The positive effect of directional hearing aids for speech understanding in noise is well documented. The literature, however, also shows variations in the amount of directional benefit obtained among hearing-impaired individuals with very similar audiometric thresholds. The aim of this study was to examine whether the outcome of an aided speech-in-noise intelligibility task using hearing aids in omnidirectional mode can help predict the amount of directional hearing aid benefit. Results revealed no statistically significant correlation between the SRT in omnidirectional mode and directional benefit. Degree of hearing loss was found to influence omnidirectional and directional performance, but not directional benefit.
The inability to understand speech in noise is one of the most common complaints of hearing aid users who have a sensorineural hearing loss.1-3 Research has shown that hearing aid users may require signal-to-noise ratio (SNR) improvements of 2-18 dB, depending on the magnitude of the hearing loss, in order to achieve word recognition scores equivalent to normally hearing persons.4-9
A variety of techniques are currently available to help reduce the effects of background noise on speech understanding for hearing aid users with sensorineural hearing loss. These techniques include both clinical and specific circuitry designed techniques. Clinical techniques include binaural amplification,10-12 reduction of low-frequency amplification, and providing increased headroom for decreasing distortion at high sound levels (for a review, see Gordon-Salant13 and Schum12). User tone control, wider bandwidth, automatic low-frequency reduction, multi-memory hearing aids, compression, noise estimation, and multiple microphone arrays are some of the hearing aid circuit features which may assist speech understanding in background noise. Automatic speaker identification in which the speech of one talker is tracked and separated from the speech of other talkers in the same listening environment is also a potential future solution to this problem.12
Directional hearing aids provide an effective means of achieving better speech understanding in noisy near-field situations where signals of interest and noise are spatially separated.12,14-20 Data suggests that some hearing-impaired listeners receive markedly more directional benefit than others.21
From a clinical standpoint, it is relevant to know factors that are important for identifying hearing-impaired people who might, or might not, achieve benefit from directional hearing aids. Past investigations have suggested at least three factors that may influence benefit with directional hearing aids. These include
1) The slope of the audiometric configuration;
2) The degree of high-frequency hearing loss, and
3) Aided omnidirectional performance for a speech-in-noise intelligibility task.
Killion et al.21 suggested that the slope of the unaided audiogram and the degree of high-frequency hearing loss influence directional benefit. The authors proposed that hearing aid users with flat audiometric configurations should have less directional benefit than those with sloping hearing loss. This reasoning is based on the directional characteristics of hearing instruments and the reduced ability of some users with more severe high-frequency hearing losses to utilize high-frequency speech information.22-24 The Directivity Index (DI) improvement for the directional hearing aid setting, as compared to the omnidirectional setting, was reported by Killion et al.21 to be predominantly in the low frequencies. This finding is typical of many ITE and BTE directional hearing aids.25 Since the improvement in directivity between directional and omnidirectional conditions is greatest in the low frequencies, those hearing aid users who rely primarily on low-frequency speech information (ie, those with sloping hearing losses) will achieve greater directional benefit than those who are able to use speech information across the entire frequency range (those with less severe high-frequency loss).
Some research, including that by Killion et al.,21 has shown that hearing aid users with greater SNR loss (loss in ability to understand speech at the SNR used by those with normal hearing) appear to receive greater directional benefit. Although the differences in directional benefit measured by Killion et al.21 were later attributed to differences in threshold slope,26 it is still unclear whether this difference may have resulted from other unknown factors related to SNR loss.
Ricketts & Mueller27 examined whether the magnitude of directional hearing aid benefit could be predicted by selected audiologic factors. They reviewed results obtained in three separate investigations and compared these results to three different patient-specific factors: slope of audiometric configuration, degree of high frequency hearing loss, and aided omnidirectional performance for a speech-in-noise test. The results of this analysis revealed no significant correlation between the slope of audiometric configuration or amount of high-frequency hearing loss and the benefit obtained from directional hearing aids. There was, however, a significant inverse relationship between aided omnidirectional performance and the directional benefit obtained in one of the studies.
The presence of a significant SNR loss might indicate a higher susceptibility to the negative effects of noise. If this is the case, is it possible that increasing noise levels cause greater-than-normal decrements in performance, and increasing SNR through the use of directionality (ie, decreasing the noise level) may lead to greater-than-usual benefit?
