All parents want their kids to succeed in school, but in the U.S. there has been a persistent achievement gap between kids from lower and higher income families. Programs like Head Start are designed to address that divide — giving kids from low-income homes access to the kind of structure and experiences nursery schools can provide and that make starting school a little easier.

Children who take part in these early childhood programs not only benefit from them in preschool, but years later, a study from Penn State shows.

Over the years kids in Head Start-type programs not only did better academically, they were also more advanced in social-emotional growth, for instance, being better able to understand the feelings of others.

The team set out to understand how the Head Start REDI (Research-based, Developmentally Informed) program and a related at-home parenting program designed by the team would help kids from low-income families over the years. The home-based program was geared toward building social and emotional skills in the kids as they played imaginative games with their parents.

“Essentially, our storybooks and conversation games give parents an easy way to get their children talking with them about their feelings and their social experiences,” study author, Karen Bierman, said in a statement. “These conversations show the children that their parents are someone they can talk to about feelings and problems.”

Researchers made 10 visits to the families’ homes during preschool and six during kindergarten to provide in-person coaching. Parents were encouraged to play games like “restaurant”, with kids taking their parents’ “orders” and serving them pretend foods. Games like this allow kids to practice letters, numbers and interpersonal skills.

The team tracked the academic success and social-emotional functioning of the kids through third grade, following 200 children who’d taken part in the REDI program previously. Half the kids were then enrolled in the at-home parenting program, and the other half received a math skills program by mail, which served as a control group.

It turned out that, over the years, the kids in the parenting group did better academically than those in the control group. Most importantly, they were also more advanced in social-emotional growth, for instance, being better able to understand the feelings of other kids. They also needed fewer additional services like learning support, special education and mental health counseling. Parents also reported that the kids who were part of the program had fewer problems at home.

“This suggests that the program pays off over time,” Bierman said. “If you strengthen that parent support right at the transition from preschool to kindergarten, children are more able to navigate both the academic and social demands of school over time. Also, when children are bringing problems home, parents may be more well set up to support their children with those issues.”

Other studies have suggested similar benefits for Head Start-style programs, but typically haven’t followed the kids for so long. The team says they’ll track the kids’ progress for even longer, to see how the results hold up over time. They’re also experimenting with using technology to teach some of the methods that are currently taught in-person, to cut down on the costs of implementing the program for school districts.

The big message of this study is the importance of playing with your kid, whether they are in a preschool program or not. Games of make-believe can teach an awful lot about language, numbers and feelings, all of which will serve them well as they grow up, both in and outside of school.

The study is published in JAMA Pediatrics.