Although there have been many studies on the acoustics and vibration of voice registers, the underlying physiological mechanisms of different registers still remain unclear. This study represents a preliminary effort toward identification of regions in the physiological domain that would lead to different voice registers, using a computational phonation model. The physiological parameters under consideration include vocal fold geometry, stiffness, resting glottal angle, subglottal pressure, and vocal tract shapes. Three registers are considered, including vocal fry, chest, and falsetto. Voice registers are identified based on known or reported acoustic features, including F0, closed quotient, and harmonic structures, based on which the physiological conditions corresponding to each voice register are determined. Preliminary perceptual experiment will be performed to confirm register identification. [Work supported by NIH.]