Biodynamic measurements of human response to vibration may be nonlinear as a result of active control of the response, complex body movements, or tissue characteristics. Experiments using a random Gz stimulus were designed to explore the extent of these nonlinearities under normal conditions. The input transfer function was found to exhibit near-linearity, but transmission measures suffered from errors arising from complex body movements. The technique developed was found to be valuable for investigation of the variables affecting human response, and examples are given of the effects of body restraint and the relationship between intraabdominal pressure and input acceleration. It is argued that improved techniques and careful consideration of variables lead to a greater acceptance of the biodynamic approach.