Objective: The interactions between sleep, neck muscle activity, and cervical spinal pain were examined in a controlled study with nine patients suffering from idiopathic cervical dystonia (ICD; also referred to as spasmodic torticollis), and nine gender- and age-matched controls. Methods: From each participant, two all-night polysomnograms with additional electromyographic recordings from the sternocleidomastoid and upper trapezius muscles were obtained. The first night was for habituation to the laboratory environment; the second night for experimental data collection. Visual analogue scales were used to collect intensity and unpleasantness ratings of cervical spinal pain before and after the second sleep recording. Results: None of the standard sleep variables showed statistically significant differences between average values of both groups of participants. However, a significantly larger variance in sleep latency was obtained for the ICD patients. In general, abnormal cervical muscle activity decreased immediately when lying down without the intention to go to sleep. Subsequently, abnormal muscle contractions were gradually abolished in all ICD patients during the transition from relaxed wakefulness to light NREM sleep. Following this transition phase, no more abnormal EMG activity was found in any of our patients. Finally, cervical spinal pain intensity and unpleasantness were reduced by about 50% overnight. Conclusions: Both supine position and sleep can be associated with an improvement of symptoms of ICD, and this disorder does not induce any sleep perturbations.