An implied or stated goal of some hearing aid processing strategies is to compensate for the lost compressive function of the cochlea resulting from sensorineural hearing loss. I am in the process of writing a brief paper (as part of my doctoral coursework) discussing the ability of current devices to actually do this and the pros and cons of even attempting to do so. Initially, I am trying to describe the "compression parameters" of the normal cochlea. I have found substantial physiologic work in various animals suggesting linear function below an initial "compression threshold" of about 20-40 dB SPL and some psychophysical work in humans suggesting an initial "compression threshold" of 20 dB SPL or less. For midlevel inputs (40-90 to 110 dB SPL or so) a compression ratio of about 5:1 (7:1 to 3:1) has been suggested. For high levels (above 90 to 110 dB) some research suggests linear functioning but I am unclear on which references support this idea. I have not found any references that discuss potential "time constants" for the normal cochlea. I believe some physiologic work on outer hair cells has suggested they can function essentially instantaneously.
In short, I am interested in references that may discuss the compressive nonlinearity of the cochlea in terms of "compression characteristics". In specific, references for a second high level compression threshold and a discussion of "cochlear time constants". Any thoughts or ideas are appreciated.