Getting kids to like vegetables can feel like a losing battle. A small new study suggests that battle may start before birth.

Researchers led by Durham University and Aston University in the UK found that children exposed to specific vegetables through their mothers’ diet during pregnancy were less likely at age three to react negatively to the smell of those same foods.

The findings suggest that likes and dislikes formed during gestation can last into early childhood.

“What we see over time is that the children are still more favorable to the vegetables they were exposed to while they were in the womb,” says psychologist Nadja Reissland, from Durham University.

“From this we can suggest that being exposed to a particular flavor in late pregnancy can result in long-lasting flavor or odor memory in children, potentially shaping their food preferences years after birth.”

The children in the study had previously been exposed to carrots and kale through their mothers’ diets. At 32 and 36 weeks’ gestation, researchers used ultrasound to record fetal facial responses after mothers consumed a high-potency capsule containing either kale or carrot powder.

A 2022 study reported evidence that the unborn babies reacted to those tastes and smells.

By the third trimester, about 28 weeks into pregnancy, humans are developed enough to taste and smell flavors in the amniotic fluid surrounding the fetus, which passes through the placenta from the mother.

In the new study, researchers tested if those reactions lasted into childhood.

At age three, 12 children were presented with the odors of both kale and carrot on cotton swabs. Researchers recorded their reactions as either “cry face” for negative or “happy face” for positive.

Overall, the three-year-olds showed fewer negative reactions to the vegetable they had been exposed to in the uterus than to the other one.

Kale prompted more negative reactions overall, but early exposure appeared to reduce them.

The researchers said the study has limitations, including its very small size and the fact it included only one population.

The children did not actually eat anything or choose what to eat, and researchers did no monitoring between birth and age three to rule out the effect of later exposure.

Still, the team said the results support further research with larger groups of children and more detailed follow-up over time.

“It is important to note that flavor preferences and dietary habits are influenced by a complex interplay of genetic, environmental, and cultural factors,” the researchers write in their published paper.

“For example, genetic differences among individuals play an important role in how various foods and odors are experienced. Therefore, genetic predispositions, causing various levels of sensitivity for flavors, may change the importance of prenatal exposure effects.”

“These findings give further support to the use of prenatal exposure for improving children’s acceptance of often disliked vegetables,” says psychologist Jacqueline Blissett, from Aston University.

The research has been published in Developmental Psychobiology.

Read more from Science Alert.