The DSM-IV-TR definition of nightmares-extremely frightening dreams from which the person wakes up directly-is unnecessarily narrow. Other emotions (anger, grief) have also been reported in nightmares, and direct awakening from a bad dream seems to be unrelated to increased distress. In addition, assessment of nightmares is problematic. Polysomnographic recordings have an ameliorating effect on nightmare frequency, retrospective measurements tend to underestimate nightmare frequency, and persons with frequent nightmares may feel reluctant to fill out (daily) prospective measurements. For studying nightmares, it is necessary to distinguish idiopathic nightmares from posttraumatic nightmares, which are part of a posttraumatic stress reaction or disorder that may result from experiencing a traumatic event. Both types of nightmares have been associated with an elevated level of periodic limb movements, although only posttraumatic nightmares seem to be related to more and longer nocturnal awakenings. Nightmares have also been repeatedly associated with the general level of psychopathology, or the so-called personality factor neuroticism. Nightmare distress, the impact on daily functioning caused by nightmares, may function as a mediating variable. Several studies in the last decades have shown that nightmares can be treated with several cognitive-behavioral techniques. The cognitive-restructuring technique imagery rehearsal therapy is the treatment of choice for nightmares, although a randomized controlled trial with an attention control-group has not yet been carried out. Nightmares are more than a symptom of a larger (anxiety) syndrome and need to be viewed from a sleep medicine perspective: nightmares are a highly prevalent and separate sleep disorder that can and should receive specific treatment.