This research was completed as part of an ongoing effort to characterise human thoracic response to belt loading in a well controlled and repeatable laboratory environment. This paper presents the results of eight tests conducted on three post-mortem human subjects. The sled test environment provides realistic occupant kinematics and restraint interaction in the inertial environment of a vehicle collision, but is too complex for detailed analysis of thoracic deformation under belt loading. To study in more detail the kinematics of the chest when loaded anteriorly by a seat belt, three male post-mortem human surrogates (31-62 years of age) were mounted on a stationary apparatus that supported the spine and shoulder in a configuration comparable to that achieved in a 48 km/h sled test at the time of maximum chest deformation. The belt was positioned across the anterior torso with attachments at D-ring and buckle locations based on the geometry of a mid-sized sedan. The belt was attached to a trolley driven by a hydraulic ram linked to a universal test machine. Ramp experiments were conducted at rates of 0.5 m/s, 0.9 m/s and 1.2 m/s. Average peak sternal displacements ranged from 13% to 23% of chest depth measured at the central sternum. Belt loads and spinal reaction loads were measured along with six degree-of-freedom (DOF) displacement data of the sternum, 4th and 8th ribs anteriorly, 8th and 10th ribs posteriorly, and acromia bilaterally. Three DOF targets were mounted to the distal clavicles, 8th ribs laterally and along the path of the belt. The targets were tracked optically by a high speed 16-camera motion capture system (VICON MX™) in a calibrated space around the torso. Post-test analysis of the target motion included decomposition of the trajectories into Cartesian coordinate displacements with respect to a spine fixed coordinate system. The results showed that the chest deformation closely followed the belt loading regionally with a trough developing where the belt contacted the chest. The anterior rib targets exhibited three-dimensional (3D) translational motion. Displacements in the X direction (anterior-posterior) were the largest, however the Z (vertical) and Y (lateral) displacements comprised nearly 35% and 10% respectively of the total resultant deflection measured at the sternum. Peak posterior deformations were significantly (p < .001) lower than those observed at anterior locations and were below 7 mm except for the final injurious tests in which the peak posterior deformation averaged 13.4 mm (approximately 22% of the average peak anterior ribcage deformation in the injurious tests). Overall, the results provide a detailed 3D mapping of the chest deformation under belt loading, which should be considered in the future development of physical and computational models of the thorax.