ASSESSMENT OF (C)APD IN SUBJECTS WITH PERIPHERAL HEARING LOSS

Presentation No: 15

Part 1
Introduction & Need Auditory Neuropathy Otitis Media Other conductive pathologies Presbyacusis

 (C)APD may occur in pediatric and adult patients with normal hearing or may co-exist with, or occur secondary to, peripheral hearing loss. (ASHA, 2005a)

Need to study
To analyze results of central tests keeping in mind, hearing loss as a variable. When new tests are developed, the effect of hearing loss on the performance should be dealt with. Peripheral auditory disorders can impact language development, reading, and learning. The hearing loss needs to be ruled out as a factor that could be the cause.

 Peripheral hearing loss can affect the outcome of many auditory processing tests, audibility understanding speech.

Processing problem/ attentional deficit - reduction or distortion of auditory signals.

Stimulus
Peripheral hearing loss can reduce the intrinsic

redundancy.
Reduces the ability of the auditory system to resolve spectral detail.

Stimulus
Hence, tests that place significant demands on cochlear processing to resolve frequency and intensity transitions of the acoustic signal (e.g., CV syllable) and monaural low redundancy tests are heavily dependent on rapid frequency intensity interactions will be sensitive to peripheral hearing loss.

Basic Audiologic Test Battery

Peripheral auditory dysfunction can mimic or exacerbate APD type behavioral manifestations. 1. Pure tone Audiometry 2. Performance-Intensity Functions for Word Recognition 3. Immittance Audiometry 4. Otoacoustic Emissions (OAE) 5. Auditory Brainstem Response (ABR)

 If a bilaterally symmetrical hearing loss is present and the central test results are more depressed in one ear more than the other. Result -

(C)APD

 Unilateral/asymmetrical hearing loss present, but, the better ear shows poorer performance on central auditory testing. Result

(C)APD

 If hearing loss is present in one or both ears and the central test results fall within the normal range for both ears Normal

Electrode effect
Symmetrical hearing loss, abnormal middle/late potentials are noted from electrodes positioned over one hemisphere than the other (i.e., a significant electrode effect). Result-

(C)APD CANS involvement

Ear effect
Symmetrical hearing loss, and a ear effect is noted during the electrophysiological recording. Result -

(C)APD CANS involvement

 Asymmetrical performance is noted in speech & MLR amplitude measure, and when the stimulus is presented binaurally it results in a poorer performance or score than for the better ear for a monaural presentation condition. Result:

(i.e., binaural interference effect) CANS compromise & a (C)APD is implicated (Musiek & Baran, 1996)

AN/AD
A hallmark of an APD is the audiometrically normal-hearing childs failure to hear well in the presence of competing speech or background noise (Bellis, 1996; Chermak, 2002).

 Children with AN/AD show similar hearing behavior as children with an APD, in that there is somewhat appropriate development of speech and language, normal understanding of speech in quiet, but difficulty understanding speech in noisy environments. Therefore, both categories of children will present with a normal pure tone audiogram, normal OAEs and similar speech discrimination scores in quiet and in noise.

 Although the two disorders share the similar characteristics, children with AN/AD will present with absent MEMRs (in rare cases, MEMRs can be present but elevated) and absent or abnormal ABR waveforms, whereas children with an APD have normal MEMR thresholds and normal ABR responses to a click stimulus.

It is therefore important to include MEMR testing for all children suspected of having an APD. If MEMRs are absent or elevated, an ABR recorded with both click polarities to rule out AN/AD will then be necessary.

 Children with AN/AD will be referred for further testing once it has become apparent to the parents and/or pediatrician that there is a hearing problem.

However, children who do not show major delays in the first few years of life will be completely missed.

 Some of them will later be inappropriately diagnosed with an APD (or other type of learning disability) if a hearing evaluation includes only the pure tone audiogram, word recognition in quiet, tympanometry, and OAE measures.

