Monkeys or Mamas? It's a Toss-Up for Newborns

There’s been a lot of discussion about whether newborn humans recognize their mothers’ voices, or anything else about them, The voice issue has received a great deal of emphasis because babies can hear speech sounds before birth. Even though the sounds are somewhat muffled by the mother’s skin, muscles, viscera, and amniotic fluid, they can be heard and are apparently recognized, as babies after birth have been shown to prefer familiar speech sounds to unfamiliar ones.

In addition to preferring the familiar, newborns prefer voices over other kinds of sounds from inanimate sources. They show this when they have learned that by sucking frequently on a pacifier-like object they can control whether they get to listen to voices or other sounds, or to familiar or unfamiliar voices.

Most researchers working on this topic have not thought about the effect of sounds that are not only unfamiliar, but not likely to be heard at all by most infants in the industrialized world. However, one group looked at whether babies were attentive to hearing rhesus monkey vocalizations (Vouloumanos, A., Hauser, M.D., Werker, J.F., & Martin, A. [2010]. The tuning of human neonates’ preference for speech. Child Development,81, 517-527). They were interested in learning whether newborns were really especially attentive to speech, or whether they preferred all sounds with characteristics they share with human speech-- like the sounds of monkey vocalization. These researchers let newborn babies hear nonsense speech and also recorded vocalizations of rhesus monkeys. The neonates were equally likely to choose (by sucking at the right time) to hear human nonsense speech or to hear monkey sounds. They chose both of these over synthetic sounds.

It seems that the newborn humans could not, or at least did not, tell the difference between humans speaking and monkeys vocalizing. But adults and even older children can readily make that discrimination. How and when do they manage this? Vouloumanos and her colleagues tested 3-month-old babies, using a different technique than the one they had used with the newborns. The 3-month-olds could keep a sound playing by looking at a black-and-white checked screen for a long time; if they looked away, the sound stopped; when they looked back again, one or another sound occurred at random. Unlike the newborns, the 3-month-olds spent more time looking in a way that would allow them to hear human speech sounds. They not only could, but did, tell the difference between human and monkey sounds.

How does this work compare with the research (by DeCasper, for instance) suggesting that newborns can make very careful discriminations, not only recognizing their mothers’ voices but even telling the difference between stories that were read? How could they tell these rather subtle differences, but not discriminate human and monkey vocalizations? Is a puzzlement, yes?

Vouloumanos and her colleagues suggested that this peculiar difference might result from the fact that the nonsense speech they used was quite brief and was not the connected speech of ordinary use. It may be that spoken sentences make available to babies a good deal more of the distinctively human speech characteristics that differ from those of monkey vocalization. Connected speech contains a variety of information in the form of syllable length, intonation and emphasis, rhythm, and so on. This rich, variable material may be much more effective in catching infants’ attention than brief nonsense words. In other words, the researchers speculated that the newborns actually could tell the difference between human and monkey sounds, but did not pay much attention to what they did not find very interesting.

Whether or not newborn babies can tell monkeys’ voices from mamas’ voices-- and a definitive answer is not yet available-- it’s clear that they do discriminate between voice sounds and synthetic sounds, and spend more time listening to voices when they get the chance. They are born with a positive bias, an interest in the sounds voices make. Is this because of prenatal experience with voices, because of a built-in “tuning” of the auditory system for those important sounds, or because of a combination of the two? That isn’t clear yet, but it is clear that newborns are still very much in the process of development, and adults should not jump to any conclusions about resemblances between the very young and the older human being.

Just a little addendum: when Vouloumanos and her colleagues did one of these experiments, they tested 30 babies. Another 44 babies were available but were excluded from the study because they fussed or fell asleep or because they did not suck often enough. If you ever wonder why research on young infants is not done more often and with larger numbers of babies,do remember this: just as you can’t make a turtle come out, you can’t make a baby participate if it doesn’t feel like it.