This study examined what types of subjects may benefit from the use of
multiple memory hearing aids that offer variation in their frequency
response characteristics. Thirty subjects with varied degrees and conf
igurations of hearing loss compared an individually prescribed frequen
cy response (NAL) and two variations in which the real-ear response sl
ope was either increased (more high-frequency emphasis) or decreased b
y about 3 dB/octave over the range from 500 Hz to 4000 Hz. The evaluat
ions consisted of paired comparison judgments of pleasantness and of e
ase of understanding speech, in quiet and in three background noises w
ith substantially different acoustic spectra (traffic noise, speech ba
bble, and high-frequency noise). Twelve of the 30 subjects selected di
fferent frequency responses for different background noises and/or for
different response criteria. These subjects were characterized as hav
ing the more severe high-frequency hearing losses (averaged across 200
0, 3000, and 4000 Hz). They also tended to be those for whom the three
frequency responses provided the greatest variation in real-ear low-f
requency gain. (Despite substantial electronic variations, some subjec
ts received only small variations in real-ear gain at the low frequenc
ies). When subjects chose a different response for different condition
s, they tended to prefer responses for which the slope was negatively
related to the spectrum of the stimulus. This was true for both respon
se criteria. The conclusion is that hearing aid users with substantial
high-frequency losses, and who can be fitted with sufficient variatio
n in the low-frequency real-ear gain, have the potential to benefit fr
om having a choice of frequency response characteristics.