Monopolar (or referential) montages are channels in which one electrode (usually the active, but that’s not necessarily true) is placed over an active EEG site on the head. Bipolar (or sequential) montages are channels in which BOTH active and reference electrode are placed over active EEG sites. T3/A1 has one electrode over a head site and one over a relatively inert site (the earlobe), so it is monopolar. T3/T4 has both electrodes over head sites, so it is bipolar.

A bipolar montage does not have to be inter-hemispheric (another common confusion). T3/A2 is an interhemispheric placement (one electrode is reading from the left, the other from the right hemisphere), but the right electrode is not over an active EEG site so it’s monopolar. F3/C3 is not an interhemispheric placement (both electrodes are on the left hemisphere), but both are over active EEG sites, so it’s bipolar.

Each channel has an active and reference electrode. Therefore you can have a one-channel monopolar or bipolar (three electrodes including the ground). If you use five electrodes (on most systems, though some require a ground for each channel), you are using two channels and either one can be monopolar or bipolar.

The ground can go pretty much anywhere on the body.

Theory

It appears that in any montage you are training those pyramidal neurons (which exist only in the cortex, cingulate and hippocampus) which are between the active and reference electrodes that happen to be oriented closest to parallel to the line between them and closest to that line. My interest in this topic has pretty much been limited to the more practical elements of it. I use a variety of bipolar montages that I have seen effective over a number of clients for various types of problems. Ditto monopolar montages. It makes absolutely no sense to me that T3/T4, which I use often, is measuring this way, since I don’t understand what is being measured. Are the electrodes measuring in a straight line between the two (in which case some neurons in the temporals and perhaps in the hippocampus are all that’s being measured) or do the electrodes somehow pick up signals in a line that runs all the way up over Cz and back down again? Since the distance between an electrode and the neuron being measured reduces the effect of that neuron, what the heck are we really measuring? Since many–perhaps most–of the pyramidal neurons in that line will be oriented up-and-down or side-to-side across that line–not parallel to it–it seems like a crap-shoot.

So yes, I get the theory, but I haven’t really found that it helped most people (maybe including myself) to “understand” what they are doing. So my mantra remains, if it works, do it. Certainly sometimes I’ve worked with clients for whom bipolar montages worked really well for specific issues, and I respect Noel and Jorge who basically train ONLY that way. But I’ve also seen clients who didn’t respond as well to C3/C4 as they did to Cz/A2/g/C4/A2. Some folks I’ve worked with love T3/T4, and others prefer T3/A1/g/T4/A2 sum/difference training or just amplitude training.

References

In any channel, you are training the difference between the signal at the active and reference electrodes. In a monopolar placement with an earlobe or mastoid reference, the reference is very close to 0. Whatever signal reported is happening at the active site. Whatever changes occur in the signal during training will be due to changes at the active site.

Bipolar montages combine active and reference signals from brain sites–and report the difference between them.

Two sites could produce a theta value of 6, for example, in a bipolar montage with 15u at one site and 9u at the other; or 23 and 29, or 2 and 8. A bipolar signal really doesn’t tell us anything about the amplitudes in various frequencies. It tells us about the difference in the amplitudes. You can train to increase the difference (Reward) or decrease the difference (Inhibit) between the signals.

One of the ways to do that is to change the phase relationship between the two signals. When two signals are synchronized, with peaks being subtracted from peaks, the result is a very small number, no matter how much of the frequency is actually being produced at the two sites. Active CH1 shows 28u of low theta; Reference CH1 shows 25u. The trainer will see a reading of 3u.

ANY differential signal is actually reading all the pyramidal neurons between the active and reference electrodes that are more or less parallel to the line between them. Neurons are oriented in all different directions in the cortex, because of the folding of the cortical sheet. So the further apart the electrodes are, in general, the greater will be the average amplitude there, and the closer together they are, the smaller (on average) will be the amplitude. This is partly because of the larger number of neurons being measured and partly because most brains can produce synchronous activity more easily over shorter distances. Even those people who really understand the physics of the brain aren’t absolutely positive about this stuff (our knowledge of that organ has changed hugely in the past 20 years and is still changing, as are the various models for explaining the unexplainable) you can spend a lot of time on it–or you can just recognize that people who work with the brain to produce changes have discovered that, in many cases, certain things work to achieve certain results. If you need to know why, then keep asking. If you want to use the technology for change, then do what works.