A method is presented for investigating the response of passive and active muscle to experimentally induced deep muscle pain. Ten healthy adult males performed multiple, submaximal isometric neck flexion tasks before and after pain had been induced in the sternocleidomastoid (SCM) muscle by injecting 5 ml of a hypertonic (5%) saline solution. A similar volume of isotonic saline was injected into the contralateral muscle as a control. Cervical myoelectric signal (MES) root-mean-square (RMS) response to each injection was recorded from eight neck muscles over 8 minutes. The subject rated perceived pain at regular intervals using a visual analog scale (VAS). Sternocleidomastoid pain, which reached a mean of 36 mm on the 100-mm VAS 2 minutes after the injection, resulted in significantly increased (1-2 microV, P less than 0.05) RMS MES in the otherwise relaxed SCM muscle during the first 2 minutes. This was followed by a gradual return to control values after 5 minutes. A similar trend in MES was found for the active SCM muscle during a 10% maximum voluntary contraction (MVC) isometric neck flexion effort. Thus, acute deep muscle pain caused subtle, yet systematic, changes in motor output in both the relaxed and active painful muscle, and its synergists and antagonists.