Notched-noise measures of frequency selectivity in adults and children using fixed-masker-level and fixed-signal-level presentation.

Three experiments were performed to examine the development of frequency selectivity and to attempt to separate peripheral versus central contributions to frequency selectivity. In Experiment 1, frequency selectivity was examined in 4-year-old children, 6-year-old children, and adults, using a fixed-masker-level, notched-noise masking method. Thresholds were determined in a no-notch condition and in conditions in which the notchwidth was 0.3 times the center frequency (0.3 x fc) or 0.6 x fc. The results for the adults and 6-year-olds were similar, but the notched-noise functions of the 4-year-old listeners were relatively shallow. These results were consistent with an interpretation that frequency selectivity is relatively poor in 4-year-old listeners; however, the results of the 4-year-old listeners might also be accounted for in terms of poor processing efficiency. For example, the children may require a relatively high level of signal-related excitation in order for detection to occur. If the growth of excitation is steeper in the no-notch condition than in notched-noise conditions, then thresholds in the notched-noise conditions might be elevated because of poor processing efficiency rather than poor frequency selectivity. This interpretation is consonant with past data showing that the growth of loudness of partially masked signals steepens as a function of increasing masker level: The masker level at the output of the auditory filter centered on the signal frequency would be less for the notched-noise masker than for the masker having no notch. In Experiment 2 the growth of loudness was examined for three adult listeners in the no-notch condition and the 0.6 x fc notched noise. Results showed a steeper growth of loudness in the no-notch case. In Experiment 3, a notched-noise measure of frequency selectivity was used in which the signal level was fixed at 15 dB SL and the masker level was varied adaptively. In this method the noise level at the output of the auditory filter centered on the signal should theoretically be constant over the different notch conditions. Adult and 4-year-old listeners were tested. The notched-noise functions were similar between the adult and 4-year-old listeners. This result supports an interpretation that the shallow notched-noise, fixed-masker-level functions of 4-year-old children may be due to poor processing efficiency rather than to poor peripheral frequency selectivity.