Abstract
The torsion pendulum model predicts semicircular canal responses when angular accelerative stimuli are relatively brief, but certain discrepancies appear when accelerations are sustained for a longer time. A recently proposed revision of the torsion pendulum model accounts for many of these discrepancies on the basis of a short-term adaptation process. The revised model was tested by comparing its predictions with the observed responses of subjects exposed to a series of prolonged accelerations. The model described these data fairly accurately, although minor discrepancies remained, suggesting the existence of additional adaptation processes. A new method is described for measuring the adaptation process described by the model. This method is based on a single measurement that can be obtained with relative ease for either nystagmus or sensation responses.