Prenatal learning: Do baby food preferences begin in the womb?

© 2009 Gwen Dewar, Ph.D., all rights reserved

Fetuses begin swallowing amniotic fluid at around 12 weeks, and amniotic fluid can take on the odors and flavors of whatever mom last ate. Fetuses begin to respond to odors—which are an important component of flavor--at around 28 weeks (de Vries et al 1985).

So might babies learn about flavors before they are born?

The phenomenon has been documented in rodents and rabbits (Bilko et al 1994; Hepper 1988). And there’s evidence for prenatal learning in humans as well.

For instance, babies recognize the smell of their own amniotic fluids immediately after birth. Given the choice, newborns prefer breasts that have been dabbed with fluids from their amniotic sac (Varendi et al 1996).

But babies lose their preference for the scent of amniotic fluid within a few days after birth. What about long term effects? Can prenatal learning influence infant behavior weeks—even months—after birth?

• Drink carrot juice during the pregnancy and switch to water after the pregnancy

• Drink water during the pregnancy and switch to carrot juice after the pregnancy

• Avoid carrot juice before and after the pregnancy (i.e., drink water only)

The “carrot juice during pregnancy only” condition permitted researchers to test for the effects of flavored amniotic fluid.

The “carrot juice only after pregnancy” condition allowed researchers to test for the effects of consuming breast milk that may carry the flavor of carrots.

More than five months after the babies were born—when the babies were just starting to eat cereal, their first solid foods—the researchers tested the infants’ flavor preferences. They gave the babies two kinds of cereal, plain and carrot-flavored. Would the babies exposed to carrots in utero prefer carrot-flavored cereal? They seemed to.

Compared with their reactions to plain cereal, the prenatally-exposed babies made fewer negative facial expressions while they ate carrot-flavored cereal (Mennella et al 2006).

The results were similar for the babies who had been exposed to carrot-flavored breast milk. But there was no effect observed for babies whose mothers had never drank carrot juice.

Prenatal learning about flavors: Are the implications all good?

Prenatal learning about alcohol

But the effect may have a dark side, too. Studies show that rodents who are exposed to alcohol in utero are more attracted to alcohol-tainted water after they are born. In fact, newborn rats show as much attraction to the odor of alcohol as they do to the smell of their own amniotic fluid (Abate et al 2008).

For obvious reasons, no one has ever done a comparable experiment on humans. But one observational study suggests that babies born to women who drank moderate amounts of alcohol during pregnancy have more pleasurable reactions to the smell of alcohol (Faas 2001).

And several epidemiological studies have found a link between fetal alcohol exposure and alcoholism later in life—even when researchers control for genetics and postnatal environmental factors (see Abate et al 2008 for a review).

So in addition to the dangers that alcohol poses for fetal health and brain development, it appears that prenatal alcohol exposure might give babies a lasting taste for alcohol.

Prenatal learning about junk food?

This makes me wonder how other bad habits might affect baby taste preferences.

A recent experiment on rats suggests that a daily maternal diet of “junk” food--including potato chips and jelly donuts--might affect the food preferences of babies (Bayol et al 2007).

But in this rat study, feeding preferences weren’t affected by the mother’s diet during pregnancy—-just the mother’s diet during lactation.

It should also be noted that the carrot juice study involved the daily consumption of carrot juice.

If women drank carrot juice only 2-3 times a month, would their babies have responded the same way?

We just don’t know. But I suspect that pregnant women eating otherwise healthful diets don’t have to worry that the occasional jelly donut will program their babies to love junk food.

Other clues to prenatal learning: More reading

Studies reveal that newborn babies can distinguish their mothers' voices from those of other women. And experiments have shown that babies are also born with a preference for certain speech rhythms--those matching the language of their mothers. So it seems that fetuses are listening to their mother's voices before they are born.

You can read more about these clues to prenatal learning in my article "The social world of the newborn."

References: Prenatal learning of food preferences

Bayol SA, Farrington SJ, and Stickland NC. 2007. A maternal 'junk food' diet in pregnancy and lactation promotes an exacerbated taste for 'junk food' and a greater propensity for obesity in rat offspring. Br J Nutr. 98(4):843-51.

Bilko A, Altbacker V, and Hudson R. 1994. Transmission of food preference in the rabbit: The means of information transfer. Physiology and Behaviour 56: 907-912.

de Vries JIP, Visser GHA, and Prectl IIFR. 1985. The emergence of fetal behaviour II. Quantitative aspects. Early Human Devel 12:99-120.

Hepper PG. 1988. Adaptive fetal learning: prenatal exposure to garlic affects postnatal preference. Animal Behav 36:935-6

Mennella JA, Jagnow CP, Beauchamp GK. 2001. Prenatal and Postnatal Flavor Learning by Human Infants. Pediatrics. 107(6):E88.

Varendi H, Porter RH, Winberg J. 1996. Attractiveness of amniotic fluid odor: evidence of prenatal olfactory learning? Acta Paediatr. 85(10):1223-7.