Based on the hypothesis above and in the light of the results reported by Ricketts & Mueller,27 it is of interest to further examine the relationship between omnidirectional performance for a speech-in-noise test and the magnitude of directional benefit. Ricketts & Mueller studied the relationship between omnidirectional performance for a speech-in-noise test and the magnitude of directional benefit by reviewing results obtained in three different studies. The original aim of these studies was not to investigate whether there was a relationship between speech-in-noise performance in omnidirectional and directional benefit among hearing aid users, but rather to examine the influence of different factors on directional benefit. Among the factors tested in the three studies were different listening environments, different hearing aid models (BTE/ITE), different hearing aid types (digital/programmable/ nonprogrammable), hearing aid settings (linear/non-linear; one-/more channels, different compression thresholds and compression types and different directionality systems).
In contrast, the goal of the present study was to examine if there was a relationship between speech-in-noise performance in omnidirectional mode and directional benefit. The study was designed only for examining the relation between speech-in-noise performance in omnidirectional mode and directional benefit (excluding differences in test environment, and hearing aid models, types, and settings).
Ricketts & Muellers results showed a considerable variability among the individual participants. The hearing loss of the participants also varied to some extent. It is therefore relevant to clarify the interactions between degree of hearing loss, omnidirectional performance for a speech-in-noise test, and the magnitude of directional benefit.
The objective of the present study was to examine whether the outcome of an aided speech-in-noise recognition task using hearing aids in omnidirectional mode could be used to predict the amount of directional hearing aid benefit, if slope of hearing loss was controlled for. The effect of degree of hearing loss on the outcome was also studied.
Figure 1. Median hearing thresholds for the three subject loss groups. These median thresholds represent the combined thresholds of both left and right ears.
The study was conducted under laboratory conditions, and the participants did not have a real-world trial period with directional hearing aids prior to the speech testing. The results from this trial are, for this reason, indicative of initial directional benefit.
Participants: Thirty-two hearing-impaired adults (20 male, 12 female) participated in the study. Subjects ranged in age from 28-81 years old (median age = 68). The participants all exhibited symmetric moderately sloping sensorineural hearing impairment. Moderately sloping hearing loss was defined as <30 dB and <15 dB per octave on average from 250 Hz-4000 Hz. Symmetry between ears was defined as exhibiting no more than a 10 dB difference in pure-tone thresholds at any octave frequency from 250 Hz-4000 Hz. Sensorineural hearing loss was verified by normal (type A) tympanograms,28 by the presence of normal contralateral acoustic reflexes for a 1000 Hz tone, and by a positive Bing test. All listeners also exhibited unaided monosyllabic word recognition ability in quiet better than 50% in each ear under earphones at a comfortable level, as measured using the Dantale speech material.29
Participants were divided into three groups according to degree of hearing loss for the pure tone average of 500 Hz, 1000 Hz, and 2000 Hz. Mild hearing loss was defined as between 26 dB-40 dB HL, moderate hearing loss between 41 dB-55 dB HL, and moderate-to-severe hearing loss between 56 dB-90 dB HL.30 None of the moderate-to-severe hearing losses exceeded an average of 78 dB HL in order to keep the hearing losses within the fitting range of the hearing aid. When difference in classification of the two ears existed, the hearing loss was classified according to the better ear. Twelve participants with mild hearing loss, 10 participants with moderate hearing loss, and 10 participants with moderate-to-severe hearing loss participated in the study. Figure 1 shows the median hearing thresholds (for both left and right ears) for the three hearing loss groups.
All participants were experienced with amplification, using binaurally fitted behind-the-ear (BTE) hearing aids for at least 6 month (range: 6 months-37 years). None of the participants had previously worn a directional hearing aid.
Hearing Aids: The hearing instrument used in this study was the GN ReSound Canta 770-D BTE, a fully digital hearing aid with adjustable compression ratios in 14 overlapping bands and 4 memories (programming configurations). The syllabic compression thresholds are <45 dB SPL, with compression ratios of up to 3.0 (program dependent in each band).
|125 Hz||1.3 s|
|250 Hz||1.3 s|
|500 Hz||1.0 s|
|1 kHz||0.3 s|
|2 kHz||0.3 s|
|4 kHz||0.3 s|
|8 kHz||0.3 s|
Table 1. Reverberation time (time required for 60 dB decay after signal offset) in octave bands based on interrupted white noise. Average reverberation time was 0.7 seconds.
