Suspect they may be deaf
In young children, often the first signs of autism involve differences in perceiving sounds, Mottron said. “Parents sometimes notice that the child doesn’t turn around when called, often leading them to suspect he or she may be deaf.
“Yet hearing tests usually show the child has normal hearing and even heightened sensitivity to non-social sounds. He or she has perfectly good hearing but seems to have little interest in social sounds, especially the human voice.”
Conversely, some specific sounds can be unbearable. Two classic examples: the noise of hand dryers in public restrooms and the sound of a toilet flushing.
Similarly, research over the past decade or so has found that perfect pitch—the rare ability to identify or produce a musical note without an external reference point—is far more common in people with autism than in the general population.
“Some studies have shown that it can be up to 100 times more common,” Mottron said.
“Twenty years ago, my UdeM colleague Isabelle Peretz and I demonstrated that people with autism were also better at determining the frequency of a sound—and they were mainly people who'd experienced language delays in childhood.”
Mottron believes these observations are significant: they point to a different way of prioritizing auditory information. They also raise an important question: what if some aspects of hearing were related to how the brain processes language from the earliest years of life?
Two key dimensions
In their new study, Mottron and his Chinese counterpart examined two key dimensions of hearing.
The first, “spectral processing,” refers to the range of frequencies and intensities that the ear can detect, such as voice timbre or sound pitch. The second, “temporal processing,” involves how the brain organizes sounds over time, something that's crucial for children to pick up syntax from hearing spoken words.
The researchers compared the performance of autistic children with limited spoken language and a control group on two tasks that measured these two dimensions of hearing.
In the first, participants had to detect frequency changes—i.e. identify the point at which a sound rises or falls in pitch. In the second, they had to detect a very brief silence inserted into a continuous sound.
Compared with their neurotypical peers, the children with autism had significantly higher temporal thresholds and significantly lower spectral thresholds—that is, they found it more difficult to detect the gap in the sound but were better at discriminating variations in frequency.
“Autism-related deficits are often reported,” Mottron noted. “But it is very rare in autism research to find such a specific dissociation between a task where people with autism perform better than controls and a closely related task at which they perform worse.”
One important aspect of this two-sided finding is that both temporal and spectral processing scores are linked to linguistic ability: lower temporal and spectral thresholds, indicating higher sensitivity in each domain, were associated with stronger language skills.
Explaining a paradox
Mottron believes that autistic children’s interest in language, combined with limited interest in social sounds such as the human voice, may explain a paradox:
Some rapidly develop reading skills and the ability to recognize letters or numbers—sometimes even in another language—yet show significant delays in expressing themselves through speech.
“The autistic brain seems to prioritize stable, simultaneous information like visual symbols or pitch over the rapid sound sequences of speech,” Mottron said.
He added that people with autism often show a keen sensitivity to patterns, regularities and trends, whether they're spotting visual similarities, subtle sound differences or changes in a daily routine.
“This ability to spot patterns may help explain some traits of autistic language,” Mottron suggested.
“It makes it relatively easy to learn letters, numbers or single words, while syntax, which depends on complex temporal sequences, is harder and requires exposure to and interest in spoken language.”
Based on these findings, Mottron concludes that the autistic brain has the same major structures as a neurotypical brain, but processes information differently. He compares it to a car that runs perfectly but only drives in reverse:
Nothing is broken, it just works differently.
