Fig. 5From: Rigorous optimisation of multilinear discriminant analysis with Tucker and PARAFAC structuresInterpretation of components from the Stekelenburg&Vroomen experiment. Spatial and temporal patterns corresponding to the extracted spatial and temporal filters found from the training data without subject 5 in the Stekelenburg&Vroomen data by the following (from top to bottom) one component models: CMDA, DATER, ManTDA, BDCA. Logistic regression coefficients are shown above the spatial patterns. a CMDA b DATER c ManTDA d BDCABack to article page