Judged Noisiness of Modulated and Multiple Tones in Broad‐Band Noise

Paired‐comparison judgment tests, undertaken to investigate the subjective noisiness of single, modulated, and multiple tones plus noise, clearly confirm the need for corrections in calculating perceived noise levels of spectra containing strong discrete frequency components. The tone‐correction procedures of Little and Kryter‐Pearsons both provided a considerable improvement in calculating perceived noise levels that agreed with judgments. The tests indicate that the tone corrections may continue to increase in magnitude as a function of tone‐to‐noise levels even for levels greater than 25 dB as measured in $13‐oct$ bands. However, the rate of increase diminishes for higher tone‐to‐noise levels. Little difference was found in the judged noisiness of unmodulated and either frequency‐ or amplitude‐modulated tones, except at rates of 5 Hz, where the modulated tones were found to be more annoying because of envelope irregularities. For multiple tones, little difference was noted in the judged noisiness of harmonic and nonharmonically related tone complexes. There was also found to be little consistent trend in the judged noisiness of two‐ or five‐tone complexes of the same frequency range, except for the relatively narrow frequency range of $110 oct$ where the five‐tone complexes were judged slightly noisier than the two‐tone complexes.