The role of spread excitation and suppression in simultaneous masking

This experiment was intended to clarify the relative role of spread of excitation and suppression in simultaneous masking, for masker frequencies just below and well below the signal frequency. The experiment had two stages. In stage 1, growth-of-masking functions were measured in simultaneous masking for a 2200-Hz sinusoidal signal and a sinusoidal masker with frequency of either 1800 Hz or 500 Hz. Straight lines fitted to these data were used to determine masker levels that would give 10, 20, and 30 dB of masking. In stage 2, thresholds for detecting a brief 2200-Hz signal were measured using forward masking. It was reasoned that the threshold of the signal would give an indication of the amount of excitation evoked by the masker in the frequency region of the signal. Three forward maskers were used: (1) a 2200-Hz sinusoid at 10, 20, or 30 dB sensation level (SL); (2) a 2200-Hz sinusoid at the same levels as in (1) together with a sinusoid with frequency 500 or 1800 Hz at a level just sufficient to mask the 2200-Hz sinusoid. We refer to this as the "combined masker," (3) a 500-Hz or 1800-Hz sinusoid at the same levels as in (2) above. The 1800-Hz combined masker produced slightly less forward masking than the 2200-Hz masker (1), which might be explained in terms of suppression or as perceptual cueing. Both the 1800-Hz combined masker and the 1800-Hz component alone (3) gave significant amounts of forward masking (up to 18 dB), indicating that these maskers produced substantial excitation at 2200 Hz. This is consistent with the idea that the simultaneous masking of the 2200-Hz component in stage 1 was produced by spread of excitation rather than by suppression. The 500-Hz combined masker produced much less forward masking than the 2200-Hz component alone, indicating strong suppression of the 2200-Hz component of the combined masker by the 500-Hz component. However, both the 500-Hz combined masker and the 500-Hz component alone produced some forward masking. This is not consistent with the idea that masking of the 2200-Hz component in stage 1 (simultaneous masking) was produced solely by suppression.