The attack time is <5 msec, and the release time is 70 msec (120 msec at 250 Hz). The maximum power output limiting threshold is adjustable between 0 dB and -9 dB relative to maximum power output (program dependent in each band) with a >15.0 compression ratio and <5 msec attack and 70 msec release times.
Directivity/Signal Processing: In addition to the described wide dynamic range compression (WDRC), Canta 770-D includes optional microphone directionality. Directionality is achieved electronically through a dual-microphone system and is selectable among three fixed polar patterns (hypercardioid, cardioid, and bi-directional) plus an adaptive pattern. The adaptive directionality system seeks to change the directional characteristics during use of the hearing aid to maintain optimum directional characteristics by adjusting the node (or greatest region of attenuation) in the polar pattern to point towards the strongest noise source in the rear hemisphere. If more competing noise sources exist, the node is designed to be placed where maximum noise attenuation can be achieved.
In accordance with the manufacturers recommendation, the adaptive polar pattern was used in this study. The fitter can compensate for the normal 6 dB per octave low-frequency roll-off associated with directionality by adding low-frequency gain. The fitting options are no bass boost, normal bass boost (3 dB), and maximum bass boost (6 dB). In this study a 3 dB boost (normal boost) in the lower frequencies was added consistent with the manufacturers recommended setting.
The instrument also has a noise reduction system, a digital feedback suppression circuit, and a spectral enhancement algorithm, but these were not engaged for any of the participants in this trial, as it was not part of the objective for this study.
Fitting: The hearing instruments were fitted binaurally and according to the manufacturers threshold-based prescription rule for users experienced with amplification, Audiogram+. The fitting was fine-tuned according to the participants preferences. The hearing aids were fitted using the Aventa version 1.1 software. The aids were coupled to the ears with custom, acrylic full-shell earmolds, standard #13 tubing, and appropriate venting (typically, 1 mm or 1.5 mm).
Figure 2. Mean directional benefit (in dB) for the three loss groups. The median directional benefit is statistically significant (p<0.05) in all three groups.
Test Conditions: The study design included two test conditions. The hearing aids were programmed with two programming configurations. Program 1 was an omnidirectional microphone program. Program 2 was a directional program (adaptive pattern). The test order of the two programs was counterbalanced across the participants in each of the three above-mentioned hearing loss groups to minimize order effects.
Speech Recognition Testing: The Dantale II sentence test31 was used to determine the SRT for each test condition. In this article, SRT is defined as the signal-to-noise ratio necessary for 50% correct performance. Speech-shaped unmodulated masking noise at 65 dB SPL served as the competing signal. Wagener et al.31 reported a standard deviation of 1.75 dB in their normative study of this speech-in-noise test.
The Dantale II Hagerman sentences and masking noise were presented from a Philips CD 930 compact disc player. Participants were seated in the center of the listening environment. The speech material was presented at 0° azimuth from a Dali (102) 2-way bass reflex loudspeaker. Three noise sources (mono noise source delayed 0 ms, 400 ms, and 800 ms) were placed at 90°; 180°, and 270° azimuth and presented from three Dali (102) 2-way bass reflex loudspeakers. All sound sources were placed at a distance of 1.25 m from the participants.
The three noise speakers were calibrated so that they all had the same level at the center of the participants head without the head present (ISO 8253-2), which was -4.8 dB relative to the speech source at 0 dB SNR. The SPL produced by the three sources playing at the same time was power summated (the noise sources can be considered as uncorrelated sources because of the delay) and produced 0 dB relative to the speech source.
The listening environment was a carpeted room with sound treated panels on three walls. The room measured 6.30 meters (length) x 5.63 meters (width) x 2.41 meters (height). Reverberation time (time required for 60 dB decay after signal offset) was measured as an average of three spatial positions without the subject (the position of the participants head included). The test stimulus used was abruptly gated white noise (Table 1). The average reverberation time was 0.7 seconds.
