Registrations of the summed action potential (SAP) of the tympanal nerve are more or less distorted by high and low pass filters, presumably positioned between excited tissue and recording electrode. Originally, i.e. undistorted, the summed receptor excitation forms a relatively rapid decreasing oscillation around a slowly decreasing DC-level (Fig. 3a). It is likely that the DC-level corresponds to mean activity, whereas the oscillation is a result of synchronous activity of the tympanal receptors. The time course of this receptor group activity, i.e. the oscillation of SAP, resembles—after reducing random variations by averaging—a sine curve. Intensive stimuli magnify the DC-level and the amplitude of oscillation (Fig. 5) as well as the number of periods (maximally 10–20). However the duration of periods is essentially constant: It is influenced neither by duration nor intensity of sound (between threshold and saturation of the SAP, Fig. 5) and neither by stimulus ramp nor frequency (Fig. 4). An additional oscillation of smaller amplitude is seen at high stimulus intensities. At all physiological intensities, and when rise times are chosen which are similar to those occuring in species specific stridulatory song, the time difference between oscillations coming from both tympanal nerves varies so extensively with the direction of the sound source that its localization on the basis of this physiological time difference is possible. Oscillatory SAPs are optimal for this purpose because, and similar to vertebrates, the measurement of the time difference may take place several times, i.e. between each pair of the consecutive peaks of both oscilla
展开▼
