Fibromyalgia Syndrome (FMS) is a chronic, widespread, musculoskeletal pain condition that affects approximately 2% of the United States’ population. Nearly half of those diagnosed with FMS experience an intense fear of pain and movement. The Fear Avoidance (FA) Model, has become the dominant framework used to explain the mechanism that perpetuates chronic pain and the development of pain-related disability. Research with chronic low back pain patients suggests that graded in-vivo exposure therapy, targeting fear avoidance, may improve patients’ health outcomes. However, graded in-vivo exposure has not been tested as a treatment for FMS. Thus, the purpose of the present study was to determine the applicability of the FA model to FMS patients and to determine whether the relationships between the FA model factors differed between participants who received a graded in-vivo exposure intervention (GIFT) and those who received an educational control intervention (Stanford Nutrition Action Program; SNAP). Participants were 19 people with FMS, recruited for a pilot study. Structural equation models were used to examine participants’ responses to measures that assessed the variables included in the FA model. The results indicated that the FA model did not fit the data provided by the sample, and that the relationships between the FA model factors did not differ significantly between the GIFT and SNAP groups at the post-assessment (p’s > .05). However, the relationships between fear of movement and pain-related anxiety, along with depression and disability, were significantly different between the GIFT and SNAP groups at the follow-up assessment (p’s < .05). As fear of movement increased, the increase in pain-related anxiety was greater among those in the SNAP group than among those in the GIFT group; as depression increased, the increase in disability was greater among those in the SNAP group than among those in the GIFT group at the follow-up assessment. Although there were not significant differences among these relationships during the post-assessment, it is possible that the effects of the intervention took longer to manifest within the sample. Future researchers should continue to modify and test the GIFT protocol among larger and more diverse samples.