OM/OME
A history of otitis media in early childhood is another factor that may indicate the need for an auditory processing evaluation, as otitis media can have an adverse effect on the development of auditory processing abilities. The age of onset, number of episodes, and duration, are important factors.

Hohn and Kunze found that auditory skills were significantly depressed in those children with a significant history of otitis media in early life.

Long standing OME

Risk for abnormal auditory development. Impact of OME on central auditory processing have effects on functions believed to originate in the lower brainstem and require equal (binaural) hearing by the two ears. Inconsistent and decreased auditory stimulation during the sensitive period of auditory pathway maturation may lead to permanent disturbance, damage or changes to the pathways.

MLD
Masking level difference is the most commonly used behavioral test in OME population. Many investigators have reported reduced MLD in children with previous history of otitis media

(Pillsbury, Grose, Joseph and Hall, 1991; Moore, Hutchings and Meyer, 1991; Hall and Grose, 1993).

Pillsbury, Grose, and Hall (1991) - MLD

Reduced MLDs in children just prior to surgical intervention for prolonged histories of OME in comparison to MLDs obtained from children of the same age. While reduced MLDs could have been the result of the presurgical conductive hearing loss, many children continued to demonstrate lower-thannormal MLDs after surgical restoration of hearing.

Masking Level Difference

Effects of hearing loss will persist even after the results have been corrected for the loss (Hall and Grose, 1993a, 1993b). 1) Poorer encoding of the temporal cues inherent to the signal at the periphery (Schoeny & Carhart, 1971). 2) Long term deprivation or degradation of the acoustic signal at the periphery may lead to abnormal tuning at the brainstem level (Miller & Knudsen, 2001),

3) The frequency tested, duration of hearing loss and

age of onset.
4) In adults the presence of high frequency hearing loss does not affect the performance of 500 Hz stimuli, although MLD for speech stimuli may be diminished (Olsen & Noffsinger, 1976).

 Children with early history of OM are known to have: Abnormal binaural interaction (Moore, Hutchings
& Meyer, 1991),

Sound localization (Besing & Koehnke, 1995), Central conduction time (Gunnarson & Finitzo, 1991) Auditory closure/speech perception in noise
(Jerger et al, 1983).

 Early OM affects hearing sensitivity, ability to attend to or process auditory information in background will be affected.

Gravel and Wallace (1992)

Auditory perception in school children with a history of OM, the tests used were: Speech in noise, Filtered speech, Binaural fusion, Dichotic speech and Auditory memory.

 The effect of OM was significant only with respect to speech identification on speech in noise. Children with OM in their first year required a more advantageous signal-to-competition ratio to achieve a score of 50% in sentence intelligibility than did their peers without history of OM.

Effect of early OM on Speech processing

The primary purpose of the study was to document the effects of early onset OM on speech identification and signal processing. Speech identification scores were obtained for natural, time compressed and spectrally distorted speech stimuli.

Tests done:
Dichotic digit testing. ABR and LLR was recorded for clicks & natural, time compressed and spectrally distorted speech stimuli - /ta/.

 Spectral distortion was achieved by adding speech noise to achieve 0 dB signal-to-noise ratio. Major findings were:

 Speech identification scores for natural speech was not significantly different between the groups (Control & experimental).

Time compressed and spectrally distorted: Early OM poorer scores. Lower double correct scores on dichotic test (compromised binaural integration process) Which means:

Those with early Otitis Media could not compensate for missing information in time compressed speech to the same extent as normal hearing children.

Evoked potentials were also done of natural syllables /ta/ and /ka/ in dichotic presentation mode, participants were found to have normal ABR for dichotic stimulation.

ABR results
Mean I-III and I-V intervals were longer and mean I and III peak amplitude was reduced in 3;1 to 3;6 year old participants with early onset OM. Physiologically, prolonged interwave intervals reflect slowing down of the central conduction time in the lower brainstem.

