Postgraduate student (Audiology),
Department of Audiology,
All India institute of Speech and Hearing,
Naimisham campus,
Manasagangothri,
Mysore - 570006.
kaditee@yahoo.com
N.Kartik,
Postgraduate student (Audiology),
Department of Audiology,
All India institute of Speech and Hearing,
Naimisham campus,
Manasagangothri,
Mysore - 570006.
C.S. Vanaja,
Lecturer, Department of Audiology,
All India institute of Speech and Hearing,
Naimisham campus,
Manasagangothri,
Mysore - 570006.
Abstract:
The present investigation was undertaken to study the perceptual deficits in a population of sensorineural hearing loss (SNHL) patients. Auditory brainstem responses (ABRs) were obtained on 20 ears with mild to moderate SNHL for this preliminary study. Two stimuli were used to evoke the ABR; a standard acoustic click and the burst portion of the syllable |t|. The results of the present study indicate that while click evoked ABRs exhibited latency values within normal limits, speech burst evoked ABRs showed more deviant results, perhaps suggesting that using speech sounds to elicit the ABR offers an opportunity to better isolate normal speech processing from abnormal speech processing, and we suggest how this might be useful for patients with possible auditory processing disorders.
This study aims to highlight the clinical application of speech burst ABR in possibly identifying perceptual deficits in SNHL patients, and discusses the possible use of this test in objectively identifying auditory processing disorders.
Introduction:
Hearing impaired individuals presumably process complex signals (i.e., like speech) in a manner different from those with normal hearing. Processing deficits are likely present due to abnormal representation of complex speech signals at the cochlea, the eighth nerve, the brainstem, and/or the auditory cortex.
Abnormal processing at the level of cochlea is evident through poor frequency selectivity as portrayed by broadening of tuning curves in the hearing impaired population (Leek & Summers, 1996). Performance of those with SNHL deteriorates with increased demands for finer degrees of spectral resolution (Summers & Leek, 1994).
The literature reveals features like minimum contrast for vowel identification, frequency selectivity, preservation of spectral contrast in noise and internal representation of spectral contrasts are all affected in individuals with SNHL (Summers & Leek, 1994; Leek & Summers, 1996).
Hearing impaired individuals respond to auditory stimuli with normal or near normal reaction times when stimulus loudness is predictable. However, with the introduction of uncertainty, performance of individuals with SNHL deteriorates. There is always better performance for higher intensity signals for subjects with SNHL (Seitz & Rakerd, 1997). With regard to spectral resolution, although subjects with SNHL are able to use the temporal envelop information within a single channel in a normal fashion, when combining information across multiple channels, abnormal/deviant performance occurs (Turner, Chi & Flock, 1999).
Abnormal frequency resolution is associated with SNHL and produces a smearing of spectral detail with regard to internal representation of complex acoustic stimuli. Individuals with SNHL may have difficulty differentiating spectral peaks within stimuli, and these peaks appear to be critically important as markers and indicators of the true acoustic identity.
A multitude of differences exist between normals and individuals with SNHL, with regard to many psychoacoustic measures not limited to; auditory filter characteristics and spectral contrast detection, as a sequalae of poorer frequency resolution highly related to SNHL. (Summers & Leek, 1994).
Objective audiology tests are extremely useful in assessing deficits in the auditory pathway, especially "below" the cortical level. However, there are very few objective options to identify normal versus abnormal processing of speech and complex stimuli within the auditory system.
Auditory brainstem response (ABR) is widely used in audiology and neurotology as an objective tool for assessing hearing sensitivity and auditory nerve function. The ABR represents the initial processing and transmission of acoustical signals through the auditory nerve and brainstem (Moller, 1994). Brief acoustic signals, such as clicks, tone bursts and tone pips have been used to elicit the ABR. The ABR response, a measure of neural integrity, has been shown to be extraordinarily useful as a screening tool for hearing loss and other auditory system abnormalities, and as a screening tool to identify retrocochlear disorders, such as acoustic neuroma.
However, according to the motor theory of speech perception (Liberman, Cooper, Shankweiler,& Studdert-Kennedy, 1967), speech events are not processed and transmitted in the same way as non-speech signals. Several studies have used speech as a stimulus while recording evoked cortical responses (Kraus, Mc Gee, Carrell, Sharma, Micco & Nicol, 1993; Naatenen, Lehtokoski, Lennes, Cheour, Huotilainen, Iivonen, Vainio, Alku, Ilmoniemi, Luuk, Allik, Sinkkonen & Alho, 1997). These studies have promoted a deeper understanding of neurophysiology of speech processing at the cortical level.
It can be argued that the processing of speech and speech sounds is potentially more "meaningful" with respect to psychological and linguistic issues, than the processing of clicks, tone bursts and pips. As such, speech-evoked ABR recordings may have diagnostic and clinical management implications to help screen or identify patients with abnormal speech processing issues, or perhaps those with auditory processing disorders (APD).
The present study aims to evaluate the potential of speech burst evoked ABR to potentially identify speech processing deficits in individuals with SNHL.
