Friday, 11 December 2009

Good basic info regarding Bi-lateral CI....

From a thesis by Carol A. Sammeth, Ph.D, CCC-A

Please be careful that the statements selected are representative of the paper's overall conclusions.  Also, the part posted is only a piece of a longer work. (Full pdf here..)

The Psychoacoustic Literature
There is a fairly voluminous literature in psychoacoustics (hearing science) illustrating the benefits in normal hearing persons of having two-eared rather than one-eared input. When hearing loss disrupts the ability of the brain to process binaural inputs, whether due to large differences in the degree of loss between the ears, or a failure to provide amplification or a cochlear implant to one impaired ear, these benefits can be severely degraded or lost. There are three primary effects ascribed to binaural listening: the head shadow effect, the binaural summation effect, and the binaural squelch effect (e.g. Durlach & Colburn, 1978), producing benefits that range from improved speech recognition in noise, to the ability to localize the direction of a sound, to more “natural” perception. The following briefly describes the key benefits of binaural functioning.

Head Shadow Effect
When speech and noise come from different directions (i.e. are spatially separated, as typically occurs in the real world), there is always a more favorable signal-to-noise ratio (SNR) at one ear than at the other because of the head shadow effect and different sound distances to the ears. The head shadow effect is primarily seen in frequencies higher than 1500 Hz (e.g. Shaw, 1974), with the amount of attenuation of sounds from the opposite side of the head dependent on frequency but ranging from about 7 dB in the speech range up to 20 dB or more at the highest frequencies. If both ears are participatory, the ear with the most favorable SNR is always available so that the patient can selectively attend to this ear. This is compared to the unfavorable situation where only the ear with the poorer SNR is functional. Persons with unilateral hearing loss can become very frustrated when people are talking on both sides of them because they must constantly turn their “good ear” to whomever they want to hear best at the time, and then they miss sounds on the deaf ear side.

As will be shown later in this paper, a primary benefit of bilateral cochlear implants appears to be related to the beneficial aspects of hearing from both sides, and always having the ear with the more favorable SNR available. This is generally tested with speech from a frontal speaker and noise from a side speaker - - when the second ear is added that is contralateral to (opposite side of) the noise source, performance benefit comes primarily from the head shadow effect. Note, however, that there is discrepancy across the published studies in how to quantify the head shadow effect, with some researchers merely examining differences in scores with sound ipsilateral versus contralateral to a unilateral ear under test, and others comparing the score for listening with bilateral inputs to that for unilateral listening with the noise presented ipsilateral to the ear under test.

Binaural Summation & Redundancy
Sounds that are presented to both ears rather than just one are perceived as louder due to binaural summation of the information received at each ear. In fact, the threshold of hearing is known to improve by about 3 dB for binaural versus monaural presentation to normal ears, resulting in a doubling of perceptual loudness and improved sensitivity to fine differences in the intensity and frequency domains. This latter effect is sometimes referred to as binaural redundancy, and it is believed that it may translate into improved speech perception scores. When listening to speech with only one ear in a difficult listening situation or with one ear with greater sensorineural hearing impairment than the other, there is a loss of the redundancy in cues across the ears that may reduce performance.

The benefit of the binaural redundancy aspect of bilateral inputs is typically tested by presenting speech alone or having speech and noise emanate from the same loudspeaker frontally - - when the second ear is added, benefit is possible through redundancies or overlaps in representation at the two ears. In a normal hearing ear, this effect produces about a 1 to 2 dB improvement in SNR (Bronkhorst & Plomp, 1988). At this time, there is only limited evidence for true binaural redundancy effects on speech perception results in the
bilateral cochlear implant literature reviewed herein. This effect is probably not stronger either because such subtle cues are not able to be utilized by ears that have severe to profound hearing loss, or simply because the signal processing available in today’s cochlear implants (with two implants processing independently) does not adequately maintain these interaural cues.

Binaural loudness summation has been shown to occur, however, and is a potential confounding factor in comparing across studies. While most researchers have adjusted the loudness of the implant processing for binaural presentation versus monaural presentation (and made sure loudness is reasonably balanced across the ears), some have not. In a clinical bilateral implant fitting, it would generally be presumed that loudness would be adjusted so that the patient’s overall loudness comfort level is reasonable, and thus any purely binaural summation effects would be reduced or negated for bilateral listening compared to a previous unilateral implant.