Level of brainstem
The latency - longer & amplitude - reduced in ABRs for speech stimuli (natural, time compressed and spectrally distorted) in children with OM. The deviance in the ABR indicates that these children have difficulty in processing complex acoustic spectrum at the level of the brainstem. The deviance was also found to be more for the distorted stimuli.

Click early & Speech - delayed

LLR for click and speech stimuli showed different characteristics: LLRs for clicks in the early onset OM children occurred earlier than in normal hearing while LLRs for speech was delayed.

 This findings support reports that LLRs for clicks and speech have different neural generators. The earlier response to clicks may suggest that the cortical auditory system is compensating for the prolonged conduction time at the brainstem level. The delay for speech may be because the cortex receives deviant input from the lower centers.

Dichotic ABR & LLR

Normal ABRs and LLRs for the dichotic stimulation were observed in both the control and the experimental group

Other conductive hearing loss

The effect of conductive hearing loss is not just limited to attenuation of overall energy but it may have effects on higher central auditory nervous system at least when the pathology is a long lasting one.
(Dobie and Berlin, 1979; Webster and Webster, 1979; Gunnarson and Finitzo, 1991)

 Chronic conductive impairment in adults, due to Otosclerosis, tympanic perforation, chronic middle ear infection and Cholesteotoma may lead to reduced MLDs independent of hearing thresholds. Due to central type changes, as reflected by changes in the auditory brainstem response (ABR) (Ferguson, Cook, Hall, Grose & Pillsbury 1998)

Hall & Grose (1993a) - MLD

Subsequent restoration of hearing thresholds may lead to gradual recovery of the MLD over time, although not in all subjects.

Adult listeners with unilateral otosclerosis before stepedectomy surgery, one month following surgery, and one year following surgery.

Unilateral otosclerosis & stepedectomy

Their results indicated that the MLD in this group improved significantly over each of the subsequent tests; however two of their eight subjects did not show recovery to a normal MLD value over this time period. The authors postulated that a period of exposure to abnormal binaural auditory input can impair sensitivity to binaural cues.

There appears to be a plasticity type readjustment or adaptation of the central auditory nervous system after restoration of normal hearing, such that binaural cues restoration may return to normal/ near normal level one year after restoration of hearing.

Presbyacusis
HF HL results in decreased audibility of certain spectral components of the acoustic signal, & leads to disruption in the tonotopic organization of the CANS (Willott, 1986).
Neurons at the level of the IC that may have responded previously to HF sounds exhibit low characteristic frequencies (CFs) following the advent of cochlear hearing loss, resulting in a greater representation of low frequencies in the CANS.

 It can be predicted that this can lead to degraded encoding of high frequency spectral components of the speech signal.

LF auditory channels exhibit inherently poor temporal resolving abilities (Stuart & Phillips, 1996). Tonotopic reorganization secondary to peripheral auditory dysfunction may result in degradation of temporal encoding of the acoustic signal.

Configuration of hearing loss Cox, McCoy, Tun and Wingfield (2008)

Effects of varying degrees of peripheral hearing loss on APD tests used for the assessment of older adults, while controlling for age and cognitive abilities. They tested three subgroups, based on hearing acuity; (1) normal hearing from 500 through to 4000 Hz, (2) high frequency sloping hearing loss, and (3) hearing loss in both the low and high frequencies.

 They compared performance on six monotic APD tests: 1)Low pass filtered speech test with a fixed cut-off frequency set at 750 Hz (Auditec), 2) Pitch Pattern Sequences (PPS) adult test, 3) QuickSIN Speech-in-Noise Test, 4)SSI-ICM test, a time-compressed sentence test, and the 5) NU-6 Time-Compressed Speech test. But the SSI-ICM, low-pass filtered speech test, and timecompressed tests were significantly influenced by peripheral hearing loss.

The mean low-pass filtered speech score was 69.1% - Normal hearing, 56.3% - High frequency sloping hearing loss and 54.1% - Low- and high frequency hearing loss groups respectively.