The outcome of the test was analyzed using the ANOVA Single Factor analysis.32 Statistical significance was defined at the 0.05 level. Figure 2 shows the mean directional benefit in dB for the three hearing loss groups. The mean directional benefit for the hearing loss group with mild hearing losses was 3 dB (range: 2 dB-4 dB). The directional benefit for this group was statistically significant (P<0.01). The participants with moderate hearing losses obtained a mean directional benefit of 4 dB (range: -1 dB-6 dB). This directional benefit was also statistically significant (p<0.01). The mean directional benefit for the moderate-to-severe hearing loss group was 2 dB (range: 0 dB-4 dB). The directional benefit for this group also reached statistical significance (p<0.01).
Figure 3. Relation between PTA and SNR in omnidirectional mode. The correlation is 0.49 which is statistically significant (p<0.01).
An ANOVA Single Factor analysis32 was made in order to see whether there was a statistical significant difference in the benefit achieved by the three hearing loss groups. Statistical significance was defined at the 0.05 level. A statistical significant difference between the groups was not found at the 0.05 level, but at the 0.25 level. The difference in benefit obtained by the three hearing loss groups was, in other words, not statistically significant.
Figure 4. Relation between PTA and SNR in directional mode. The correlation is 0.56 which is statistically significant (p<0.001).
The relationship between PTA and SRT in omnidirectional mode is depicted in Figure 3. The correlation was 0.49, which is statistically significant (p<0.01). The relationship between PTA and SRT in directional mode is depicted in Figure 4. The correlation was 0.56, which is statistically significant (p<0.001).
The relationship between PTA and directional benefit is depicted in Figure 5. The correlation was -0.30, which is not statistically significant (p>0.05).
The correlation between SRT in omnidirectional mode and directional benefit for the mild hearing loss group was 0.50; for the moderate hearing loss group it was 0.16; and for the moderate-to-severe hearing loss group it was 0.08. The correlation between SRT in omnidirectional mode and directional benefit for all hearing loss groups is depicted in Figure 6. The correlation was -0.04 and not statistically significant (p>0.05).
Degree of directional benefit: The directional benefit obtained in this study was statistically significant (p<0.01) and comparable to directional benefit obtained in other studies measuring directional benefit in more realistic environments.4,6,33,34
The difference in mean benefit between the three hearing loss groups was, however, not statistically significant (p>0.05). A problem in comparing the mean directional benefit in the three hearing loss groups is that the audiogram slopes in the three hearing loss groups were not fully matched. The mean audiogram slope of the moderate and moderate-to-severe hearing loss group was nearly identical, whereas the mean audiogram slope of the participants with mild hearing losses was steeper than the mean slope in the two other hearing loss groups. This resulted in the mild hearing loss group having a larger mean high-frequency hearing loss than the moderate hearing loss group.
The steeper audiogram slope in the mild hearing loss group is perhaps not so strange, as people with this degree of hearing loss probably wouldnt experience hearing problems great enough to seek help with a degree of slope seen in the two other groups. The mild hearing loss group showed a tendency (although not statistically significant) to obtain a smaller mean directional benefit as compared to the moderate hearing loss group.
The larger mean high-frequency hearing loss of the mild hearing loss group might indicate inner hair cell loss.35 Total loss of inner hair cells in a certain region might prevent extraction of useful speech sound information from that region.36,37 Reducing background noise using directional microphone systems cannot solve this problem, which might explain, or partially explain, the tendency of the mild hearing loss group to obtain less directional benefit.
The moderate-to-severe hearing loss group, like the mild hearing loss group, showed a tendency (again, not statistically significant) to obtain a smaller mean directional benefit as compared to the moderate hearing loss group. More severe hearing losses often include inner hair cell loss.35 The presumed greater involvement of inner hair cell damage in this hearing loss group is believed to be the reason or part of the reason that this group showed a tendency for obtaining less directional benefit.
Data from previous studies suggest that some hearing-impaired listeners receive markedly more directional benefit than others.21 The results above show that the three hearing loss groups obtained a mean directional benefit that isnt statistically different between the groups; in other words, they obtained the same amount of benefit.
Degree of hearing loss: It was also of interest to examine whether there was a relation between degree of hearing loss and SRT in omnidirectional mode, SRT in directional mode, and directional benefit.
Figure 5. Relation between PTA and directional benefit. The correlation is -0.30, which is not statistically significant (p>0.05).
