4.2: Classes of Filters

Last updated
Save as PDF

Page ID: 87946

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

There are four main classes of filters used in EEG/ERP research. They’re typically named in terms of the frequencies that they pass, not the frequencies that are filtered out (much as an air filter passes air and filters out dust).

A low-pass filter passes low frequencies and filters out high frequencies (e.g., muscle activity).
A high-pass filter passes high frequencies and filters out low frequencies (e.g., gradual drifts resulting from skin potentials).
A bandpass filter passes an intermediate band of frequencies and filters out the lower and higher frequencies. A bandpass filter is the same as filtering twice, once with a low-pass filter and once with a high-pass filter.
A notch filter passes all frequencies except for a narrow band (e.g., 60 Hz). A notch filter is typically used during an EEG recording when AC electrical devices produce so much contamination of the EEG that it’s hard to see the signal.

I don’t ordinarily recommend applying notch filters (unless they are necessary during the recording process); it’s usually better to use a low-pass filter that attenuates all the high frequencies. However, if you don’t want to use a low-pass filter with a cutoff of 20 or 30 Hz (e.g., because you are interested in relatively high-frequency activity), a very sophisticated line noise filtering approach (Mitra & Pesaran, 1999) is available in EEGLAB as the cleanline plugin (see Bigdely-Shamlo et al., 2015 for important details about implementing this tool). Another tool called Zapline is can also be used for this purpose (de Cheveigné, 2020; Klug & Kloosterman, 2022), but it is newer and hasn't yet accumulated a strong track record.