The authors suggested that a mild to moderate low- and high frequency peripheral hearing loss may significantly influence APD test results.

 They further suggested that a mild high frequency sloping hearing loss, typical of early presbycusis, may not have a significant influence on APD test outcomes. Peripheral hearing sensitivity over the frequency range of 500 to 2000 Hz as a key predictor.

Neijenhuis, Tschur and Snik (2004)

Effect of mild hearing loss on APD test outcomes. They examined participants who were divided into two groups: Experimental group consisting of subjects with a mild, relatively flat, symmetrical sensorineural hearing loss; and a Control group consisting of subjects that had normal peripheral hearing.

6 APD tests were administered, dichotic-digits test frequency and duration pattern test sentence-in-noise test words-in-noise test filtered speech test binaural-fusion test The latter four tests were administered at two presentation levels, the usual presentation level used clinically, and at a level that was adjusted according to the subjects speech reception threshold.

 The filtered speech test used 22 words that were filtered using a low-pass filter with a fixed cut-off frequency of 500 Hz and a high-pass filter with a fixed cut-off frequency of 3000 Hz, both with a slope of 60 dB per octave, presented at 65 dB SPL.

Comparison between the two groups

The hearing impaired subjects scores were significantly lower than the control subjects scores in all six tests presented at the normal level.
After the adjustment of the presentation level, scores on words-in-noise, filtered-speech, binauralfusion tests and sentences-in-noise improved significantly in subjects with a mild hearing impairment.

 However, scores were still deviant with the exception of scores on the sentences-in-noise test which fell within normal limits. These results showed that even a mild sensorineural hearing loss can influence certain APD test results.

 A newly developed adaptive monaural, low-redundancy, low-pass filtered speech test, known as the University of Canterbury Adaptive Speech Test Filtered Words (UCAST-FW) To provide an effective measure of assessing APD in the elderly population, regardless of the listeners degree of high frequency peripheral hearing sensitivity.

 The UCAST is an adaptive speech test platform developed by Dr Greg O'Beirne and written using LabVIEW 8.20 (National Instruments, TX, USA).

 The UCAST-FW aims to eliminate the effect of any high frequency peripheral hearing loss on test performance by using low-pass filtered speech, resulting in test items with spectral content almost entirely below 1 kHz.

Processes
Binaural interaction Temporal processes Dichotic listening tasks

Binaural Interaction
MLD
Quranta, Cassano and Cervellera (1978) concluded that MLD (for 500 Hz tones) were not useful diagnostically to detect central impairment unless peripheral hearing sensitivity was normal.
Caihart (1976) demonstrated that persons with peripheral impairments show reduced MLDs.

Binaural Fusion Test

Affected by the presence of hearing loss. Subjects with mild hearing loss, and flat audiometric configuration, score significantly poorer than subjects with normal hearing. Even though the test stimuli were presented at an adjusted level (i.e., SL) and when scores were corrected on the basis on pure tone average by means of linear regression analysis (Neijenhuis et al., 2002).

Rapid Alternating Speech Perception

Affected by peripheral hearing loss (Dawson et al., 1978) The goal of the study was to examine the effect of peripheral hearing loss on auditory tasks that are used to assess dysfunction within the central auditory pathways.

Dawson et al. (1978)

Each subject received a central auditory processing (CAP) test battery that consisted of: (1) a dichotic sentence listening task; (2) a monosyllabic filtered word task; (3) a spondaic word binaural fusion task; and (4) a rapidly alternating speech task. All of these tasks were affected by certain degrees/configurations of SNHL, with the monosyllabic filtered word task being the most seriously affected.

Temporal processes
Of all the temporal processing tests, the Duration Patterns Test appears to be the most resistant to hearing loss (Musiek et al., 1990).

 In their 1990 study, they demonstrated that the performance of the subjects with cochlear hearing loss was almost indistinguishable from the performance of those without hearing loss. 90% of the individuals with hearing loss, including those with moderate degrees of hearing impairment performed within normal limits.