There was a statistically significant (p<0.01) correlation (0.49) between PTA and SRT in omnidirectional mode indicating that greater hearing loss requires better SNR. These results are in accordance with the results found by Killion,4,6 Moore,7 Peters, Moore & Baer,8 and Plomp.9
Figure 6. Relation between SNR in omnidirectional mode and directional benefit. The correlation is -0.04 which is not statistically significant (p>0.05).
The correlation between PTA and SRT in directional mode was 0.56, which is statistically significant (p<0.001). This means that, on average, a greater SNR is needed to achieve 50% correct performance on the speech task as greater hearing losses are encountered.
The correlation between PTA and directional benefit was -0.30, which isnt statistically significant (p>0.05). These results indicate that directional benefit is not affected by degree of hearing loss, and that hearing-impaired participants should enjoy the same degree of directional benefit regardless of degree of hearing loss. This result supports the finding of no statistically significant difference between the benefit obtained by the three hearing loss groups described above. So, the result was as expected: a directional hearing aid giving a certain amount of improvement in SNR should give this improvement no matter the degree of hearing loss of the subject.
Dirks, Morgan and Dubno38 found that there is a greater variability in the ability to understand speech in noisy surroundings among people with sensorineural hearing loss, as compared to intersubject variability among normal-hearing people. This is not surprising as hearing-impaired people have different degrees of hearing loss while normal-hearing people have more similar audiometric thresholds.
The results in this study suggest that the degree of performance in both omnidirectional and directional mode vary with degree of hearing loss. It is possible that differences in suprathreshold processing abilities are a factor in speech recognition, as well as audibility. A variety of explanations for the increased difficulty in understanding speech in noise surroundings have been advanced. It is likely that, for any given person with sensorineural hearing loss, a combination of factors is at play. These may include decreased audibility, decreased signal-to-noise ratio due to reduced ability to hear binaurally (also called the squelch effect), upward spread of masking, and decreased discrimination between the timing of a wanted speech signal and unwanted speech signals (also called temporal smearing; see Schum12 for a discussion).
Prediction of directional benefit based on SNR in omnidirectional mode: It was of interest to examine whether the degree of directional benefit could be predicted from the SRT with an omnidirectional response. The correlation between SRT in omnidirectional mode and directional benefit (-0.04) was, however, not statistically significant. In other words, it is not possible to predict directional benefit from the SRT in omnidirectional mode.
The results in this study show that individuals with hearing-impairmentindependent of the degree of hearing loss degree and performance in omnidirectional modegain the same amount of directional benefit. It is not possible to predict the amount of directional benefit from the hearing threshold or SRT hearing loss.
Prediction of directional benefit from audiological factors appears to be difficult using todays commonly used audiologic test methods. It is likely that predictability of directional benefit is to be found from the characteristics of the listening situations encountered by the hearing-impaired, rather than from audiological factors. Degree of directional benefit/performance is influenced by factors such as placement of the speech signal and placement and number of competing noise sources, as well as the distance from the speech signal to the hearing aid microphone. In addition, reverberation time also influences directional benefit/performance, as do the type of noise and the sound pressure level of the noise sources.18,33,39-43
For this reason it is most likely that the amount of directional benefit that a hearing aid user will receive varies with the characteristics of his/her listening environment. A hearing-impaired individual who receives, for example, 4 dB of SNR advantage in directional mode (as compared to in omnidirectional mode) is likely to get varying degrees of benefit in real life depending on the specific characteristics of the listening environment. Directionality might well provide the hearing aid user with a substantial directional benefit in a challenging listening situation, like in a noisy classroom. But, when at home talking around the dinner table, the same individual may experience less directional benefit. In this case, he/she might already be performing near the optimal level in omnidirectional mode, thus limiting the amount of potential directional benefit in this situation.
The superiority of directional microphone technology for improved speech understanding in noise in controlled laboratory settings has been demonstrated clearly.19,44,45 However, this dramatic directional microphone advantage has not been observed in field measures of perceived benefit in everyday life as reported by Walden et al.46 Results from a study made by Cord et al.47 suggest that the disparity between laboratory and field measures, to a large extent, can be explained by the specific characteristics of the listening situations encountered in daily living. In other words, directional technology is an exceptional tool for increasing the SNR of hearing aid users who are frequently confronted with challenging listening environments, but there is no way to predict or guarantee across-the-board benefit (in dBs SNR) for any or all listening situations using standard tests in a soundbooth.
