| back to text |
|
|
Although our slow frequency sweeps yielded excellent measures of phase-shift as a function
of frequency, we could not be certain they were providing good measures of the linear
transfer ratio (transducer gain) and its dependence on frequency. We would need that in
order to draw strong inferences concerning the peripheral dynamics of the saccular receptors.
Therefore we began to make use of fixed-frequency measurements at very low levels-- where
the units with background spike activity yielded robustly linear behavior. This is an example
of the sorts of a tuning curve obtained in that way. Again, we used Mike Chin's fast Fourier
transform. The steep slopes on the high- and low-frequency sides of the pass band imply
relatively high dynamic order. In 1983, this was about to become a major issue for us.
updated 07/20/07