Although restless legs syndrome (RLS) is a common neurological disorder, it remains poorly understood from both clinical and pathophysiological perspectives. RLS is classified among sleep-related movement disorders, namely, conditions characterized by simple, often stereotyped movements occurring during sleep. However, several clinical, neurophysiological and neuroimaging observations question this view. The aim of the present review is to summarize and query some of the current concepts (known knowns) and to identify open questions (known unknowns) on RLS pathophysiology. Based on several lines of evidence, we propose that RLS should be viewed as a disorder of sensorimotor interaction with a typical circadian pattern of occurrence, possibly arising from neurochemical dysfunction and abnormal excitability in different brain structures.