Now, scientists at the University of Florida have added a piece to the puzzle. In a series of unique experiments on a pregnant ewe designed to record exactly what sounds reach the fetal ear, UF research has bolstered previous findings suggesting that human fetuses likely hear mostly low-frequency rather than high-frequency sounds. That means they hear vowels rather than consonants and are more sensitive to the melodic parts of speech than to pitch, said Ken Gerhardt, a UF professor of communication sciences and disorders and an associate dean of the Graduate School.
As for music, "they're not going to hear the violins, but they will hear the drums," said Gerhardt, who led the research reported in the November-December issue of the journal Audiology and Neuro Otology.
Anthony DeCasper, a professor of developmental psychology at the University of North Carolina-Greensboro, said the UF results are noteworthy because they were obtained from the inner ear, which presumably would register sounds exactly as a sheep fetus would hear them. The findings - which resulted from implanting a tiny electronic device in the inner ear of a fetal sheep that tapped into the signal the ear sends to the brain - dovetail with what other researchers have concluded based on less invasive studies involving human fetuses, he said.
"The way I put it is, the way the mother's voice would sound in utero would be like Lauren Bacall speaking from behind a heavy curtain," DeCasper said.
The research also may have implications for the care of premature babies. Neonatal care units are traditionally relatively noisy places, replete with beeping machines and the hum of conversation, and research by Gerhardt, DeCasper and others is raising awareness and questions about how the sounds "premies" are exposed to may influence normal growth and development, researchers said.
Gerhardt said he began researching fetal hearing in response to inquiries from law enforcement agencies and the U.S. Navy. Both were concerned about how loud noises, such as gunfire or the rumble of a ship's engine, could affect the hearing of babies being carried by pregnant servicewomen.
Gerhardt and Robert Abrams, a UF professor emeritus of obstetrics and gynecology, decided to study the issue using sheep because experiments by other researchers had shown that properties of sound transmission in pregnant women and sheep are similar. Their experiments revealed the womb dampened all but the loudest sounds. Even loud rock concerts are probably not noisy enough to pose a threat to fetal hearing development, Gerhardt said. The research, which has received more than $1 million from the Navy, the National Institutes of Health and the March of Dimes, contributed to federal workplace safety guidelines that today limit the duration of extremely loud noise exposure for pregnant women.
In their recently reported research, Gerhardt and Abrams focused specifically on the kinds of sounds fetal sheep hear. Their method was unique: They implanted an electronic "pickup" inside the inner ear of a fetal sheep, then played 64 recorded sentences on a loudspeaker in the open air near the mother sheep. The sounds the pickups detected were recorded and played back to 30 human adult listeners in order to determine how much or what portions of the sentences the fetus actually heard. For comparison, the researchers also placed a microphone inside the uterus of the ewe and in the open air, then performed identical tests with human listeners.
The results were surprising, with the intelligibility of sentences "actually much higher than we anticipated," Gerhardt said.
In part, that was unexpected because much of the noise that reaches a fetus comes from its mother's own voice, movement, breathing and digestive processes. Even in a quiet room, the womb can be noisy place, he said. Also, fetuses don't "hear" as much with their ears as children or adults do because their ears are filled with fluid, he said. Rather, much noise is transmitted to their inner ears through vibrations in their skulls. As a result, the mother's voice tends to be the most dominant and recurring sound in the womb.
The listeners understood all of the sentences recorded in the open air, about 70 percent of the sentences recorded in the womb and about 30 percent of the sentences recorded in the fetal sheep's inner ear.
The recordings revealed the reason the inner ear-recorded sentences proved so much less intelligible was that the higher-frequency consonants in words tended to be absent or confused. In other words, "ship" could easily be heard as "slit" or "sit." Lower-frequency vowels, by contrast, tended to penetrate the inner ear to a much greater extent.
Charlene Krueger, a UF assistant professor of nursing who works with premature babies, said that while a fetus hears its mother's voice whenever she speaks and is insulated from higher frequencies, infants born prematurely don't hear their mother's voices all the time because their mothers usually can't be at the bedside 24 hours a day. Also, while unborn children are insulated from many frequencies, premies are exposed to all the frequencies of the sounds in the nursery. She is interested in finding out how this added exposure, combined with the loss of the mother's voice, might influence the premies' development.