 The frequency pattern sequences test (Humes et al, 1996; Musiek & Pinheiro, 1987) and the auditory duration pattern test (Humes et al, 1996; Musiek, Baran & Pinheiro, 1990) have also been found to be: Resistant to confounding effect of mild to moderate peripheral hearing loss Provided the stimuli are audible.

 The frequency pattern test may be used with mild relatively flat hearing losses, Because there is a frequency or a spectral element involved it may not be completely resistant to hearing loss (Musiek and Pinherio, 1987).

 As the Gap in Noise test incorporates a broad band stimulus, it may also be used with individuals with cochlear hearing loss.

The broad band stimulus may not vary as a function of peripheral hearing loss (Musiek et al., 2005)

 The tonal stimuli used in both the AFTR (by McCroskey and Keith (1996)) And the RGDT (Random Gap Detection Test) is beneficial in its ability to determine an individuals frequency specific temporal resolution abilities as well as its ability to help in testing individuals who present with peripheral hearing loss.

 Tonal stimuli may enable, even individuals with peripheral hearing losses, to have their temporal resolution abilities tested without possible contamination or influence of a peripheral hearing loss. The assumption is that these individuals must have at least one of the frequencies at normal to near normal hearing acuity.

 The click stimuli utilized in the RGDT are beneficial in that clicks are spectrally complex, enabling testing of individuals regardless of the presence or absence of most configurations of peripheral hearing losses. This may be an excellent screening tool for auditory processing abilities (Jerger and Musiek, 2000).

Psychoacoustic considerations
Gap Detection Thresholds have been found to be similar for normal hearing and hearing impaired adults (Lister et al., 2000). This finding lends support to the theory that perceptual channels for processing gap detection are centrally located because damage to the peripheral system does not appear to affect the gap detection thresholds. Conversely, there are studies that show the largest GDT when a SNHL is present (Florentine and Buus, 1984)

Sensorineural Hearing loss

Listeners may have difficulty detecting a brief pause or gap in a continuous noise (Boothroyd, 1973; Fitzgibbons & Wightman, 1982). Explanations provided were 1) Information required to perform the task was below the subjects absolute threshold (Boothroyd, 1973). 2) Experiments on gap detection on octave band of noise have shown that temporal resolution is better at high than low frequencies (Buus & Florentine, 1982; Fitzgibbons & Wightman, 1982).

 This indicates that the high frequencies are responsible for the detection of gaps in a broadband noise and that hearing impaired listeners cannot hear the high frequency portions of the BBN. These findings point to the role of audiometric configuration in gap detection.

 Temporal resolution is poorer than normal. The deficit with condition equaled for SL must be attributed to processing distortion imposed by cochlear damage. It might indicate increased persistence of sensation in the cochlear impaired listeners.

 The influence of frequency content on subjects performance suggests that the configuration of hearing loss may be a determining factor of temporal resolution (Glasberg and Moore, 1988).

Temporal Resolution in Regions of Normal Hearing and Speech Perception in Noise for Adults with Sloping HFHL
Temporal resolution in the low-frequency region with near-normal sensitivity seems to be deteriorated in subjects with HF SNHL.

They were more sensitive to increases in speech rate, suggesting that poorer temporal processing may be related to speech perception deficits in noise.

Dichotic Listening Tasks Dichotic Digit Test

Musiek (1983a, 1991) used pairs of digits presented dichotically and found that normal hearing obtained 90% scores, or more on the DDT. He reported that the test was relatively unaffected by peripheral hearing loss. Mild moderate hearing loss, a cut-off of 80% is required.

Speaks, Niccum and Van Tassel (1985)

Effect of stimulus material on the dichotic listening performance of patients with bilateral symmetrical mild sensorineural hearing loss. They tested the performance with 4 tests: Digits, Vowel words (e.g., key vs. cow), Consonant words (fan vs. pan) CV non-sense syllables.

