Mismatch negativity (MMN) is a neurophysiological response measured as an auditory event-related potential (ERP) generated automatically by the brain when a deviant sound occurs rarely and unpredictably amongst regular background sounds. A reduction in the amplitude of MMN in schizophrenia is one of the most replicated results in schizophrenia studies first reported approximately 30 years ago. Considering the large effect size of the MMN reduction and other characteristics of MMN in schizophrenia, it has been proposed that MMN have the status of a neurophysiological biomarker of the disorder. In this paper, we review the evidence that has accrued to support the assertion of biomarker status, and then evaluate what that means for diagnosis, prognosis, and prediction of medication response, but most importantly, its use as a translational tool in pre-clinical studies. Although considerable knowledge has accrued over the last decades about the optimal stimulus conditions for exposing MMN impairments in schizophrenia, the association with cognitive deficits and functional status, and the involvement of the glutamate NMDAr system, little is known about the neurobiological basis of MMN reduction in schizophrenia. Pre-clinical research using animal models is essential to make advances in this area. Fortunately, paradigms have been developed for use in the rat that strongly suggest that the rat brain is capable of generating the equivalent of a human MMN response or MMR. We review evidence supporting that a late component of MMR in the rat exhibits many of the characteristics of the human MMR. What is needed now is an animal model of schizophrenia that produces a robust and replicable reduction in this component that can be investigated using molecular approaches to support the development of novel pharmaceuticals to rescue MMR in these animals.