Here is a typical EEG reading:

(source: frontalcortex.com)

I understand that each row corresponds to the signal read between two sensors on a standard 10-20 (or 10-5) distribution setup (e.g. C3-P3).

What I'm missing here is: what waves are we actually seeing here - alpha, beta, etc.? Or do EEGs not really contain wave types, and instead just show voltage differentials between sensors?

  • 2
    $\begingroup$ Actually I believe you may be seeing all the bands and will need to do a fast Fourier transform to get the actual frequency, however, I could be wrong as I am studying the MEG currently, not the EEG. $\endgroup$
    – SolarLunix
    Feb 20, 2016 at 14:19
  • $\begingroup$ Thanks @SolarLunix (+1) - so I assume there's a way to use FFT to decompose V(t) (voltage as a function of time) back out into constituent waves? $\endgroup$
    – smeeb
    Feb 20, 2016 at 17:43

1 Answer 1


EEGs are often analyzed in the frequency domain, where signals are subjected to spectral analysis, typically by Fast Fourier Transformation, or FFT.

What an FFT basically does is decomposing a signal in the time domain into one in the frequency domain. It does this by decomposing the input signal (any signal, including EEG) into a series of sinusoids. These sinusoids are not present in the original signal; it is a mathematical trick. If you would add all the sinusoids back together, the approximate original signal is restored.

Typically not a single frequency is analyzed, but a band of frequencies, because the frequencies within these bands are associated with similar brain states. The frequency bands that are typically reported on are shown in Table 1, including the brain state they associate with as well as sample EEGs:

![eeg bands
Table 1. Typical analyzed EEG frequency bands. Source: Conorrus Somanno

- Nelson Garcia et al. Device and method for cognitive enhancement of a user (2014). Patent EP 2681729 A1


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .