When Chad, a yellow Labrador retriever, moved in with Claire Vaccaro’s family in Manhattan last spring, he already had an important role. As an autism service dog, he was joining the family to help protect Ms. Vaccaro’s 11-year-old son, Milo — especially in public, where he often had tantrums or tried to run away.

Like many companion animals, whether service dogs or pets, Chad had an immediate effect — the kind of effect that is noticeable but has yet to be fully understood through scientific study. And it went beyond the tether that connects dog and boy in public.

“Within, I would say, a week, I noticed enormous changes,” Ms. Vaccaro said of Milo, whose autism impairs his ability to communicate and form social bonds. “More and more changes have happened over the months as their bond has grown. He’s much calmer. He can concentrate for much longer periods of time. It’s almost like a cloud has lifted.”

Dr. Melissa A. Nishawala, clinical director of the autism-spectrum service at the Child Study Center at New York University, said she saw “a prominent and noticeable change” in Milo, even though the dog just sat quietly in the room. “He started to give me narratives in a way he never did,” she said, adding that most of them were about the dog.

The changes have been so profound that Ms. Vaccaro and Dr. Nishawala are starting to talk about weaning Milo from some of his medication.

Anecdotes abound on the benefits of companion animals — whether service and therapy animals or family pets — on human health. But in-depth studies have been rare. Now the Eunice Kennedy Shriver National Institute of Child Health and Human Development, part of the National Institutes of Health, is embarking on an effort to study whether these animals can have a tangible effect on children’s well-being.

In partnership with the Waltham Center for Pet Nutrition in England (part of the Mars candy and pet food company), the child health institute is seeking proposals that “focus on the interaction between humans and animals.” In particular, it is looking for studies on how these interactions affect typical development and health, and whether they have therapeutic and public-health benefits. It also invites applications for studies that “address why relationships with pets are more important to some children than to others” and that “explore the quality of child-pet relationships, noting variability of human-animal relationships within a family.”

The national institutes’ interest in this type of research goes back at least two decades. Valerie Maholmes, who directs research on child development and behavior at the children’s health institute, said that at a broad-ranging meeting in 1987 on the health benefits of pets, the N.I.H. “concluded that there needed to be much more research,” especially on child development.

Other sessions confirmed the need for research, but most studies focused on negative interactions, like the ways pets could spread disease, said James A. Griffin, the institute’s deputy chief of child development and behavior.

Meanwhile, the Waltham Center was expanding its own research to do some small studies about human-animal interaction, said Catherine E. Woteki, global director of scientific affairs for Mars Inc. “We are a pet food company and pet care company,” Dr. Woteki said, “and we’re interested in seeing that that relationship stays a strong one.”

Reviews of the Waltham research program indicated that larger studies over longer terms with appropriate control groups were needed. When Mars became aware of the institutes’ interest in this type of research, a public-private partnership was established, with the company committing more than $2 million. The National Institute of Nursing is also providing money.

Peggy McCardle, chief of the institutes’ child development and behavior branch, said the money from Mars helped jump-start the efforts. Dr. McCardle added that the N.I.H. had established protocols for public-private partnerships and that all proposals got two levels of review before being approved.

People working with animals expect the research to back up their observations. At Children’s Hospital of Orange County in Southern California, for instance, dozens of volunteers regularly take their dogs to visit patients. Children being treated for serious illnesses often have the blues, anxiety or depression. “The dogs brighten them up,” said Emily Grankowski, who oversees the pet therapy program at the hospital.

Some patients who have refused to speak will talk to the dogs, she said, and others who have refused to move often reach for the dogs so they can pet them. So the animals become part of the therapeutic program, especially in the areas involving speech and movement.

“The human-animal bond bypasses the intellect and goes straight to the heart and emotions and nurtures us in ways that nothing else can,” said Karin Winegar, whose book “Saved: Rescued Animals and the Lives They Transform” (Da Capo, 2008) chronicles human-animal interactions. “We’ve seen this from coast to coast, whether it’s disabled children at a riding center in California or a nursing home in Minnesota, where a woman with Alzheimer’s could not recognize her husband but she could recognize their beloved dog.”

Such observations are not new at Autism Service Dogs of America, which brought Milo and Chad together. “Many children with autism can’t relate to a human,” said its director, Pris Taylor, “but they can relate to a dog.”