One of your questions is easy to answer. Even listeners with normal hearing show decreased intelligibility at constant SNR as overall level is raised. Fletcher and his group saw this in their work, and Studebaker et al. (1999) provide a more recent set of data on this. I’ve seen different arguments for why this might be the case but I’m not sure it’s completely understood. Arguments would include processes such as changing filter bandwidth, changing compression amounts, and changing saturation of nerve fibers as level is changed. (Judy Dubno’s group has also been looking at this more recently.)
As Pierre mentioned in an earlier post, Judy Dubno and Larry Humes have done a great deal of work in the area of understanding speech performance with aging and hearing loss, and they have a very nice review chapter in a 2010 Springer book, “The Aging Auditory System”. I’d say their chapter is a great place to start if you want to get a good overview of the current state.
One main issue Dubno and Humes (2010) point out is that basic audibility accounting (using articulation index) goes a long way in capturing a large portion of the variance of speech performance across listeners. From this perspective, one way in which the phone test results may yield correlation with absolute thresholds is if the noise used is not above absolute threshold in some important speech frequency region (I assume the noise level is constant and the speech level is adapted in the phone test, as in the Dutch version). When this is the case, absolute threshold itself, and not the noise, will be contributing to the speech level required for 50% correct, thus increasing correlation with pure tone average (PTA) directly.
Audibility alone doesn’t always capture all the variance, however, and there is ongoing research looking to understand the influence of other contributors, such as reduced inner hair cell and/or auditory nerve fiber count, which may manifest as poor performance on basic, supra-threshold auditory tasks such as frequency modulation detection (see, e.g., Strelcyk and Dau, 2009). Note that these other contributors may affect speech performance in quiet as well as in noise (see, e.g., our 2013 paper on a related issue).
In the end, the health of many of the processes contributing to speech intelligibility performance in quiet and in noise may all be correlated with age to some degree, and thus also correlated with each other. It would be interesting to hear from the Phone Test folks how much variance in predicted PTA is reduced (or how much categorization into their three categories is improved) when the phone test results are added to predictions based on age alone.
Starkey Hearing Research Center
Humes, L. E., and Dubno, J. R. (2010). “Factors affecting speech understanding in older adults,” in The Aging Auditory System, edited by S. Gordon-Salant, R. D. Frisina, A. N. Popper, and R. R. Fay (Springer, New York), pp. 211–257.
Strelcyk, O., and Dau, T. (2009). “Relations between frequency selectivity, temporal fine-structure processing, and speech reception in impaired hearing,” J. Acoust. Soc. Am. 125, 3328–3345.
Studebaker, G. A., Sherbecoe, R. L., McDaniel, D. M., and Gwaltney, C. A. (1999). “Monosyllabic word recognition at higher-than-normal speech and noise levels,” J. Acoust. Soc. Am. 105, 2431–2444.
Woods, W., Kalluri S., Pentony S., and Nooraei, N. (2013). “Predicting the effect of hearing loss and audibility on amplified speech reception in a multi-talker listening scenario,” J. Acoust. Soc. Am. 133, 4268–4278.
From: AUDITORY - Research in Auditory Perception [mailto:AUDITORY@xxxxxxxxxxxxxxx]
On Behalf Of Richard F. Lyon
I took the test just now and got the expected result: "slightly below normal range" in both ears (expected, based on previous tests that show somewhat worse than typical hearing for my 61-year age).
It would be great if there was better quantification of the result, like dB of SNR degradation relative to normal, rather than just "slightly below normal range". Charles, is there a numeric range associated with these words?
It does seem like a pretty effective test of hearing in noise. I presume it's doing an adaptive SNR process, since it would usually give me a relatively easy one after each one where I was pretty much guessing.
What I'd like to understand better is exactly how the mechanisms causing threshold elevation also cause degradation of SNR threshold. I presume that the auditory filter bandwidths are wider, and the compression
less, with hearing loss. But they're also wider at high levels, and high levels don't cause a degraded SNR threshold, do they? Or maybe they do, in normal hearing, at levels high enough to cause this much bandwidth widening?
On Wed, Oct 30, 2013 at 9:02 AM, Watson, Charles S. <watson@xxxxxxxxxxx> wrote:
Sent: Wednesday, October 30, 2013 6:01 AM
Actually, my comments about telephone service are based on having lived in Germany for nearly five years and having a number of friends in Europe. I do believe their telephone system superior to ours. This is especially true of their
cell phone system but is to a lesser degree of their land lines.