| Description |
Amplitude modulation (AM) detection measures a listener's sensitivity to temporal envelope fluctuations. AM signals are ecologically relevant because the amplitude of speech fluctuates over time. The post-cochlear representation of AM may be influenced by processes that occur in the cochlea, where signals are subject to cochlear compression and adaptive mechanisms that modulate the cochlear response such as the medial olivocochlear (MOC) reflex. Specifically, cochlear compression may reduce the difference between high-intensity peaks and low-intensity valleys (i.e., effective modulation depth) of AM. Furthermore, gain reduction of the cochlear amplifier via the MOC reflex is hypothesized to decompress the cochlear input-output function and thus improve the AM effective modulation depth at moderate levels. To test these hypotheses, AM detection was measured for a narrow-band, high-frequency carrier (5000 Hz) for conditions that do or do not elicit the MOC reflex. These conditions take advantage of the sluggish onset of the reflex, which exhibits an onset delay (?25 ms) upon stimulation. Specifically, AM detection was measured as a function of level for a 50 ms carrier in the presence and absence of a long ipsilateral notched-noise precursor. A longer carrier (500 ms) without a precursor was also included. For no-precursor condition, AM detection thresholds at moderate carrier levels are poorer compared to low and high levels, consistent with a reduced effective modulation depth due to cochlear compression. In the precursor condition, AM thresholds improved monotonically with carrier level, with the largest improvements seen at moderate levels. This improvement is consistent with decompression of the cochlear input-output function via the MOC reflex. For 500 ms carriers, AM detection thresholds improved by a constant (across all carrier levels) relative to AM thresholds with a precursor, consistent with the longer carrier providing more "looks" to detect the AM signal. In a second experiment, AM thresholds were measured as a function of modulation frequency to examine whether the effects of the precursor depend on the modulation frequency. The results showed that the improvement in AM detection with compared to without a precursor is limited to low modulation frequencies (<60Hz). The experiment in Chapter 3 was designed to examine the effects of cochlear compression on the inherent fluctuations of narrow-band noise carriers. To test this, AM detection was measured for short and long, high- and low-fluctuating noise carriers as a function of carrier level. The results showed that AM thresholds for short, low-fluctuating noise carriers worsened as carrier level increased from low to mid carrier levels and then improved with further increases in carrier level, as found in the previous experiment. This is consistent with greater cochlear compression at moderate levels. For high-fluctuating carriers, AM thresholds were roughly constant across carrier levels. For high-fluctuating carriers, low-level linear and mid-level compressive cochlear response growth may have resulted in constant envelope signal-to-noise ratios, due to the cochlear response growth equally affecting the target modulation and inherent carrier fluctuations. Thus, AM detection for high-fluctuating carriers is constant as a function of carrier level. |
