Estimation of the autonomic nervous activity from pupillary response

A possible method for estimating the activity of the autonomic nervous system (ANS) innervating the pupillary muscle is proposed. The method utilizes a model for human pupillary muscle plant as an inverse dynamic model so that the ANS activity can be estimated from pupillary responses. In general, the inverse problem of the dynamics determined by antagonistic muscles results in an ill-posed one. We showed, however, that in case of the pupillary muscle plant, it can be solved uniquely by utilizing certain constraints based on physiological evidence. It was shown that the nervous inputs estimated from experimental step responses indicated fairly similar characteristics to the result of physiological experiment. Thus, the method is useful to examine properties of the ANS activity and to reveal the mechanism of the pupillary control system. Further more, this method can be applied as a new non-invasive monitor of the ANS activity