ROBERT SIEGEL, host:
From NPR News, this is ALL THINGS CONSIDERED. I'm Robert Siegel.
MICHELE NORRIS, host:
And I'm Michele Norris.
No blood test or brain scan can identify a child with autism but scientists are trying to change that, using technology that can detect unusual electrical signals in the brain. The approach could help diagnose autism earlier.
And as NPR's Jon Hamilton reports, it's also providing hints about what therapies are likely to help.
JON HAMILTON: Research shows that kids with autism do better if they start getting therapy before age two. But Charles Nelson, a professor of pediatrics at Harvard, says that rarely happens.
Professor CHARLES NELSON (Children's Hospital Boston, Harvard Medical School): Right now, the earliest we can reliably identify a child is, say, three years of age. We've now lost three years of time to do an intervention. Our work is designed to see, can we do that in early infancy, long before any signs or symptoms of autism are apparent in the child's behavior?
HAMILTON: Nelson is part of a team that's studying the brains of autistic children using electroencephalography, or EEG. The technology records the electrical signals produced by brain cells. Nelson says kids wear a sort of hairnet that holds several dozen electrodes against their scalp.
Dr. NELSON: Once that's on, the child can either sit on a parent's lap or sit in a chair by themselves if they're old enough.
HAMILTON: While a computer analyzes the signals coming from their brain.
William Bosl, of Children's Hospital Boston, says the goal has been to find some kind of patterns in those signals that indicate autism.
Dr. WILLIAM BOSL (Pediatrics and Neuroscience, Children's Hospital Boston, Harvard Medical School): So what we're looking for are markers or features in the signals that tell us something about the brain as it develop and differences in development.
HAMILTON: And they've found at least one promising marker. It involves very high frequency brain waves known as gamma oscillations. Bosl and Nelson have been studying those oscillations in about 80 children from the time they were six months old.
Nelson says some of the kids are at high risk for developing autism because they have an older sibling with the disorder. The rest have no special risk factors.
Dr. NELSON: What we've observed is that starting as young as six months, maybe even younger, infants who have a high risk for developing autism show dramatic reductions in gamma activity. So they show less gamma activity than do children who have no risk or a very low risk for developing the disorder.
HAMILTON: That suggests EEG could help identify very young children who will go on to develop autism. But the researchers won't know for sure, until the children in their study reach the age when the disorder becomes obvious.
In the meantime, other research using EEG is revealing how autism affects the brain and how intervening early might make a difference. Sophie Molholm, of Albert Einstein College of Medicine in New York, has been studying how autism impairs the brain's ability to integrate information from the senses. She says a typical brain does this with ease.
Professor SOPHIE MOLHOLM (Cognitive Neurophysiology, Albert Einstein College of Medicine): We walk through the world and our sensory systems are continuously presented with stimulation: visual stimulation, auditory stimulation, touch, smells. And we very automatically put this information together to make sense of the world and to navigate it.
HAMILTON: For example, when a person is listening to someone at a noisy cocktail party, they'll focus on that person's face. And their brain will automatically combine what they see with what they hear.
But Molholm says people with autism often have trouble doing this.
Prof. MOLHOLM: So we set up a very simple version of this where we present words and noise.
HARRIS: Words like bear.
(Soundbite of static)
Unidentified Woman: Bear.
HAMILTON: Molholm played the words for children with autism, as well as typically children.
At times, the audio was accompanied by a video of person saying the word, at other times it wasn't. Molholm wanted to know whether the brains of children with autism react differently than did other kids' brains. And they do.
Prof. MOLHOLM: What we find is that children with autism don't benefit from the visual signal nearly as much as do typically developing children.
HAMILTON: And research using EEG indicates that's because their brains are having trouble integrating what they're seeing with what they're hearing. Molholm says the surprising thing is that the brains of kids with autism seem to get much better at this integration when they become teenagers.
Prof. MOLHOLM: The beauty of that finding is it suggests that the neural circuitry to do this is there, it's intact. But it's just not coming online during the earlier childhood years.
HAMILTON: Molholm says that neural circuitry might come online sooner if these children were diagnosed early and encouraged to practice watching faces while listening to words.
Jon Hamilton, NPR News.
NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.