A computational head-neck model was developed to more efficiently study dynamic responses of the head and neck to near-vertex head impact. The model consisted of rigid vertebrae interconnected by assemblies of nonlinear springs and dashpots, and a finite element shell model of the skull. Quasi-static flexion-extension characteristics of ten human cadaveric cervical spines were measured using a test frame capable of applying pure moments. The cadaveric motion segments demonstrated a nonlinear stiffening response without a no-load neutral zone. Computational model parameters were based upon these measurements and existing data reported in the literature. Geometric and inertial characteristics were derived from three-dimensional reconstructions of skull and vertebral CT images. The model reproduced the shape and timing of the cervical spine buckling deformations observed in high speed video of cadaveric studies of near-vertex head impact [ 1 ]. Head and neck force histories and head acceleration histories agreed with those reported in the cadaveric studies. A sensitivity analysis of the model parameters revealed that head and neck responses were most sensitive to changes in head stiffness, head mass, and flexion-extension properties, suggesting that an appropriately configured deformable head and accurate experimental characterization of motion segment flexion-extension behavior are critical to reliable model predictions. This validated, computationally efficient model is well suited for large-scale parametric studies of the role of impact surface properties on injury risk. It also serves as a foundation for future model enhancements such as the incorporation of the cervical musculature.