Marshall Chasin, Back to Basics (previously published in Hearing Review)
Whenever I start to read about the impedance of structures, my eyes tend to glaze over. I understand that impedance is an important part of audiology and I certainly remember the formulae that we needed to learn in school which explains why stiffness and mass have different effects, but other than our clinical admittance/impedance tests that we perform on each client, that’s typically where my interest stops—and I have a degree in mathematics!
Below you will find a great picture which helps me, at least in part, understand what the engineers are talking about: the eardrum is a brick wall for higher frequencies. Or stated differently, low frequency sound does not see the eardrum. It’s one of my favorite pictures (right up there with my wedding picture, picture of my kids, and my cats Broccoli and Cauliflower).
We can instantly remember (without having to resort to formulae) that higher frequency sounds “see” the head as an obstruction which is one reason why binaural hearing aids are so useful; low frequencies go around the head, whereas a head shadow exists for the higher frequency (shorter wavelength) sounds. The head is a brick wall for higher frequency sounds.
The same is true of the ear canal. High frequency sounds do not see the tympanic membrane (and middle ear structures) as easily as do the lower frequencies. For lower frequency sounds the outer ear, tympanic membrane, and middle ear structures all contribute to the “equivalent volume” generated in an occluded outer ear. In contrast, for higher frequency sounds, only the outer ear contributes to the “equivalent volume” such that a constant volume ear coupler/simulator will generate a higher level output for higher frequencies (Boyle’s law) than for lower frequencies. That is, the ear canal behaves as if it is a smaller volume for higher frequency sounds.
This has caused some in the standards side of the field of audiology to consider a smaller volume ear canal simulator for higher frequency sounds (0.4 cc) and a second larger volume ear canal simulator for lower frequency sounds (1.4 cc). The smaller volume has relevance for wider bandwidth measures as well (to avoid a 10,500 Hz anti-resonance found in a 2 cc couplter), but that’s another Back to Basics. Of course, we are not typically interested in ear canal simulators clinically- the 2 cc coupler (ANSI S3.7, 1995) that has been in use since before I was born- isn’t a simulator but is a coupler that can provide repeatable and reliable results.
The hearing aid test standard (ANSI S3.22, 2003) uses the 2 cc coupler and because it is not a simulator of the ear canal (i.e., it doesn’t take in to account the various impedances and resonances of the outer and middle ear system) real-ear to coupler differences (RECDs) need to be established for each client. But a coupler is just a coupler- we could have just as easily used a coke can or a 2.1 cc coupler. If our goal is to test a hearing aid according to some well-defined criteria then as long as we are consistent and use the same coupler that was used in the manufacturing facility, or the same coupler we used last year, then this is fine.
If we want to understand why real-ear coupler differences are not zero then the brick wall analogy becomes quite useful.