- Speech recognition of participants using the directional program was significantly better than when using the omnidirectional program. The magnitude of this directional advantage was in agreement with past investigations conducted in difficult (multiple source, moderate reverberation), near-field listening environments.
- SNR in omnidirectional and directional mode is positively correlated with the PTA.
- PTA cannot predict directional benefit.
- SNR in omnidirectional mode cannot predict directional benefit.
Preliminary results from this study were presented at the American Academy of Audiology Convention held in Philadelphia during April 2002, and at the Nordic Audiological Society Congress held in Helsinki, Finland, during May 2002.
|Charlotte Thunberg Jespersen, MA, and Steen Østergaard Olsen are research audiologists in the department of clinical audiology at GN ReSound AS in Taastrup, Denmark.|
1. Kochkin S. MarkeTrak III: Identifies key factors in determining consumer satisfaction. Hear Jour. 1992;45(8):39-44.
2. Kochkin S. MarkeTrak III: Why 20 million in US dont use hearing aids for their hearing loss. Hear Jour. 1993;46, (1):20-27; 46(2):26-31; 46(4):36-37.
3. Kochkin S. "Why my hearing aids are in the drawer": The consumers perspective. Hear Jour. 2000;53, (2), 34-42.
4. Killion MC. The SIN report: Circuits havent solved the hearing-in-noise problem. Hear Jour. 1997;50, (10):28-32.
5. Killion, M.C. SNR loss: I can hear what people say, but I cant understand them. Hearing Review. 1997;4(12): 8-14.
6. Killion MC. Hearing aids: Past, present, future: Moving toward normal conversations in noise. Brit Jour Audiol. 1997;31:141-148.
7. Moore BCJ. Perceptual Consequences of Cochlear Damage. Oxford: Oxford University Press; 1995.
8. Peters RW, Moore BCJ, Baer T. Speech reception thresholds in noise with and without spectral and temporal dips for hearing-impaired and normally hearing people. J Acoust Soc Am. 1998;103:577-587.
9. Plomp R. A signal-to-noise ratio model for the speech-reception threshold of the hearing impaired. J Speech and Hear Res. 1986;29:146-154.
10. Agnew J. Optimizing binaural cues for speech understanding. Hearing Review. 2000;7(5):20-26,52.
11. Kochkin S, Kuk F. The binaural advantage: Evidence from subjective benefit & customer satisfaction data. Hearing Review. 1997;4(4):29-34.
12. Schum D. Speech understanding in background noise. In: M.Valente, ed. Hearing Aids: Standards, Options, And Limitations. New York: Thieme; 1996: 368-406.
13. Gordon-Salant S. Effects of reducing low-frequency amplification on consonant perception in quiet and noise. J Speech Hear Res. 1984;27:483-493.
14. Agnew J, Block M. HINT thresholds for dual-microphone BTE. Hearing Review.1997;4 (9):26-30.
15. Agnew J. Challenges and some solutions for understanding speech in noise. In: Kochkin S, Strom KE, eds. High Performance Hearing Solutions, Vol 3. Hearing Review. 1999; [suppl] 6 (1): 4-9.
16. Mueller HG, Ricketts TA. Directional-microphone hearing aids: An update. Hear Jour. 2000; 53(5):10-19.
17. Preves D. Directional microphone use in ITE hearing instruments. Hearing Review. 1997;4(7):21-27.
18. Ricketts TA, Dhar S. Comparison of performance across three directional hearing aids. J Am Acad Audiol. 1999; 10(4):180-189.
19. Valente M, Fabry D, Potts L. Recognition of speech in noise with hearing aids using dual microphones. Jour Amer Acad Audiol. 1995;6:440-449.
20. Voss T. Clinical evaluation of multi-microphone hearing instruments. Hearing Review. 1997;4(9):36, 45, 46, 74.
21. Killion M, Schulien R, Christensen L, Fabry D, Revit L, Niquette P, Chung K. Real-world performance of an ITE directional microphone. Hear Jour. 1998;51(4):24-38.
22. Rankovic CMR. An application of the articulation index to hearing aid fitting. J Speech Hear Res. 1991;34:391-402.
23. Skinner MW. Speech intelligibility in noise-induced hearing loss: Effects of high-frequency compensation. J Acoust Soc Am. 1980;67:306-317.
