Spectral summation precedes temporal summation in loudness processing

Evidence

Because of the two kinds of integration, spectral and temporal, we have to know which one takes place first. Searching for an answer to this question in anatomy, it seems reasonable that spectral summation comes first because this distribution is available along the basilar membrane, where the stimuli of the sensory cells are transferred quickly into the corresponding neural excitation. Temporal summation seems to take place in more central areas, a fact which has been also clarified for postmasking using otoacoustic emissions. The results of such experiments confirm that the decay of the cochlear excitation is much quicker than the decay of postmasking, which may be assumed to be a counterpart of loudness. Therefore it was assumed that temporal summation comes second and spectral summation first. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1296

Temporal summation is another factor in the development of loudness. Although spectral summation produces larger effects in everyday sounds, the temporal effect is also quite important. The model can be extended so that temporal effects are taken into account. In doing so, the excitation level and specific loudness have to be treated as time-dependent values. Excitation is correlated with masking, therefore the masked thresholds produced in premasking, simultaneous masking and postmasking can be used and transformed into time functions for specific loudness. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1289

The approximation of the temporal effects in the model represents a kind of incomplete temporal integration. Corresponding to the cut-off frequency of the special low-pass system, the integration time lasts roughly about 45 ms. The spectral distribution of the sound leads also to an integration, namely the summation along critical-band rate. Because of the two kinds of integration, spectral and temporal, we have to know which one takes place first. Searching for an answer to this question in anatomy, it seems reasonable that spectral summation comes first because this distribution is available along the basilar membrane, where the stimuli of the sensory cells are transferred quickly into the corresponding neural excitation. Temporal summation seems to take place in more central areas, a fact which has been also clarified for postmasking using otoacoustic emissions. … Therefore it was assumed that temporal summation comes second and spectral summation first. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1296

Categorization

• XREF: The cochlear-vs-central place-of-processing distinction ties into the broader peripheral-to-central processing hierarchy one meets in psychoacoustics and neuroscience reading.
• DEFINE: Clarifies the ordering of two integration mechanisms (spectral at the basilar membrane, temporal more centrally) that a loudness model must respect.
• corroborated: 5 independent passages support this claim — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1296; Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1289