Partial Least Squares (PLS) Analysis was first developed in the late 60’s by Herman Wold, and works on the assumption that the focus of analysis is on which aspects of the signal in one matrix are related directly to signals in another matrix. It has been developed extensively in chemometrics, and has recently been applied to neuroimaging data. In the application to imaging data, it has been used to identify task-dependent changes in activity, changes in the relations between brain and behaviour, and to examine functional connectivity of one or more brain regions. PLS has similarities to Canonical Correlation in its general configuration, but is much more flexible.
