Simply EEGLAB

ICA, simply explained

Station 4 · Lesson 1 · EEGLAB 2026.1

Script

eeglab nogui;   % puts runica on the MATLAB path
srate = 128;
t = (0:10*srate-1) / srate;
brain = sin(2*pi*10*t);                        % steady 10 Hz rhythm
n = round(0.3*srate);
win = 0.5 - 0.5*cos(2*pi*(0:n-1)/(n-1));       % smooth bump for one blink
blink = zeros(size(t));
blink(round([1.5 4 6.5 9]*srate)) = 1;
blink = 5 * conv(blink, win, 'same');          % four large, slow blinks
muscle = 0.5 * randn(size(t));                 % fast, irregular activity
S = [brain; blink; muscle];
A = [1.0 0.8 0.3;     % frontal channel: strong blink
     0.9 0.2 0.4;     % central channel
     0.7 0.1 1.0];    % temporal channel: strong muscle
X = A * S;
[weights, sphere] = runica(X, 'extended', 1);
act = weights * sphere * X;
figure;
for k = 1:3
    subplot(3, 2, 2*k-1); plot(t, X(k, :));   title(sprintf('Channel %d (mixture)', k));
    subplot(3, 2, 2*k);   plot(t, act(k, :)); title(sprintf('ICA component %d', k));
end

Which value should I choose?

In the idea	In EEGLAB	What it means
The way to recombine the channels	`EEG.icaweights` and `EEG.icasphere`	What ICA learns. Multiplying them by the data gives the components.
The shape of a component in time	`EEG.icaact`	The time course of each ICA component.
Where a component is on the head	`EEG.icawinv`	Each column is the scalp map of an ICA component: how much it contributes to each channel.
Number of components	Equal to the data rank, which is never more than the number of channels	In D1 after average reference and with no interpolated channel: 31 components from 32 channels. Each interpolated channel removes one (see Running ICA).

Check

You can explain the party
You know the two parts of an ICA component
If you ran the script

Script

Which value should I choose?

Check

More