Traumatic spinal cord injuries clinically occur in a heterogeneous fashion, including at different spinal levels, injury velocities, and injury mechanisms. Clinical treatment options, such as early surgical decompression produce inconsistent recovery outcomes in the patient population, despite demonstrating effectiveness in preclinical animal models. The most common biomechanical factors, such as cervical level, high-energy impact and dislocation injury mechanism, are not adequately represented in preclinical models, which may explain the lack of agreement between clinical studies. The overall objective of this thesis was to investigate the biomechanics of a high-speed cervical dislocation rat model at acute stages, refine the model, and incorporate residual compression. The temporal progression of acute SCI was investigated in different injury mechanisms, where dislocation injuries demonstrated the fastest loss of white matter tissue. To refine the dislocation model, new vertebral injury clamps were designed with a feature allowing the clamps to pivot and self-align when tightened. The vertebral kinematics during a dislocation injury were analysed using high-speed x-ray and clamp slippage was significantly reduced with the self-aligning clamps, compared to the existing clamps. This study also emphasized the importance of validating injury displacements against input parameters, particularly when comparing results or reproducing injuries. In order to implement residual compression within the dislocation model, injury parameters were independently investigated. Electrophysiology techniques were implemented to determine a minimum residual compression depth that affects signal conduction following a traumatic injury. Continuously holding the residual compression following the initial injury induced a significantly different physiological response compared to when the injury was immediately reduced. Behavioural outcome was used to identify severities following a range of displacements, and four hours of residual compression was survivable following a ‘mild’ traumatic injury, indicating suitable parameters for future studies. Rats of the same weight were identified to have different anatomical dimensions and structural properties of the spinal column, potentially influencing injury outcomes in closedcolumn models. The continued development and implementation of the cervical dislocation injury model in the rat will deepen understanding of SCI biomechanics and provide an additional clinically-relevant injury model for testing the robustness of potential treatment therapies.