Babies who have a high risk of developing autism may benefit when their parents receive some video-based lessons on how to work with their infants, a new study finds.

Researchers found that the babies of parents who completed the lessons were moderately more engaged with other people, did a better job of paying attention and showed more social behaviors, compared with babies whose parents didn't complete the lessons.

The results suggest that although early intervention does not prevent autism, it may lessen its features in some children who have a high risk of developing the disorder, according to the study, published online today (Jan. 21) in the journal The Lancet Psychiatry.

"We preach this idea that intervention changes something in the brain, but we rarely have proof of that," said Mayada Elsabbagh, one of the study's researchers and an assistant professor of psychiatry at McGill University in Montreal. "This is one of the first times in my career that I've seen that so clearly."

In the study, the researchers looked at 54 babies who had an older sibling with an autism diagnosis. Past studies have shown that about 20 percent of such babies will be diagnosed with autism spectrum disorder, and another 20 percent to 30 percent will be diagnosed with other social and communication disorders.

Therapists visited the homes of 28 of the families, and made videos of the parents' interactions with the babies, who were 7 to 10 months old at the time. The therapists then reviewed the videos with the parents, and showed them how to improve their interactions in a way that would help develop the children's attention spans, communication abilities, language development and social engagement. The other 26 families did not receive the visits, and served as a control group.

"The idea is that in autism, very subtle, atypical behaviors of the infant may cause them to be less sensitive and attentive to those social signals from their mothers," Elsabbagh told Live Science. "In turn, mothers are not able to maintain synchrony in the interaction."

For example, a parent may "insist on a given toy or activity that the infant is not interested in, to try to get the infant to respond," Elsabbagh said. "So this is the cycle the intervention focuses on breaking very early on, before the child becomes more and more distant."

Improved prognosis

After five months and at least six of these therapy sessions, the babies in the intervention group were rated better in their engagement, attention and social behavior than those who did not receive the intervention. The parents also improved in their interactions with the babies, the researchers said. In a video clip, for instance, the babies in the intervention group showed fewer signs of early autism-related behaviors, such as not responding to their names, than the babies in the control group did.

"We also showed that the babies had improved their social behavior with other people apart from their parents," Jonathan Green, lead researcher and a professor of child and adolescent psychiatry at the University of Manchester in England, said at a news conference in advance of the release of the findings.

However, the babies in the intervention group did not respond as well to people's language sounds as those in the control group. The researchers said the reason for this finding was unclear, but they noted that the intervention group had more male babies and more babies from bilingual families than the control group, and both of these factors may have contributed the discrepancy, the researchers said.

"We know that these groups develop language more slowly," Teea Gliga, a researcher at the University of London and one of the co-authors, said at the news conference. "So it might not be due to the intervention itself, and may not persist."

The study is the first such randomized therapy trial given to siblings of kids with autism during their first year of life, Green said. The babies are still too young to be assessed for autism — diagnoses typically happen at age 2 or 3 — but the intervention suggests that the plasticity in young brains may help lessen later symptoms related to autism, he said.

The findings need to be replicated in larger studies before other people turn to the video-based therapy as a potential treatment, Green said.

What's more, "We would never want to say that early intervention is the only thing that's needed in autism," he said. "But there's something about early development that might be amenable to intervention."

The research "is an innovative idea and an innovative group," said Joe Piven, a professor of psychiatry at the University of North Carolina at Chapel Hill, who was not involved with the study.

"They tried to intervene, and they had some success," Piven said. "It's moving in the right direction."