Independent component analysis is an unsupervised learning algorithm that aims to decompose a dataset into a set of underlying additive signals, assuming they are statistically independent. The signals are found by optimizing an objective function that measures their non-Gaussianity. The result is a set of spatial weight maps and corresponding temporal profiles. It is particularly useful for data lacking other well-defined stimulus or behavioral structure. Here, ICA was applied to "spontaneous" activity data, and each of the resulting spaital maps are depicted in different colors, and superimposed in a maximum intensity projection.