Hippocampus is implicated in several major episodic learning and memory disorders and many therapies have worked in mice but they did not translate in humans. I hypothesize that a part of the reason is the way we probe learning and memory in rodents, using spatial navigation tasks. Related, place cells are found in rodents but not found in primates, despite extensive efforts. On the other hand, primate and rodent hippocampal neurons show a diversity of non-spatial responses. I will share a novel theory of the hippocampus that can reconcile these observations, and show supporting experimental evidence obtained using virtual reality. This would provide an effective technique to diagnose, translate and treat learning and memory disorders.