 Monotic performance intensity function for each ear was defined. The 4 dichotic tests produced reliable differences among scores for the left ear and the right ear, in terms of performance level and ear advantage.

 The digit test was found to be most resistant to sensorineural hearing loss, because performance for digits was only slightly affected. It was not possible to estimate with confidence, the true contribution of the hearing loss on the dichotic test results because the scores for the left and right ears, that would have been obtained for the sample patients in the absence of the hearing loss was not known.

 As the difficulty of the test stimuli increased the ear advantage also increased. The use of digits seem to be least affected by sensorineural hearing loss (0% ear advantage).

 CV syllable were most affected by peripheral hearing loss, as it is dependent on fast resolution of intensity and frequency transitions (Chermak, 2001).

Chermak & Musiek, 1997

Stimuli with greater complexity of intensity and frequency interactions, especially over a restricted time period, place a greater demand on cochlear processing, and is influenced more by HL.

Dichotic CV test
Absence of REA was found in subjects with bilaterally symmetrical moderately severe SNHL, as reported by Roeser et al. (1976) who carried out dichotic CV testing at 30 dBSL. They also found large ear differences (right or left) suggesting that SNHL may affect the size and direction of ear advantage in dichotic listening tasks.

 According to, Roeser et al. (1976) and Cattley (1977) subjects with bilaterally symmetrical moderately severe SNHL failed to demonstrate REA.

Both studies indicate that sensorineural hearing loss affects the central processing of dichotic CV syllables.

 In contrast, Cattey (1981) who also administered dichotic CV test to moderate SNHL cases at 75 dB SPL found a right ear advantage even when the lagging syllable was presented to the left ear, and it was not enhanced when the lagging syllable was presented to the right ear.

Speaks, Bauer and Carlstrom (1983)

Assessed the extent to which a peripheral loss may confound the interpretation of dichotic listening tests the assessment of central auditory dysfunction.

They tested normal hearing subjects with CVs monotically and dichotically in two conditions:

 Conductive hearing loss simulated with ear plugs(unilateral): This testing was carried out at 78 dB SPL. Without ear plugs: This testing was carried out at 40110 dB SPL.

 With the plug, the magnitude and direction of the REA varied with the test intensity even when the monotic speech recognition in both ears was greater than 95%. Thus, when dichotic tests are used to assess central auditory dysfunction in patients with HL, the authors recommended that the intensity be at least 10 dB from both the upper and lower knees of the monotic performance intensity function.

 Conductive hearing loss does not seem to affect the performance significantly as long as the stimuli is presented at an intensity of 12 dB above the monotic knee for the affected ear(s). The monotic knee represents that point in the performance intensity function at which the listener reaches 95% accuracy level for monotic presentation of the stimuli (Speaks et al., 1987). Thus, research on the dichotic CV in peripheral HL, reveal changes in the ear advantage.

Dichotic sentence identification test for HI adults

STIMULI? These dichotic sentence materials were then administered to 14 normal listeners and 48 hearingimpaired subjects to determine the influence of peripheral HL on test performance. Results suggest that the Dichotic Sentence Identification test is resistant to the influence of peripheral hearing loss until the pure-tone average of 500, 1000, and 2000 Hz exceeds approximately 50 dB. Beyond this level, degree of peripheral hearing loss limits its value for detecting central auditory disorder.

Staggered Spondaic Words

Flynn et al. (1984) investigated effects of intensity on mild to moderate cochlear hearing impaired adults performance.

The SSW test was presented at 20, 30, 40, and 50 dB SL reference PTA. The responses were used to generate performance intensity functions.

 No significant differences were found for these HI listeners performance for items presented at the standard 50 dB SL and those at low sensation levels. Thus, the authors concluded that the SSW may be administered on cochlear HL patients at SLs as low as 20 dB SL. Arnst, 1982; Katz, 1968 found SSW to be resistant to peripheral hearing loss.

