Digital Image Correlation techniques are applied to sequences of images of argillite rocks samples at various saturation states at both the global scale of the sample and the local scale of the microstructure. Not only it is confirmed that the clay matrix deforms much more than the other mineral inclusions, but it also appears that the deformation is very inhomogeneous in the matrix, with some areas almost not deformed, while others exhibit deformation twice the average overall strain, depending on the local distribution of the inclusions. Strain heterogeneities are also detected at the scale of the sample of saturated rocks and can be correlated with the presence of a network of cracks. On such samples, DIC analysis shows that the overall strain results both from the bulk deformation of the sound rock and the closing or opening of these cracks, which is confirmed by the analysis of acoustic emissions.