Playing recordings of a mother’s voice to premature babies may help their brains mature faster, according to the first randomised-controlled trial of this simple intervention. This approach could ultimately improve language outcomes for babies born too early.

Being born prematurely is associated with altered brain structures, which is linked with language difficulties, sometimes compromising later communication and academic achievement. The sound of a mother’s voice and her heartbeat . But it isn’t always possible for a parent to be with or hold their baby in a neonatal ward.

To find out if their presence could be mimicked by a recording, at Weill Cornell Medicine in New York and her colleagues enrolled 46 premature babies, born between 24 weeks and 31 weeks of gestation, while they were in neonatal intensive care.

The mothers recorded themselves reading extracts from the children’s book A Bear Called Paddington. A 10-minute audio clip was then played to half the babies twice every hour between 10pm and 6am, increasing that child’s exposure to their mother’s voice by an average of 2.7 hours each day, until around their original due date. The rest of the babies received the same care, but without the recordings.

Once the babies had reached term, they were given two types of MRI scan, which showed how organised and connected their brain networks were. The scans revealed that those who heard their mother’s voice at night had stronger and more organised connections in and around the left arcuate fasciculus, one of the major areas supporting language processing. It was more mature, says Travis. “Its structure looks more like what we would expect to see in an older or more developmentally advanced infant.”

The scans also suggest that this maturation may be driven by increased myelination – the formation of fatty sheaths that insulate nerve fibres and help signals travel faster and more efficiently around the brain. “Myelination is a key aspect of healthy brain development, especially in pathways that support communication and learning,” says Travis.

Free newsletter

Sign up to The Daily

Sign up to newsletter

Prior studies have and learning difficulties. The new results hint that targeted speech exposure could help improve those outcomes.

But was there anything uniquely important about the babies hearing their mother, rather than anyone else’s voice? This study hasn’t answered that, but previous research has shown how , and continuous exposure to their mother’s voice in the womb is thought to explain why they prefer it over other voices after they are born. “It’s the most familiar and biological meaningful voice for an infant,” says Travis. “Because this voice is so well-established even before birth, it may be especially engaging for the developing brain.”

That said, variability in speech is also important for language development, she says, so it is possible that speech from other caregivers could provide similar benefits. The team intends to explore this idea in future studies.

The intervention is simple and could be easily added into the care system. However, at Evelina London Children’s Hospital warns that the results shouldn’t be overinterpreted. “It’s a very small sample size and I think some more control groups are needed – other sources of speech, other forms of auditory stimulation, other forms of increased stimulation,” he says.

Travis and her team are now hoping to confirm the results in larger trials and in babies who are more medically fragile. They will also follow the current participants to see if the brain differences observed translate into meaningful benefits in language and communication skills as they grow.

Latin inscriptions from the ancient world can tell us about Roman emperors’ decrees and  – if we can read them. Now an artificial intelligence tool is helping historians reconstruct the often fragmentary texts. It can even accurately predict when and where in the Roman Empire a given inscription came from.

“Studying history through inscriptions is like solving a gigantic jigsaw puzzle, only this is tens of thousands of pieces more than normal,” said at the University of Nottingham in the UK, during a press event. “And 90 per cent of them are missing because that’s all that survived for us over the centuries.”

The AI tool developed by Sommerschield and her colleagues can predict a Latin inscription’s missing characters, while also highlighting the existence of inscriptions that are written in a similar linguistic style or refer to the same people and places. They named the tool Aeneas in honour of the mythical hero, who, according to legend, escaped the fall of Troy and became a forebear of the Romans.

“We enable Aeneas to actually restore gaps in text where the missing length is unknown,” said at Google DeepMind, a co-leader in developing Aeneas, during the press event. “This makes it a more versatile tool for historians, especially when they’re dealing with very heavily damaged materials.”

The team trained Aeneas on the largest ever combined database of ancient Latin texts that machines can interact with, including more than 176,000 inscriptions and nearly 9000 accompanying images. This training allows Aeneas to suggest missing text. What’s more, by testing it on a subset of inscriptions of known provenance, the researchers found that Aeneas could estimate the chronological date of inscriptions to within 13 years – and even achieve 72 per cent accuracy in identifying which Roman province an inscription came from.

“Inscriptions are one of our most important sources for understanding the lives and experiences of people living in the Roman world,” says at the University of Exeter in the UK, who wasn’t involved in the research. “They cover a vast number of subject areas, from law, trade, military and political life to religion, death and domestic matters.”

Free newsletter

Sign up to The Weekly

Sign up to newsletter

Such AI tools also have “high potential to be applied to the study of inscriptions from other time periods and to be adapted for use with other languages,” says Tupman.

During testing with inscriptions that were deliberately corrupted to simulate damage, Aeneas achieved 73 per cent accuracy in restoring gaps of up to 10 Latin characters. That fell to 58 per cent accuracy when the total missing length was unknown – but the AI tool shows the the logic behind the suggestions it makes so researchers can assess the validity of the results.

When nearly two dozen historians tested the AI tool’s ability to restore and attribute deliberately corrupted inscriptions, historians working with the AI outperformed either historians or AI alone. Historians also reported that comparative inscriptions identified by Aeneas were helpful as potential research starting points 90 per cent of the time.

“I think it will speed up the work of anyone who works with inscriptions, and especially if you’re trying to do the equivalent of constructing wider conclusions about local or even empire-wide patterns and epigraphic habits,” says at the University of Virginia. “At the same time, a human brain has to look at the results to make sure that they are plausible for that time and place.”

“Asking a general-purpose AI model to assist with tasks in ancient history often leads to unsatisfactory results,” says at the University of Vienna in Austria. “Therefore, the development of a tool specifically designed to support research in Latin epigraphy is very welcome.”

The “dream scenario” is to enable historians “to have Aeneas at your side in a museum or at an archaeological site”, said Sommerschield at the press event. Aeneas is now freely online.