Abstract
The sensitivity of a contraction of the tympanic muscles in man elicited by sine-wave bursts is determined by recording the change in the acoustic impedance measured in the auricular canal. The sensitivity is defined as the intensity required to give a certain percentage of the maximal impedance-change. The sensitivity determined in that way is well-defined for impedance-changes above ten per cent of the maximum. The re-producibility is good in contrast to what is the case when the sensitivity of the muscle contraction is defined as the threshold of contraction. The sensitivity as a function of frequency of the muscle contraction is compared with the threshold of audibility. The sensitivity, defined as intensity required for ten per cent of maximum impedance-change is found to be almost parallel to the threshold of audibility, with a difference of about 80 db for contralateral stimulation. The sensitivity defined in this way is also compared with the response of a network model of the middle ear. The intensity required for ten per cent of maximum impedance-change is compared with the electrical input voltage for constant current output of the model. The electrical current output represents the volume velocity at the stapes. Good agreement between model and ear is found between 200 and 1500 c/s except around 500 c/s, where a peak was found in the sensitivity of the muscle contraction in the ear. A corresponding peak was not found in the response of the model. This peak was assumed to originate from the leakage between the ear and the cushion of the dynamic earphone which was used for stimulation. This was approximately compensated for by using a calibration of the telephone, while it is applied to the ear measuring by sound pressure at the entrance of the auricular canal instead of using the calibration made in a 6 cc coupler. After this compensation is made a closer agreement is obtained between the network response and sensitivity for a ten per cent impedance-change.

This publication has 8 references indexed in Scilit: