Temporary threshold and loudness shifts: frequency patterns and correlations.

Two experiments studied the frequency pattern of TLS (temporary loudness shift) as a function of the level and frequency of the fatiguing sound. In experiment 1, the fatiguing tones were intermittent 375-, 1500-, or 3000-Hz tones (10 s on/10 s off) at 75, 80, 85, 90, or 95 dB SPL. The TLS patterns were established for a continuous, 60-phon test tone at different frequencies presented simultaneously with the intermittent fatiguing tone. In experiment 2, a 1000-Hz exposure tone with an intermittency of 10 s on/20 s off was used with a continuous 60-dB test tone at different frequencies. In both experiments, the total exposure duration was 60 s; TLS was measured 5 s after the exposure ended. For the lowest two exposure levels, the TLS pattern had one peak centered on the exposure frequency. As the exposure level increased, a two-peak pattern became evident, with the second peak at higher test frequencies. This finding could be interpreted as psychoacoustical evidence for the double (passive and active) mode of displacement of the basilar membrane. In experiment 2, a TTS (temporary threshold shift) measurement after exposure to a 45-min, 1000-Hz tone at 90 dB was added to the TLS sessions. The correlations between maximum TTS after a 45-min exposure and the TLS obtained after a 60-s exposure were calculated for each of the exposure levels and test frequencies used in TLS measurements. The correlation reached as high as 0.9 for TLSs measured at 1120 Hz after a 90-dB exposure; it was smaller but significant for TLSs at the exposure frequency. Despite these correlations, differences in the overall patterns of TTS and TLS suggest that they stem from two different mechanisms.