Methods:
Subjects:
Our subjects include 11 individuals. Each individual was tested on their right and left ears for a total of 22 ears. The age range of our subjects was between 15 and 50 years, the mean age was 30 years. All our subjects had mild to moderate sensorineural hearing loss (SNHL). Their pure tone averages (PTAs) ranged from 25 to 50 dB HL. All subjects presented with normal middle ear function and with bilaterally present acoustic reflexes.
Instrumentation:
The OB922 dual channel clinical audiometer (Version 2) was used for pure tone testing and speech audiometry. The GSI Tympstar middle ear analyzer was used for tympanometry and acoustic reflex measurement and recording.
ABRs were recorded using the Intelligent Hearing Systems (IHS) Smart EP system, version 2.39. This instrument was used to record and analyze ABRs. Eartone 3A insert earphones were used to present stimuli.
Materials:
The burst portion of the syllable /t/, with a duration of 1.05 msecs, as recorded by Reddy, Kumar & Vanaja (2004) was the speech burst evoked stimulus. The standard acoustic click, was also used in this study.
Figure 1. (above) shows the spectrum of the speech burst |t|
Standardized Kannada paired words were used to determine the Speech Reception Thresholds (SRT) and a common list of monosyllables developed by Mayadevi (1974) was used to establish the Speech Identification Scores (SIS). (Note, in the USA and other countries, the commonly used word recognition score (WRS) and/or the Speech Discrimination Score, would be anticipated to be analogous to the SIS.)
Test Procedure:
On the day of their tests, each subject was evaluated using the tools noted above, and otoscopy was performed on all subjects too, to assure that no visible external or middle ear abnormalities were present on the day of the test. Pure tone thresholds were acquired from 250 to 8000 Hz via air consuction, and when clinically appropriate, bone conduction thresholds were also acquired to 4000 Hz. Speech audiometry was carried out using live voice and the SIS (analogous to the word recognition score) was obtained at 40 dB SL, re: SRT. Therefore, if the SRT was determined to be 30 dB, the SIS test was accomplished at 70 dB. As indicated above, tympanometry and acoustic reflexes were recorded to rule out middle ear pathology.
ABR recording:
Subjects were seated in a reclining chair in an electrically shielded and acoustically treated room. Silver chloride electrodes (AgCl) were placed at the recording sites, after cleaning those sites with an abrasive gel. Electroencephalography (EEG) paste and surgical adhesive tape was used to hold the electrodes firmly in place. In essence, standard and well accepted ABR protocols were used throughout all ABR acquisitions.
The burst of syllable /t/ was presented as a stimulus. After the speech burst recording was obtained, the click evoked ABR was obtained. All stimuli were presented through earphones and 1500 sweeps were obtained for both stimuli at 30 dB SL, for stimulation of both ears. Stimuli were presented at a repetition rate of 30.1 per second for all recordings. A filter setting of 10 Hz - 3000 Hz was used and responses were amplified 100 K times. The analysis window was set to post stimulus epoch.
ABR Analysis:
Recorded ABR waveforms for click and speech burst stimuli were analyzed with respect to latency and peak-to-peak amplitude of wave V. Latency results were correlated with performance on SIS to evaluate their relationship. Comparisons were also made for latencies and amplitudes of wave V evoked by click stimulus versus speech burst. Comparisons were made across normals (reference data from Reddy, Kumar & Vanaja, 2004) and the clinical population (described above) to investigate any significant differences between the two groups.
Results & Discussion:
Wave latency:
Reddy, Kumar & Vanaja (2004) reported the mean values for latency measures for ABRs elicited using speech burst of |t| and click in individuals with normal hearing sensitivity. Table 1 (below) compares the result of the present study with Reddy, Kumar & Vanaja, and provides insight into the differences noted in the mean values across normal listeners and individuals with sensorineural hearing impairment.
Table 1. Comparison of results obtained for latency measures in normal listeners (Reddy, Kumar & Vanaja, 2004) and individuals with SNHL elicited with clicks and speech bursts.
Both stimuli (clicks and speech burst) evoked ABR waveforms. While click evoked ABR showed normal or slightly delayed latencies, speech burst evoked ABR exhibited statistically significantly (independent sample t-test) delayed latencies for wave V (see Reddy, Kumar & Vanaja, 2004).
Pearson's product moment correlation (bivariate) between the latency values for the click and speech burst elicited ABR & SIS were evaluated.
A statistically significant correlation (p
Although delayed latencies may indeed be indicative of SNHL, increased thresholds or lowered hearing sensitivity, the significantly delayed latencies with speech burst elicited ABR may potentially indicate processing abnormality at peripheral and higher cortical levels.
The importance of using speech burst in identification of processing holds great promise. Speech burst elicited ABR may indicate abnormal neurophysiologic representation of speech at the level of the cochlea, eighth nerve and brainstem, which routine click evoked ABR fails to highlight.
