Abstract
In this paper we discuss the measurement of the temporal resolution factor (TRF). The major merits of this measurement are that it is simple enough to be used clinically and that the results are, within certain limits, independent of presentation level. Addition of a continuous background noise has been suggested to enhance the discriminative power of TRF between normal-hearing listeners and listeners with poor temporal resolution. However, we showed through a simple model that this background noise may introduce a strong dependence on presentation level, which must be taken into consideration. To test our model we developed a new measurement protocol. This protocol was applied in measurements of 5 normal-hearing and 9 sensorineurally hearing-impaired listeners. Their results could reasonably well be fitted by the model. The model predicts large measurement errors for TRF values determined with background noise. We found the standard deviations of TRF values, calculated from test and retest measurements, to be smaller in conditions without a continuous background noise. In addition, our results did not show the predicted enhancement in discriminative power between normal-hearing and hearing-impaired listeners. Therefore, the primary conclusion is that the determination of the TRF is not improved by adding a background noise.