BACKGROUND: Chronic neck pain is a common problem and is often associated with changes in sensorimotor functions, such as reduced proprioceptive acuity of the neck, altered coordination of the cervical muscles, and increased postural sway. In line with these findings there are studies supporting the efficacy of exercises targeting different aspects of sensorimotor function, for example training aimed at improving proprioception and muscle coordination. To further develop this type of exercises we have designed a novel device and method for neck coordination training. The aim of the study was to investigate the clinical applicability of the method and to obtain indications of preliminary effects on sensorimotor functions, symptoms and self-rated characteristics in non-specific chronic neck pain METHODS: The study was designed as an uncontrolled clinical trial including fourteen subjects with chronic non-specific neck pain. A new device was designed to allow for an open skills task with adjustable difficulty. With visual feedback, subjects had to control the movement of a metal ball on a flat surface with a rim strapped on the subjects' head. Eight training sessions were performed over a four week period. Skill acquisition was measured throughout the intervention period. After intervention subjects were interviewed about their experience of the exercise and pain and sensorimotor functions, including the fast and slow components of postural sway and jerkiness-, range-, position sense-, movement time- and velocity of cervical rotation, were measured. At six-month follow up, self-rated pain, health and functioning was collected. RESULTS: The subjects improved their skill to perform the exercise and were overall positive to the method. No residual negative side-effects due to the exercise were reported. After intervention the fast component of postural sway (p = 0.019) and jerkiness of cervical rotation (p = 0.032) were reduced. The follow up showed decreased disability (one out of three indices) and fear of movement, and increased general health (three out of eight dimensions). CONCLUSION: The results support the clinical applicability of the method. The improvements in sensorimotor functions may suggest transfer from the exercise to other, non-task specific motor functions and justifies a future randomized controlled trial.