Purpose: Although sham acupuncture as well as real acupuncture are known to reduce low back pain (LBP), the brain correlates of placebo effect in acupuncture is not clearly understood. So we dissociated the brain correlates of somatosensory needling from needling credibility effect induced by a novel form of sham acupuncture “phantom acupuncture”. Methods: Forty three LBP patients were randomized into the real (REAL, n=23) and phantom (PHNT, n=20) groups. In acupuncture session, the REAL got real acupuncture at left ST36, left SP11 and bilateral SP13 points (five times stimulation per each point in a random order with inter-stimulus interval of 17.8±1.7 seconds) at around 2Hz for two seconds per stimulation, while the PHNT sawonly a videoclip (recorded needling manipulation in REAL session) to create needling credibility. LBP intensity (VAS) was measured before and after acupuncture session. Results: PHNT experienced needling credibility with visual stimulation while REAL experienced both needling credibility and somatosensory stimulation. In PHNT we found activation in somatosensory processing areas including SI, SII and anteri or cingulate cortex but not in posterior insula and thalamus where activated in REAL. In unpaired t-test, REAL shows greater activation in sensory motor areas. Also activation in rewarding and pain processing areas (periaqueductal gray, nucleus accumbens and putamen) were observed in both REAL and PHNT. Activation in pain evaluation area (inferior frontal gyrus) is observed in PHNT while signal deactivation in cognitive processing area (dorsolateral prefrontal cortex) in REAL. Greater signal deactivation in default mode network (DMN) was observed in REAL. In difference map, differences in activation intensity in ACC, posterior insula were strongly observed. Conclusion: Somatosensory afference/needling induces greater deactivation in DMN area as well as sensory motor area, while needling credibility, a contributing factor for placebo effect, activated pain and rewarding processing areas and interestingly somatosensory areas which probably due to the visually induced sensory expectation. Contact: Meena M. Makary, mmakary@khu.ac.kr