The right hemisphere is predominant in the perception of slow acoustic transitions, whereas neither hemisphere dominates the discrimination of nonspeech sounds with fast acoustic transitions. In contrast, the perception of speech stimuli with similarly rapid acoustic transitions is dominated by the left hemisphere, which may be explained by the presence of acoustic templates (long-term memory traces) for speech sounds formed in this hemisphere (Shtyrov, Kujala, Palva, Ilmoniemi, and Naatanen, 2000).
Wave amplitude:
Peak to peak amplitude measures were obtained for waves V and are shown in table 2. Wilcoxon signed rank test showed significant difference between the amplitudes of wave V evoked by click and speech burst in the hearing impaired group. This was in contrast to the results reported by Reddy, Kumar & Vanaja (2004). This difference in the results can be attributed to the spectral content and the temporal envelope of the speech burst stimulus in comparison to the click stimulus. The amplitude of the peak is principally contributed to by the number of nerve fibers firing at the onset of the stimulus. The population of fibers activated in turn depends on the spectrum of the stimulus, and hence the results.
Table 2. Comparison of results obtained for amplitude measures in normal listeners (Reddy, Kumar & Vanaja, 2004) and individuals with SNHL.
Figure 2. Displays the acquired waveforms for click and burst |t| stimulus in one subject with SNHL.
Conclusion:
The present study compares the electrophysiological responses to different stimuli; speech burst of syllable |t|, and click elicited ABR, with consideration to SIS. Speech burst ABR appears to more sensitive than conventional click stimuli in potentially identifying abnormal perceptual processing attributes in subjects with SNHL.
Although this study is preliminary, and conclusions cannot yet be made, it does indicate a possibility that speech burst stimuli may offer a different opportunity to identify auditory abnormalities in people with SNHL, than does ABR.
Additional studies should address the multitude of factors introduced here. Areas for study include; correlation of speech burst elicited ABR with more specific types and degrees of hearing loss, analysis of speech evoked ABR to tests of auditory processing disorders (APD) and various spectro-temporal speech burst options and alternatives.
Regarding the potential for an electrophysiologically based APD test, this perhaps holds the greatest potential. Currently, the majority of APD tests are behavioral and in that respect, subjective. Additionally, it is sometimes difficult to find trained personnel to facilitate APD tests and their recommended aural rehabilitative management techniques.
If the speech burst evoked ABR test proves to be reliable in identifying abnormalities, and if these abnormalites are highly related or correlated with APD findings, the speech burst evoked ABR might be used as a screening, or diagnostic tool for patients with a high index of suspicion for APD. In that situation, if the speech burst evoked ABR finding was positive, a more in-depth APD evaluation would be recommended.
References:
Kraus, N., McGee, T., Carrell, T.D., Sharma, A., Micco, A., Nicol, T. (1993). Speech evoked cortical potentials in children. Journal of American Academy of Audiology, 4, 238-248.
Leek, M.R., Summers, V. (1996). Reduced frequency selectivity and the preservation of spectral contrast in noise. Journal of Acoustical Society of America, 100(3), 1796-1806.
Liberman, A.M., Cooper, F.S., Shankweiler, D.S. & Studdert-Kennedy, M. (1967). Perception of the speech code. Psychological review, 74, 431-461.
Mayadevi, C. (1974). The development and standardization of a common speech discrimination tests for Indians. Unpublished Master's dissertation submitted to University of Mysore.
Moller, A.R. (1994) Neural generators of auditory evoked potentials. In J.T. Jacobson (Ed.), Principles and applications in auditory evoked potentials (pp. 23-46). Boston: Allyn & Bacon.
Naatanen, R., Lehtokoski, A., Lennes, M.M., Cheour, M., Huotilainen, M., Iivonen, A., Alku, P., Ilmomiemi, R.J., Luuk, A., Allik, J., Sinkkonen, J. & Alho, K. (1997). Language specific phoneme representations revealed by electric and magnetic brain responses. Nature, 385, 432-434.
Reddy, S.M., Kumar, U.A., Vanaja, C.S. (2004). Characteristics of ABR evoked by speech bursts. Scientific paper presented at the 36th National conference of the Indian Speech & Hearing Association (ISHA).
Sietz, P.F., Rakerd, B. (1997). Auditory stimulus intensity and reaction time in listeners with long standing sensorineural hearing loss. Ear and Hearing, 18(6), 502-512.
Shtyrov, Y., Kujala, T., Palva, S., Ilmoniemi, R.J., and Naatanen, R. (2000). Discrimination of speech and of complex nonspeech sounds of different temporal structure in the left and right cerebral hemispheres. Neuroimage, 12(6), 657-663.
Summers, V., Leek, M.R. (1994). The internal representation of spectral contrast in hearing impaired listeners. Journal of Aoustical Society of America, 95(6), 3518-3528.
Turner,C.W., Chi, S.L., Flock, S. (1999). Limiting spectral resolution in speech for listeners with sensorineural hearing loss. Journal of Speech Language and Hearing Research, 42(4), 773-784.