Binaural Squelch/Unmasking
A person with only one functioning ear can usually understand conversation well when listening in a quiet environment, as long as the sounds of speech are made loud enough. However, even a normal hearing person who is listening in high levels of background noise can find speech understanding to be difficult in an adverse listening situation (consider, for example, competing conversations with multiple persons seated at a long table in a very high noise level restaurant). This occurs partly because of direct masking and partly because of upward spread of masking on the basilar membrane of the cochlea (whereby low-frequency sounds have a greater impact on reducing perception of higher-frequency sounds than vice versa). Speech recognition in such noisy environments is even harder for a person with sensorineural hearing loss both because of the inherent distortion and loss of normal nonlinearities introduced by cochlear damage, and because these patients show even greater amounts of upward spread of masking effects than do normal ears.

Fortunately, the auditory nervous system is wired to help in noisy situations as long as there is functional input from both ears - - that is, the auditory system and brain can combine information from both ears so that there is a better central representation than would be had with only information from one ear (e.g. Zurek, 1993). This effect, commonly referred to as binaural squelch (but also sometimes called binaural unmasking), results from the brainstem nuclei processing timing, amplitude, and spectral differences between the ears to provide a clearer separation of the speech and noise signals. The squelch effect takes advantage of the spatial separation of the signal source and the noise source(s) and the differences in time and intensity that these create at each ear. This is generally tested with speech from a front speaker and noise from a side speaker - - when the second ear is added that is ipsilateral to (same side as) the noise source, any benefit comes from the binaural squelch effect. There is some limited evidence of improved speech understanding in noise in bilateral cochlear implant patients due to binaural squelch effects, although the effect is not seen across all bilateral implant users or studies, and is not as large as the head shadow effect.

Note that binaural summation and squelch are signs of the ability of the auditory nervous system to integrate, fuse, and use information from the two ears. In contrast, the head shadow effect merely results from the physical attenuation of sound across the head and does not require central nervous system integration - - This does not negate the fact, however, that the head shadow effect is a substantial factor in everyday performance for those listeners with unilateral versus bilateral devices.

Finally, perhaps the most well-known practical binaural benefit is the ability to localize (i.e. determine the direction that a sound is coming from). This function is dependent on auditory Bimodal Devices and Bilateral CIs, page 9 system perception of interaural (between ear) differences in time, intensity, and phase (e.g. Yost & Dye, 1997). Localization ability can be a safety consideration. For example, when crossing a busy street, it is important to know the direction that a car is coming from. Persons with significant unilateral hearing impairment can also attest to the frustration of hearing their name spoken but not knowing which direction to turn in order to find the person calling them.

Research to date has focused on localization of sound sources in the horizontal azimuth, but keep in mind that it is also possible for a listener to differentiate sound sources in the vertical plane (by elevation) and in terms of the distance from the listener. It is well known that interaural timing differences provide the information necessary to locate the direction of low frequency sounds - - specifically, those less than about 1500 Hz. For sounds that are higher in frequency, the main cue for horizontal plane localization is the interaural intensity difference that occurs because of the head shadow effect. In addition, head and pinna shadow effects,
pinna filtering effects, and torso absorption properties can all contribute to spectral differences that can be particularly helpful in determining elevation of a sound. For a listener with only one functional ear, there are very few cues to assist in sound localization although some rudimentary localization ability can still exist. The literature on bilateral cochlear implants provides significant and substantial evidence that localization abilities are enhanced with the use of both ears versus just one.

No comments:

(Some) Milestones

  • 2013-08: Grade 6
  • 2012-08: Grade 5
  • 2011-08: Grade 4
  • 2011-03: BTE's on the ear
  • 2010-08: Grade 3
  • 2009-08: Grade 2
  • 2008-08: Mainstream School (6y. old)
  • 2006-10: All-hearing Kindergarten (4y. old)
  • 2004-11-22: CI activated (27 m. old)
  • 2004-10-04: Bi-lateral CI (26 m. old)
  • 2003-08: Deaf/HOH/CI Pre-school/"DEAF" Kindergarten (12m. old)
  • 2003-07: HA's fitted (11 m. old)
  • 2003-06: Diagnosed deaf. Start sign-language (10m. old)
  • 2002-11: Suspicion loss of hearing (4 m. old)
  • 2002-08: Born - A fierce LION
--- Google Analytics