EEG sampled at 120 Hz. On the left is a series of eye blinks ending in eye closure (arrow 1). On the right is a train of alpha. Arrow 2 shows eye opening.
Below is a sample of my wife's EEG using the simplest amplifier and CB on a Macbook Pro. There is an eye closure on the left,
with alpha most prominent just afterwards. Note the high frequency hash throughout. This is 60 Hz noise with an unfiltered amplifier.