While belonging to distinct diagnostic categories, patients with schizophrenia and major depression share symptoms, such as anhedonia, the reduced ability to experience and anticipate pleasure. Though recent neuroimaging findings begin to unravel the neurobiology behind this symptom dimension the extent to which anehedonia can be characterized by a similar underlying pathology in both patient groups is largely unknown. The Virtual Brain is a powerful tool ideally suited to answer this question, by combining neurocomputational models defining the function of cell populations with large scale structural connectomes derived from the patient’s brain scans. Calculating these connectomes, however, is computationally intensive. Here, we aim to use imaging data from 25 patients with schizophrenia, 40 patients with major depression and 55 healthy controls, in conjunction with neurocomputational modeling to predict empirical data such as symptom ratings, neurophysiological measures of brain function, as well as the effect of modulation of the neural networks affected in anhedonia on these outcomes. Particularly, we want to see whether the same network models can predict these markers in both patient groups.