(Inside Science) -- Cochlear implants are powerful tools for people with hearing loss. Using electrodes implanted in the ear that transmit sound directly to the brain, they can give even the profoundly deaf a sense of sound.
But their success often depends on how early the implants are placed. People who are born deaf and receive implants as adults have worse outcomes than those who are fitted with the implants as children, said Andrea Warner-Czyz, an audiologist at the University of Texas at Dallas who studies development in children with hearing loss.
This is at least partly because as people with hearing loss grow older, the parts of their brain that are normally used to process sounds are reassigned to other jobs, such as visual processing. Once these reassignments occur, it is difficult to re-train them to do anything else.
The brains of children, by contrast, are much more flexible, and can adapt quickly to process the signals coming from their implants, so cochlear implants are the most successful when implanted at a young age.
The Food and Drug Administration advises that children should be at least 12 months old before receiving a cochlear implant, but Warner-Czyz wondered whether they would benefit from receiving the implants even earlier.
"We always want to push the envelope, we’re trying to figure out if we get them implants before 12 months, is that going to be better?" she said. Some children do receive their implants earlier than the FDA-approved age.
The question that Warner-Czyz wanted to answer was: Can infants at that age properly process the information from the implants? She and her colleagues explored this question in a study recently published in the Journal of the Acoustical Society of America.
Cochlear implants have two main parts. First, sound is collected by a microphone that sits outside the ear like a traditional hearing aid. The sound signal is analyzed in a speech processor where it is coded for intensity, frequency and duration. Then, the coded signal is transmitted across the skin to an array of electrodes that have been surgically implanted in the cochlea, which is located in the inner ear. That array directly stimulates the auditory nerve to transmit the sound signal to the brain.
But the implants do not reproduce sounds exactly. The implant groups similar sounds into channels, collecting information on the sounds' general frequencies, but losing their finer details. The more channels there are, the more information is provided about the sounds. Typical implants have between 12 and 22 channels. Those are more than enough for adults, who only need eight or nine channels to understand speech. But younger people, whose brains are less developed, need more. Preschoolers need between 12 and 16 channels to reliably understand speech.
So Warner-Czyz wanted to find out whether infants, whose brains are even less developed, would need even more information to distinguish different sounds.
She took a group of 6-month-old children with normal hearing and played them one of two sounds, either "ti" or "ta." Once they became accustomed to that sound, they listened to a series of both sounds; this time, the sounds were played either through a normal speaker, or processed by an audio device known as a vocoder to sound like a 16- or 32-channel cochlear implant, to see if the children could tell the difference.
When the sounds were unprocessed, or played through 32 channels, the children could easily distinguish between the two sounds. But with 16 channels, they could not.
"Infants may need more information than cochlear implants are able to give them at this time," said Warner-Czyz.
Having more channels could help.
"If we can increase the channels, and improve the signal that they’re getting, then maybe we can improve the outcomes for language and speech and hearing for those getting implants at early ages," she explained.
Mario Svirsky, a speech and hearing scientist at New York University in New York City, said that the work tells us important things about how infants understand degraded speech, but he cautions that the standard method of using a vocoder to mimic what a cochlear implant user hears is "woefully inadequate."
"Given how poorly validated, or even downright inappropriate, noise vocoders are as models of cochlear implants, I don’t think the study allows you to draw any conclusions regarding the optimal number of electrodes in cochlear implants for children," he said.
Warner-Czyz acknowledges that the method does not perfectly mimic a cochlear implant, and her findings do not necessarily mean that more electrodes would be needed in the ear. The software that runs the speech processor can be configured to provide more detailed information through "virtual channels."
But most importantly, her work shows that a person’s brain development should be taken into consideration when making decisions about when and how to use implants.
"Right now we’re using a one-size-fits-all strategy for people with cochlear implants, rather than basing it on developmental age," she said.
Brian Owens is a freelance science journalist in St. Stephen, New Brunswick, Canada.