Speech in noise tests

Can provide information of how an individual performs with a background of competition, functional impact of auditory disorder and may be affected by hearing loss. Several speech in noise tests have been developed for use with HIs:

Quick-SIN (Etymotic Research, 2001) has a list with 30 dB high frequency emphasis for use with ski-slope losses (If PTA is 45 dB HL presentation level should be 70 dB HL or at a loudness level of loud but comfortable for those with greater degree of hearing loss Hearing In Noise Test (HINT, Nillson, Soli and Sullivan, 1994) Use of self assessment hearing handicap forms.

Thank you for listening

Part 2
Localization
Listening in Spatialized Noise Conductive HL Sensorineural HL Pinna cues

LISN-S
The study supports the hypothesis that a high proportion of children with suspected (C)APD have a deficit in the mechanisms that normally use the spatial distribution of sources to suppress unwanted signals. The LISN-S is a potentially valuable assessment tool for assessing auditory stream segregation deficits, and is sensitive in differentiating various forms of auditory streaming.

 The Listening in Spatialized Noise--Sentences test (LISN-S) is a revised version of the Listening in Spatialized Noise (Continuous Discourse) test (LISN; Cameron et al, 2006a). The software produces a 3-D auditory environment under headphones and was developed to assess auditory stream segregation skills in children. A simple repetition response protocol is utilized to determine speech reception thresholds (SRTs) for sentences presented from 0 degree azimuth in competing speech. The competing speech is manipulated with respect to its location in auditory space (0 degree vs. + and -90 degrees azimuth) and the vocal quality of the speaker(s) (same as, or different to, the speaker of the target stimulus). Performance is measured as two SRT and three advantage measures. The advantage measures represent the benefit in dB gained when either talker, spatial, or both talker and spatial cues combined are incorporated in the maskers.

 Correlation between performance on the LISN-S and a traditional (C)APD test battery was also compared. The LISN-S did not correlate significantly with any test in the traditional test battery, nor were the nonspatial and spatial performance measures of the LISN-S correlated.

Facts
Low-frequency interaural time cues are disrupted in conductive/mixed hearing loss. Conductive hearing impairment leads to particular problems in horizontal plane localization, apparently because of disturbed interaural time difference information.

Hausler, Colburn and Marr (1983)

The sound localization impairments found in conductive hearing losses are interpreted as boneconduction effects.

Results found in sensorineural hearing losses are interpreted as consequences of impaired or preserved spectral processing.

Noble, Byrne and Lepage, 1994

Loudspeakers 18 degrees apart

Finding
Independently of the degree of hearing loss, the presence of a conductive component produces a further decrement in localization performance. The variation in localization performance of hearingimpaired listeners is only partly explained by HL. This suggests that aspects of impairment, other than attenuation, may be operative

 There are significant differences in localization performance between the sensorineural and conductive/mixed groups as a whole; there are also significant differences in average HTLs between them. To assess the true effect of type of hearing loss, samples were derived matched for degree of impairment.

Normal & SNHL = BC > AC Conductive HL -- BC = AC With BC Both cochleae stimulated equally

Hausler et al. (1983) - Disturbed time discrimination results from loss (or reduction) of cochlear isolation.
Conductive/mixed HL, transmission by air is reduced and may be approached or equaled by bone conduction.

Cochlear Isolation is reduced

Normal & SNHL = BC > AC Conductive HL -- BC = AC With BC Both cochleae stimulated equally

Cochlear Isolation - Amount of interaural time difference

Zurek (1986)
Showed that intercochlear phase cues increase as the sensitivity of the bone conduction path approaches that of the air conduction path. This would be more likely to happen for lowfrequency sounds, where interaural attenuation is less, than for higher frequency sounds.

Distortions" - A term used to cover collectively a range of impaired psychoacoustic abilities (Plomp, 1978) The relationship of localization to "distortion" aspects of hearing impairment may help to explain speech recognition in noise, which is not very predictable from hearing loss (Duquesnoy, 1983)