I found some online literature regarding an audiometric study on young hearing with iPod music here. That study suggests that, at least for young healthy hearing, a nearly full exposure dose produces a depressed hearing, most prominent in the 4-8 kHz range, of around 6 dB. That produces a Temporary Threshold Shift (TTS), that recovers after you let your ears relax in quiet for several hours. (… but look at the size of their error bars… )
Their graphs indicate a more or less constant 6 dB shift at all presentation levels, which I find quite odd, given what I know about hearing now. I’ll have to think about how to reconcile our differences. That seems like an induced decruitment condition.
We already know that, after listening at loud levels, you feel like your head is under water. Sounds are muffled. That implies a loss of high frequency response that recovers gradually after ceasing exposure.
But, that is some guidance. But what struck me about that article was that they were using Etymotics insert headphones and the SPL levels were out of sight! 94 dBA SPL. I wouldn’t dream of hitting my ears with that as the o dBVU level in my playback.
But they also point out, if you read between the lines, that the way they measure the SPL in the ear canal would imply significantly lower free-field SPL, perhaps down around 84 dB SPL.
It is interesting to contemplate that disparity between free-field (room level) versus ear canal levels.
From a physics standpoint, we know the ear canal acts as a quarter-wave resonator at around 4 kHz, with one end closed at the eardrum. That means that the pressure will be highest right at the eardrum, and has to match the room free-field pressure at the opening of the ear canal. Boundary conditions.
That resonance at 4 kHz is what causes the apparent slight increase in hearing sensitivity in that frequency region. If you look at Fletcher-Munson equal loudness contours, they all show a dip in that region.
So how to find the pressure increase at the eardrum? Well, that depends on the acoustic efficiency or Q of the ear canal as a tuned cavity. I can only guess that the Q is around 3, since they report 10 dB difference from room free field to tympanic membrane.
But those young, healthy, ears are listening at sustained levels of 85 dBSPL (free field equivalent) for 4 hours solid. Personally, I wouldn’t want to subject my own hearing to those kinds of levels for that long. Perhaps young ears are more resilient?
[ Sorry… I misinterpreted their graphs a bit. What seems “clear” (?) is that there is a threshold shift in the treble range, most strong at about 6 dB around 4 kHz. (again… consider those error bars…)
The DPOAE graphs show the resulting otoacoustic emissions being lower at all presentation levels of the primary tone, for IMD products of 2F1-F2. This is a decrease in IMD distortion products (a good thing?). Not sure how IMD OAE’s figure into my theory, if at all… To my mind, that states that the gamma term in the stiffness of the EarSpring equation diminishes a bit, meaning less resistance to loud sounds. Not a good thing… and could lead to further damage if you persist with loud sounds. The ear isn’t protecting itself as well anymore.
So, on the basis of that, we can expect to see a 6 dB threshold shift, or vTuning increase, needed after a full dose of audio.
Frankly, after you have a full dose of audio, you should put your audio toys away for the rest of the night, and come back tomorrow refreshed. ]