Researchers look to predict developmental difficulties in babies

Brain scientists are experimenting with tests to show in infancy whether a child will later develop autism, trouble speaking, or even problems reading.

Scientists based at Rutgers University in New Jersey report that they can predict with 90 percent accuracy whether a 6-month-old will have speech problems by age 3, simply by measuring how quickly the child's brain processes patterns of rapidly played ''beeps."

In another study, among children born premature, those who were slower at processing information and weaker at remembering it at 7 months old ended up scoring worse on mental development tests when they were 2 and 3, according to work just published by Albert Einstein College of Medicine in New York.

And at Brigham and Women's Hospital, scientists plan within weeks to resume scanning the tangerine-sized brains of peacefully sleeping babies, using state-of-the-art MRI technology capable of capturing even the delicate, shifting pathways that connect neurons.

Though researchers there and elsewhere aren't ready to put their work into widespread practice, in theory, such early diagnoses would allow for early intervention when the brain is most malleable, teaching an autistic child how to connect with others, for instance, and helping another child avoid the stigma of falling behind in school.

Right now, too many of the answers doctors give nervous parents are frustratingly vague, said Dr. Steven Ringer, chief of newborn medicine at the Brigham. The prognosis, Ringer said, is often something like: '' 'Well, he's at risk of problems, and the range runs from normal to very significant, to something in-between.' "

On the downside, though, early brain-based diagnoses could end up giving children labels that later prove inaccurate. Specialists note that the restorative powers of a baby's brain are legendary.

''With the understanding that we have in 2005, we don't know whether the damage [we see] will be there or whether it will be compensated for," said Gary Zientara, a medical imaging specialist at Brigham who is a pioneer in the use of MRI to scan the brains of babies.

The goal is to establish more exactly what the damage they are beginning to detect will mean for the child's future. Their plan is to ''use MRI as like a flag" of possible trouble, said Seung-Schik Yoo, an imaging colleague of Zientara's and Ringer's at Brigham.

Much of the brain research is driven by concern over babies born premature, a record half-million last year, and whose brains take a little extra time developing outside the womb.

MRI scans show that a premature birth damages white matter, the critical connections between neurons. The effect is ''a bit like bombing the railway lines" in war, said Dr. David Edwards, whose newborn intensive care unit at Hammersmith Hospital in London is one of the few in the world to have an MRI scanner adjacent to its nursery.

The work on premature infants also fits into a more general, theoretical picture: The basic building blocks of mental ability, such as processing speed and memory, are already in place and can be tested quite early in life. And if these basic skills are faulty, the trouble can cascade into problems with speaking, reading, and remembering later on.

In the latest issue of the journal Child Development, Susan Rose of Albert Einstein writes that ''specific deficits in attention, speed, and memory can be detected within the first year of life" and ''these very early deficits are critical to general cognitive compromise," that is, difficulty learning, ''at later ages."

In particular, she found that the speed at which a baby's brain processes information seems to be critical.

At Rutgers, neuroscience professor Dr. Paula Tallal and Dr. April Benasich use simple and high-tech methods to measure how quickly a baby processes a series of beeps of differing pitches.

By following the babies for years, Tallal said, they have found that ''the single best predictor of language development across the population is individual differences in how fast the brain can organize simple incoming auditory information" at extremely young ages. That speed was an even better predictor of later trouble than whether a baby came from a family with known language problems.

The Rutgers group works with 6-month-olds, but German researchers have found a similar effect in 2-month-olds, and a Finnish group even found differences in the way 2-week-olds processed sounds, Tallal said.

In an upcoming issue of the journal Neuropsychologia, Benasich and colleagues report that children who processed sounds slowly at 6 months also tended to have less developed language skills at age 2. Such testing ''during infancy, may, in the not too distant future be a useful tool for early screening," she wrote.

The next question, say Benasich and Tallal, is whether a baby identified as a slow processor can be helped to become a faster one.

Benasich's lab is now setting up pilot studies for slower-processing infants, she said, to try ''to nudge the baby very gently onto an optimal processing path" -- by changing the child's focus to sounds it hears around it, for example.

Benasich is also beginning to look at autism, and whether precursors can be detected in siblings of children with autism, by observing the infant's brain waves, behavior, and emerging language, as well as the baby's brain structure as seen on an MRI.

Other researchers have been working on detecting autism far earlier than when its signature lack of social connection-building usually becomes obvious, at 18 months or older.

Autism is very varied, and some suggest that babies who will become autistic show unusual patterns of lying, sitting, crawling, walking, and other movements. Others list early signs of autism as including a lack of response to sounds, an aversion to eye contact, and a failure to learn to point or wave. Scientists also link unusually rapid brain growth to autism. But its basic causes remain a mystery.

Autism is complex, but ''you may be able to see the beginnings of abnormal patterns of processing and/or differences in brain organization if you look early enough and follow the children through to age two or three," Benasich said.

Though young babies are notoriously difficult to study, their brains are less complex than older children's, she said, and ''sometimes you can see differences in how abilities are being set up as they come 'online' that give you valuable clues about what's really happening developmentally."

Carey Goldberg can be contacted at goldberg@globe.com.