24. Turner CW, Cummings KJ. Speech audibility for listeners with high-frequency hearing loss. Am J Audiol. 1999;8:47-56.
25. Ricketts T. Directivity quantification in hearing aids: Fitting and measurement effects. Ear Hear. 2000;21:45-58.
26. Killion MC, Christensen LA. The case of the missing dots: AI and SNR loss. Hear Jour. 1998;51(5):32-47.
27. Ricketts T, Mueller HG. Predicting directional hearing aid benefit for individual listeners. Jour Am Acad Audiol. 2000;11:561-569.
28. Jerger JF. Clinical experience with impedance audiometry. Arch Otolaryngol. 1970;92:311-324.
29. Elberling C, Ludvigsen C, Lyregaard PE. Dantale: A new Danish speech material. Scand Audiol. 1989;18:69-175.
30. Yantis PA. Puretone air-conduction threshold testing. In: Katz J, Gabbay WL, Gold S, Medwetsky L, Ruth RA, eds. Handbook of Clinical Audiology. 4th ed. Baltimore: Williams & Wilkins; 1994: 97-108.
31. Wagener K, Josvassen JL, Ardenkjær R. Design, optimization, and evaluation of a Danish sentence test in noise. Intl Jour Audiol. 2003; 42(1):10-17.
32. Montgomery DC. Design and Analysis of Experiments. New York: John Wiley & Sons; 1997.
33. Amlani AM. Efficacy of directional microphone hearing aids: A meta-analytic perspective. J Am Acad Audiol. 2001;12:202-214.
34. Hawkins DB, Yacullo WS. Signal-to-noise ratio advantage of binaural hearing aids and directional microphones under different levels of reverberation. J Speech Hear Disorders. 1984;49:278-286.
35. Moore BCJ, Huss M, Vickers DA, Glasberg BR, Alcántera JI. A test for diagnosis of dead regions in the cochlea. Brit Jour Audiol. 2000;34:205-224.
36. Baer T, Moore BCJ, Kluk K. Effects of low pass filtering on the intelligibility of the speech in noise for people with and without dead regions at high frequencies. J Acoust Soc Am. 2002;112, 1133-1144.
37. Vickers DA, Moore BCJ, Baer T. Effects of low pass filtering on the intelligibility of speech in quiet for people with and without dead regions at high frequencies. J Acoust Soc Amer. 2001;110:1164-1175.
38. Dirks DD, Morgan DE, Dubno JR. A procedure for quantifying the effects of noise on speech recognition. J Speech Hear Disorders. 1982;47:114-123.
39. Beck L. Assessment of directional hearing aid characteristics. Audiological Acoustics. 1983;22, 178-191.
40. Mueller HG, Grimes A. A Review of experimental results with directional hearing aids: The relationship to the testing paradigm. Audiol and Hear Education. 1977;3(6):26,28.
41. Nielsen HB, Ludvigsen C. Effect of hearing aids with directional microphones in different acoustic environments. Scand Audiol. 1978;7, 217-224.
42. Ricketts T, Mueller HG. Making sense of directional microphone performance. Am Jour Audiol. 1999;8:117-127.
43. Ricketts T. Impact of noise source configuration on directional Hearing aid benefit and performance. Ear Hear. 2000;21:194-205.
44. Preves DA, Sammeth CA, Wynne MK. Field trial evaluations of a switched directional/omnidirectional in-the-ear instrument. J Am Acad Audiol. 1999;10:273-284
45. Wouters J, Litiere L, van Wieringen A. Speech intelligibility in noisy environments with one- and two-microphone hearing aids. Audiology. 1999;38:91-98.
46. Walden BE, Surr RK, Cord MT, Edwards B, Olson L. Comparison of benefits provided by different hearing aid technologies. J Am Acad Audiol. 2000;11:540-560.
47. Cord MT, Surr RK, Walden BE, Olson L. Performance of directional microphone hearing aids in everyday life. J Am Acad Audiol 2002; 6: 295-307.
Correspondence can be addressed to HR or Charlotte Thunberg Jespersen, MA, GN ReSound AS, Maarkaervej 2A, PO Box 224, DK-2630 Taastrup, Denmark; email: [email protected].