The National Highway Traffic Safety Administration (NHTSA) has utilized acceleration sled testing for decades to allow for the evaluation of occupant kinematics, occupant protection systems, and occupant exposure in many modes of accidents. Such systems have the advantage of being non-destructive, having relatively low turnaround time between tests, and precise control of input parameters. These systems usually involve placing some portion of a vehicle occupant compartment on a rigid sled and then accelerating the sled with a desired linear acceleration pulse profile. Such systems have been effective in the development and evaluation of air bag systems, seat belt restraints, and occupant interior surfaces. A roll simulator has been developed to allow for similar parametric testing of vehicle/machine occupant compartments. Such occupant compartments range from those in which the operator is driving the vehicle in a similar arrangement to an on-highway vehicle, such as a recreational off-highway vehicle (ROV), to those on construction, maintenance or material handling equipment. The use of this roll simulator as an occupant kinematic and occupant protection system parametric evaluation tool relies on the ability of this system to first, provide repeatable and applicable dynamic inputs, and second, to allow for the discrimination between differing occupant protection systems. For the current work, a series of tests was completed with a recreational off-highway vehicle to determine the effects of varying occupant protection device configurations. Evaluation of the results shows that the roll simulator is an effective tool for observing subtle differences in occupant response and interaction with various restraint systems.