The Measurement of Critical Masking Bands The characteristics of the critical masking band, that spectral region of a wide-band noise that is effective in masking a pure-tone signal, were inferred by measuring detectability (d′) for tonal signals as a function of the cutoff frequency of a low-pass or high-pass noise masker. As the cutoff frequency of ... Research Article
Research Article  |   September 01, 1975
The Measurement of Critical Masking Bands
 
Author Affiliations & Notes
  • Robert H. Margolis
    UCLA School of Medicine, Los Angeles, California
  • Arnold M. Small
    University of Iowa, Iowa City, Iowa
Article Information
Research Articles
Research Article   |   September 01, 1975
The Measurement of Critical Masking Bands
Journal of Speech, Language, and Hearing Research, September 1975, Vol. 18, 571-587. doi:10.1044/jshr.1803.571
History: Received June 21, 1974 , Accepted May 12, 1975
 
Journal of Speech, Language, and Hearing Research, September 1975, Vol. 18, 571-587. doi:10.1044/jshr.1803.571
History: Received June 21, 1974; Accepted May 12, 1975

The characteristics of the critical masking band, that spectral region of a wide-band noise that is effective in masking a pure-tone signal, were inferred by measuring detectability (d′) for tonal signals as a function of the cutoff frequency of a low-pass or high-pass noise masker. As the cutoff frequency of a low-pass noise was decreased from the wide-band (100–7000 Hz) condition toward the signal frequency, (500, 1000, or 4000 Hz) detectability maintained a constant minimum until a further reduction in cutoff frequency increased detectability, presumably due to a reduction in masker power within the critical band. As cutoff frequency was reduced further, detectability increased monotonically until detection reached 100%. This usually occurred when the cutoff frequency is 0.04 to 0.06 octaves below the signal frequency. The range of cutoff frequencies over which detectability changes occurred defines the critical masking band. These ranges correspond closely to well-known critical ratio data. The dependence of d′ on noise cutoff frequency did not differ at the two signal levels (15 and 25 dB SL) used in this experiment. The critical band appeared symmetrical about the signal frequency for most subjects and most experimental conditions, although some subjects displayed a marked asymmetry in the high frequency direction for some conditions.

Order a Subscription
Pay Per View
Entire Journal of Speech, Language, and Hearing Research content & archive
24-hour access
This Article
24-hour access