Background: A forward head posture (FHP) is one of the most common types of poor head posture in patients with neck disorder. A prolonged FHP might increase pressure on the posterior cranio-cervical structure and exhibit reduced performance on a cranio-cervical flexion test (CCFT). CCFT is included to activate deep cervical flexor muscles and inhibit excessive activation of superficial cervical flexor muscles. Therefore, the selective activation of deep cervical flexors is needed for effective exercise for FHP. Objects: The purpose of this study was to compare muscle thickness between longus colli (Lco) and sternocleidomastoid (SCM) using ultrasonography in subjects with FHP depending on head support. Methods: This was a cross-sectional, case-control research design study. The ultrasonographic images of Lco and SCM were taken in 17 subjects with FHP during the 5 phases of the CCFT with and without a head support. Towel was used for supporting head to make the neutral head position in supine. Changes in muscle thickness during the test were calculated to infer muscle activation. Data were analyzed using repeated measures of two-way analysis of variance with the significance level of .05. Results: When subjects performed the CCFT with head support, there was a significant difference in muscle thickness of Lco and SCM (p<.05). According to a post hoc paired t-test, change of thickness of Lco was greater at all phases, and change of thickness of SCM muscle was less at phase 4 and 5 in condition with head support (p<.01) compared to condition without head support (p<.01). Conclusion: The result of this study suggest that applying head support for neutral head position during CCFT could be a useful method for activating Lco muscle without excessive activation of SCM muscle.