Both evoked potentials and cognitive tests may provide useful information which cannot be derived from the clinical observation. For this reason, there have been some attempts to use EPs in monitoring the natural history of the disease and in assessing the efficacy of therapeutic trials. However, no conclusion can be derived from the few available data. Although MRI is more sensitive than EPs in revealing new lesions in brain, cerebellum and brainstem, EPs are more sensitive in revealing optic nerve and spinal cord lesions. Moreover, the poor relationship between brain MRI abnormalities and disability has raised the possibility that cognitive evaluation may be an additional sensitive marker of brain involvement over time. Since the gold standard for the assessment of disease activity is uncertain, it is therefore advisable that frequent MRI, EPs and cognitive assessment may integrate clinical outcomes measured by conventional scales, both in the study of the natural disease course and in monitoring clinical trials.