Combined maskers interact predictably in simultaneous but not nonsimultaneous masking

Evidence

In many cases, there is not one masker (for example a sinusoidal tone at a lower frequency) but also a second masker which may, for example, be a narrow-band noise centred at 1 kHz. At lower frequencies, masking will be defined by the low-frequency masker. For frequencies around 1 kHz, masking will be defined by the narrow-band noise. There is a region, however, for which the masking effect produced by the low-frequency tone and the masking effect produced by the narrow-band noise, each of them presented separately, produce an equal shift of threshold. This may be, in the case mentioned, in the neighbourhood of 800 Hz. The effect expressed as “addition of masking” can be formulated by a question: does threshold masked by the two noises remain the same as threshold masked by one sound alone, or does it rise by 3 dB (intensity addition) or 6 dB (sound pressure addition), or does it rise by some other value? Results of corresponding measurements using simultaneous masking can be summarized relatively clearly. Similar measurements using nonsimultaneous masking, for example post-masking, produce effects that cannot be generalized easily. It seems that bandwidth plays an important role with a number of other variables if temporal parameters are involved. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 617

Categorization

• QUESTION: The 3 vs 6 dB question for masker addition is a concrete empirical crux worth following into the cited measurements.