Fig. 3.

Prototypical data structure and decomposition into factor loadings and factor scores. P = Participant, E = Electrode, t = voltage at a specific sampling point for a specific observation, F = Factor, λ = factor loading (representing the time course of the activity for each factor), η = factor score (representing the level of activity of each factor for a specific observation). The residuals were left out for the sake of comprehensibility; but typically loadings and scores reconstruct the participant average data only approximately.