CHICAGO -- A tide of recent research on early childhood development is inspiring prominent scientists and politicians to argue for an unprecedented investment in schooling that begins virtually at birth.
But as decades of academic studies on brain development start to land in the real world, experts are divided on whether to focus new funding on infants and toddlers, or conventional preschool.
Many now think that some policies popular with politicians and the public, such as universal pre-kindergarten, may fail to reach at-risk kids at a young enough age to achieve significant benefits.
The scientific controversy is also spilling into the presidential contest, in which the Democratic candidates have taken divergent positions on universal preschool and other early childhood issues.
Studies have suggested that intervening before children start preschool improves academic outcomes for low-income kids and may reduce the risk that they will end up in prison.
Such interventions stem from the theory that experiences in the first five years of life set a lifelong course for brain development.
Chicago has become a national proving ground for schooling during the first three years, and is home to prominent advocates such as Nobel Prize-winning economist James Heckman of the University of Chicago, who said reaching kids before preschool could offer the best long-term economic return.
"Even at age 4 or 5 you may be starting too late," Heckman said.
"I wouldn't say it's hopeless to help kids after those early years, but it's extremely expensive."
Backers of universal preschool say the evidence for even earlier intervention is not yet solid and offering conventional pre-kindergarten to everyone would help build popular support for early education.
In theory, starting to intervene soon after birth should help kids more because that's when experience starts to shape their brains, many experts said.
Children's brains change more between conception and kindergarten than at any other time.
University of Chicago neuroscientist Peter Huttenlocher has showed in studies over the past 30 years that connections between cells in most brain areas peak by age 3, then decline gradually as experiences mold the brain's circuitry.
The zero-to-3 period is not necessarily a magical and irreplaceable window for teaching children.
But studies show that babies raised in poverty get fewer of the early experiences that spur vocabulary growth and good social judgment, making it harder for them to catch up later on.
For example, toddlers whose parents speak more words to them develop bigger vocabularies than children who hear less speech, studies have found.
One University of Kansas study concluded that kids from upper-income backgrounds hear 30 million more words by age 3 than those from welfare families.
Early intervention with enrichment programs can narrow that gap, researchers and advocates say.
"The basic science of brain development says you need to start as early as possible for kids in the greatest danger to get the best outcomes," said Jack Shonkoff, director of the Center on the Developing Child at Harvard University.
That involves more than just listening to adults read stories. At the Educare Center on Chicago's South Side, which serves infants and toddlers from low-income families, caregivers engage the children in a constant dialogue about their activities -- a type of interaction kids cannot get from television or even books.
The staff also talks to parents, often first-time teen mothers, about how to give consistent rewards and punishments, which can foster a child's emotional development.
"If you can't regulate your emotions you can't get anything done," said Diana Rauner, executive director of the Chicago-based Ounce of Prevention Fund, which runs the Educare Center.
Such programs can change children's prospects dramatically.
Arthur Reynolds, director of the Chicago Longitudinal Study, has published numerous studies this decade on kids who were born in 1980 and went through Chicago's Child-Parent Centers, which offer educational services and family support for low-income children ages 3 to 9.
Reynolds found that these children scored better on math and word analysis tests and were more likely to finish high school than those who lacked that early experience. The benefit was greatest for kids who took part by age 3 or 4.
"There's a big critical mass of evidence that preschool for 3- to 4-year-olds makes a difference," Reynolds said.
Some experts think programs for even younger kids would help more, but Reynolds disagrees, noting that few studies have evaluated the effects of zero-to-3 programs later in a child's life.
Still, Shonkoff thinks the huge amount of brain development that happens from birth to 3 makes it crucial to intervene during that window.
"If you start at age 4 for kids who are at a disadvantage, you're not starting early," he said. "You're playing catch-up."
One of the best cases for starting at birth comes from the Abecedarian Project, a 1970s enrichment program in North Carolina that enrolled 111 low-income, African-American infants.
Children in the program did better on reading and math tests, were more likely to attend college and were less likely to have babies at an early age than others.
Bruce Fuller, a professor of education and public policy at the University of California-Berkeley, said he feared that focusing on universal pre-kindergarten, making preschool a middle-class entitlement, could divert assistance from the low-income families who need it most.
"Why would we use scarce public dollars to subsidize all families if we know the biggest impact is with poor kids?" he said.
The notion of intervention before age 3 even draws qualified support from author Charles Murray, who has argued that Head Start programs in general do little to improve children's outcomes.
Murray is skeptical about any government-run intervention, but said early investment makes the most sense.
Still, he said, "There is no way on God's green earth that these select programs can be expanded to a national level."
Jeremy Manier writes for the Chicago Tribune.