You’re missing that the nerve activity-reading device Nathan Crone and his team installed in a patient doesn’t use EEG, which detects voltages produced by the brain via electrodes attached non-invasively to the patients scalp, but ECoG, which detects voltage produced by the brain via an array of electrodes implanted by a neurosurgeon, invasively, in the brain.
Hmm, perhaps. But take a look at this article:
It seems to me that they do have a very good EEG(sort of), or am I missing something?
ECoG is much more accurate and sensitive than EEG because because each implanted electrode can pinpoint (pun intended) the firing of a small collection of neurons, while EEG must measure small differences between the voltages of electrodes at different places on the scalp to narrow down the area in which neurons are firing. To use Billvon’s analogy of trying to read the activity of a car’s computer, an EEG is like listening with an array of sophisticated sensors stuck on the outside of the car, while a ECoG is like soldering wires into the internal parts of the computer. The EEG is a more sophisticated system, but it’s less effective because it has to cope with so much more noise due to the distance between its sensors and what it’s trying to sense.