Backward masking: central or peripheral?

Backward masking is the masking of a signal, called probe (P), by a masker (M) that occurs later. An extensive review of this phenomenon has been given by Raab (1963). Auditory backward masking is most frequently attributed to time-dependent properties of the neural mechanism of hearing (see, e.g.,Jeffress, 1970). Jeffress states that the intervals over which backward masking occurs (<25 ms) are too long to explain in terms of energy integration in the cochlea. This statement is probably based on an expected magnitude of the decrement of the cochlear response. As recent results (e.g., Johnstone et al., 1970; Duifhuis, 1971) suggest that the frequency resolving power of the cochlea is much better than might be expected from the classical results of von Bekesy (1949), we feel that this argument deserved renewed attention.