Earlier studies have found that fetuses can hear and learn certain sounds. Nursery rhymes, vowel sounds and mothers’ voices can all influence a developing baby. But the new study, published August 26 in the Proceedings of the National Academy of Sciences, shows that a fetus can detect and remember discrete words, says study coauthor Eino Partanen of the University of Helsinki. “The fetal learning capabilities are much more specific than we thought,” he says.
Partanen and colleagues used a fake word, tatata, to test whether a particular word can worm its way into the fetal brain. Five to seven times a week during their third trimester, 17 pregnant Finnish women were instructed to blast a recording of a woman saying the word in two bursts of four minutes. The pregnant women were instructed to turn the volume up so loud that a conversation would be difficult, but not so loud that it hurt. Most of the recording was the same delivery of tatata, but every so often, there was a curveball. The pitch in the middle syllable would change, something that rarely happens in spoken Finnish.
An average of five days after their birth, babies once again heard the recordings. Electrodes attached to the babies’ heads allowed Partanen and his colleagues to look for a specific sign of recognition: An outsized neural jolt, called a mismatch response, tells the brain to pay attention because something is different. This response indicates a level of familiarity, Partanen says.Adults acquire similar neural reactions as they learn a new language, for instance.
When the recording reached the altered version of tatata, babies who had been exposed to the recordings in utero showed this mismatch response, while the 16 babies who hadn’t heard the recordings didn’t, the team found. These results suggest that babies could learn and remember the normal version of tatata.
It’s not clear how long these word memories last. In the study, the babies last heard the recording about five days before the test, but the memory could be older than that.
The study goes beyond earlier work, much of which relied on indirect behavioral changes such as sucking on a pacifier or turning the head, and instead reveals effects in the brain, says psychologist Christine Moon of Pacific Lutheran University in Tacoma, Wash. “We’ve had quite a bit of research on behavior and not so much on the brain,” she says.
The finding has implications for early intervention in kids who might be at risk of language problems, which can accompany certain kinds of dyslexia, says Partanen. Carefully designed words or features of speech played during pregnancy might prove beneficial, he says.
Editor's Note: This story was updated on September 10, 2013, to clarify the amount of time that passed between babies' hearing the recording and undergoing the testing.