Deep Brain EEG Measurement

The cortex is the “bark” on the outside of the brain–about a quarter-inch thick. There is no “deep-brain cortical structure.” There are a number of structures in the mid-brain which are heavily involved in how the brain works, but there’s a problem.

EEG is produced by one (of many) type of neuron called pyramidal. The cortex has a high concentration of pyramidal neurons, and it is on the surface of the brain, so we can record reliable EEG from it. Sub-cortical structures, like the amygdalae, the thalamus, etc. are NOT made of pyramidal neurons, and they don’t produce a measurable EEG signal. There are two sub-cortical structures that include pyramidal neurons–the cingulate, which runs beneath the cortex under the midline (all the “x” sites)–and the hippocampus, the brain’s memory center, inside the temporal lobes on each side of the brain. It’s possible to see signals from those areas on the EEG.

If you want to see directly what is happening sub-cortically, you have to use MEG (magneto-encephalography) or fMRI. These do allow us to “see” subcortical areas, but they don’t have the immediacy of EEG (which literally allows us to see changes in parts of a second) and they are very large and expensive. So researchers, wanting to know what was happening sub-cortically, developed a technique called LORETA. Essentially, they recorded EEG from the cortex and calculated a bunch of algorithms to “show” the sources of the EEG in the cortex. The LORETA basically can show a cool 3D image or video of what the calculations say is happening. The idea, however, that you are “training deep into the brain” is kind of silly. You are reading the EEG on the surface, and that’s all you can train.

Here’s the really interesting part. Back when I was starting as a trainer in the early 1990s, we had 1-channel amplifiers bigger than an encyclopedia volume that were pretty slow in terms of sampling rates and pretty rudimentary software, but you know what? We knew how to train the sub-cortex without LORETA and did it all the time. How is that possible? The same way people calculate sub-cortical sources today.

There’s an interesting difference between cortical and sub-cortical sources of brain activation. When you see beta frequencies appearing on the surface of the brain, they are local (they should appear in circumscribed areas) and they are ephemeral (they appear and disappear fairly quickly) because beta is produced by cortical neurons doing a job. In fact, they ONLY produce fast frequencies and produce when they are working. But if beta is the only frequency produced in the cortex, then what are theta and alpha and SMR and delta, etc.?

Cortical neurons in a healthy brain spend much of their time resonating to various rhythms produced from down below–like radio stations broadcasting from a single antenna that can be picked up over a wide area. For example, there are groups of neurons in the thalamus that generate slow alpha, others fast alpha, others SMR in the sensory-motor cortex and others for slow theta. Fast theta is broadcast from the hippocampus. When cortical neurons are not working but simply resonating to these frequencies, they use very little energy and tend to have access to sub-cortical areas (responsible for emotions and memories). So what we see on the surface, the cortex, gives us a hint to what sub-cortical areas are doing.

For example, when the amygdala is strongly activated by fear or rage, the temporal lobes beneath which the amygdalae appear tend to show fast activity out of the range shown elsewhere in the EEG. When the cortex is showing lots of slow theta in the sensory-motor cortex relative to SMR, we know that the slow-theta nuclei in the thalamus are dominating the cortex, so it is less active and capable. As a result, the brain’s ability to screen and monitor background noise is limited. That theta-dominant, low SMR or beta pattern shows up in the theta/.beta ratio as an indicator of ADHD. Cz is heavily connected also to the basal ganglia, which are involved in processing motor activation when we see that pattern, so physical impulse control tends to be poor.