American Scientist

Ten years ago, Blake Wilson and I described for readers of American Scientist the design and function of the first successful and widely available neural prosthesis for a sensory system, otherwise known as the cochlear implant. Today, approximately 500,000 individuals worldwide have been fitted with this technology and have experienced the significant gain in quality of life that improved hearing affords. In this post I describe two results from the newest patient group to benefit from this technology: individuals who are single-sided deaf, with one normal hearing ear and one deafened ear.

Brain Rewiring, and Reversal

Our young patient, Claire, had been born with normal hearing in both ears. At age 5, she began to lose hearing in one of her ears. By age 9, she had very little hearing, and little or no speech understanding, in that ear. Just before her 10th birthday, she received a cochlear implant in her deafened ear.

Before Claire received her implant, neuroscientist Anu Sharma at the University of Colorado, Boulder, fit a cap with 128 electrodes on her head ito take an electroencephalogram (EEG). This imaging method mapped cortical activity in response to auditory, visual, and tactile stimulation. The results showed significant reorganization of Claire’s cortex compared to what would be expected for a person without hearing loss.

For individuals with normal hearing in both ears, stimulation of one ear produces cortical activity in both brain hemispheres. For Claire, stimulation of either her hearing-impaired ear or her normal-hearing ear produced cortical activity in only one hemisphere. Moreover, stimulation of her normal-hearing ear elicited activity in a part of the brain called the inferior frontal cortex—a sign that her brain was dealing with a processing load higher than usual. Other analyses indicated the presence of cross-modal cortical reorganization, meaning that her brain had been “cross wired” for different types of stimulation. For example, tactile stimulation produced activity in areas normally activated by auditory stimulation. Thus, the evidence was clear that Claire’s brain had been reorganized by a relatively short period of unilateral auditory deprivation and the reorganization was not limited to the auditory system. Critically, cortical response to stimulation delivered to the normal-hearing ear was abnormal.

Claire’s EEG activity was assessed multiple times after she received her implant. A reorganization of this activity back toward normal was seen by eight months following surgery. After one year, stimulation of either ear elicited cortical activity in both hemispheres, as would be the case in people with full hearing, and cross-modal reorganization was either reduced or reversed to normal. Consistent with this remodeling of the cortex, Claire, when tested two and a half years later, achieved a score of 95 percent correct for understanding sentences presented to her implanted ear. Her tests also showed that she was able to integrate input from both her implanted and normal-hearing ears to improve her speech understanding, and showed that she could localize sound sources in her environment—something that listeners with a single functioning ear cannot do.

The Voice of an Implant

An additional insight provided by single-sided deaf patients is that they allow us, for the first time, to objectively determine the “voice” of an implant. That's because these patients can compare how speech sounds through their cochlear implant with what they hear in their normal-hearing ear.

Consider the video below. In it you hear first a “clean” sentence, which is analogous to the input to the implant processor. Next, you hear that same sentence modulated by different voice distortions. The intelligibility of the sentences is similar, but the voices are very different. At issue is whether any of these voices, or one of many others we have created, when presented to the normal hearing ear, sounds like the output of the implant.

The second video (below) shows what the implant sounds like to Claire. She told our researcher that “It sounds like someone is talking to me from behind a door.” This was our first clue that a slightly muffled voice captured the sound of her implant. We have now tested other single-sided deaf patients and find that, most generally, this sound quality is one characteristic of what is undoubtedly their multidimensional perception of sound.

We do not know whether this same sound quality is also experienced by conventional, bilaterally deaf implant patients. For single-sided deaf patients fit with a cochlear implant, at each moment in time when they are listening to speech, they hear a clean copy of the signal from the normal-hearing ear and a degraded copy from the implant. We wonder whether the signal from the normal-hearing ear teaches, in some sense, the cortex what the signal from the implant should sound like. If this is the case, it is possible that the voice of an implant is different for single-sided deaf patients than for bilaterally deaf patients.

Learning from Patients

For more than 150 years, neuroscience has advanced as new patient populations have become available and have helped researchers find new treatments. We expect that the latest patients to receive a cochlear implant—individuals with single-sided deafness—will continue to afford us their share of new insights into cortical function.