Toddlers can be the harshest of food critics. No matter how much care and attention a parent has put into cooking a meal, a two-year-old may still turn up their nose and refuse to eat. Many are particularly likely to reject new and unfamiliar foods, with vegetables often top of the list.
Frustrating though this is for parents, they can draw some comfort from the fact that toddlers are not the only fussy eaters in the world. Many young animals are cautious about trying new foods.
It seems bizarre at first. Food is often short, so why would a toddler or young animal refuse to eat when it is available? A full explanation is still some way away, but biologists have developed several ideas that can help to explain why.
Lucy Cooke, a psychologist at University College London and Great Ormond Street Hospital, has been studying children's eating behaviours for over 15 years.
A child who is wary of something new might have a broad repertoire of familiar foods they are perfectly happy to eat
She draws a distinction between two forms of fussy eating in young children. On the one hand there are picky eaters – children who have a limited range of accepted foods. On the other hand, there are children who are "neophobic" – avoiding any foods that are new or unfamiliar.
Cooke says there is endless debate over whether or not food neophobia and picky eating (or "food fussiness") are connected. Logic suggests they might not be. Food neophobia is a general problem that is seen in essentially all the children she works with – and is, in fact, a trait shared to some degree by most people – whereas a more general food fussiness seems to be a less common additional problem seen in some children.
"If you have a very picky child, with a very restrictive diet, they are almost by definition going to be neophobic," she says. "But flip that around, and a child who is wary of something new might have a broad repertoire of familiar foods they are perfectly happy to eat."
Cooke is a member of a research team that is exploring the subject by studying a group of twins born in England and Wales in 2007.
The parents of the children in the study – called Gemini – regularly complete questionnaires about their children's eating habits. This gives the researchers an opportunity to explore whether traits like food neophobia and food fussiness are down to nature or nurture: in other words, whether they are controlled by genes or by a child's home environment.
"Certain things are genetically determined and don't differ between people. We are almost all born with two arms and two legs, for instance," says Clare Llewellyn at University College London, another Gemini researcher. "But other things are variable: differences in height, weight, eye colour. Twin studies can tell us whether those differences are shaped by different environmental factors, or whether genetic differences between people play a role."
Some children may be born with an innate fussiness about food
The studies take advantage of the fact that identical twins share 100% of their DNA, whereas non-identical twins share just 50%. If a trait – like food neophobia – is more likely to be seen in two children who are identical twins than in two children who are non-identical twins, it hints that the trait is controlled to some extent by genes.
One paper the Gemini research team published in October 2016 made the headlines. Many of the reports missed the main message of the paper, which was that food fussiness and food neophobia are strongly correlated and might, after all, be caused by common factors.
But a broader message from the study did get through: the research had uncovered evidence that both forms of fussy eating are, to some degree, genetically determined. To put it another way, some children may be born with an innate fussiness about food, and parents are not necessarily to blame if their child becomes difficult at mealtimes.
The results were actually a little more complicated than that.
Genes do play a role in both food neophobia and food fussiness, but the study suggested genetics are not an overwhelming factor in determining a child's attitude to eating. "Food fussiness heritability was 46%," says Llewellyn. "Heritability of neophobia was a bit higher, about 58%."
This means a child who inherits "fussy" genes will not necessarily become a fussy toddler. Their upbringing and experiences can have a roughly equal impact on their eating habits.
Our ancestors gained a substantial proportion of their diet from plants
Dig deeper into the literature and things become even more complex.
The Gemini twins were 16 months old when their eating habits were explored for the 2016 study. But earlier research projects have looked at fussy eating in groups of slightly older twins, and they tell a different story.
For instance, in 2013 a research team studied food neophobia – the more common form of picky eating – in four- to seven-year-olds. They found the heritability at this age was almost 72%, which is far higher than in the 16-month-old Gemini twins.
This suggests the genes that influence a child's tendency to food neophobia gradually exert a greater influence as children grow older. "There is quite a profound increase in the genetic effect," says Llewellyn.
Why might that be? One longstanding idea offers an explanation.
Humans are, traditionally, a foraging species. Our ancestors gained a substantial proportion of their diet from plants. Unfortunately, plants – particularly those in the tropics, where our species evolved – often contain toxins.
It would be evolutionarily advantageous for those children to suddenly become far more wary of trying new foods
This left our ancestors facing a problem. They had to sample as many different plants as possible, to build a broad diet and increase their chances of survival. But they had to balance that willingness to try new plant foods with wariness, because those leaves, stems and roots might turn out to be highly toxic. This is sometimes called the omnivore's dilemma.
A child should not experience the omnivore's dilemma in their first year or so. At this stage, parents are carefully selecting all of the food a child eats, so in theory, the child should be willing to eat anything it is given.
But children gain more independence from about 24 months.
They can toddle around and explore their environment and the food it contains. It would be evolutionarily advantageous for those children to suddenly become far more wary of trying new foods, just in case the food was dangerously toxic. Consequently, it would make sense for the genes that influence food neophobia to become more active at this point in childhood.
Evolution should have "favoured" children who gobble down things like peas and broccoli with relish
Arguably, this attitudinal shift towards food should be really rather dramatic.
Adults have mature biological systems for dealing with toxins in their food. Eating the wrong plant might lead to an unpleasant stomach upset, but probably will not threaten their lives. But in general, youngsters lack those systems, so eating the wrong thing at this age could result in death. It would make sense for the omnivore's dilemma to be at its most acute in toddlers – and almost non-existent in infants.
These ideas offer explanations for some of the problems parents may have to deal with at dinnertime. Most obviously, they could help explain why children often refuse to eat their greens.
It seems baffling why so many children have a problem with green vegetables. These foods are very nutritious, so evolution should have "favoured" children who gobble down things like peas and broccoli with relish.
Food neophobia seems to kick in when children are between their second and third birthdays
But perhaps over the long expanse of human evolution it has been the omnivore's dilemma – not the nutritional value of vegetables – that has played the leading role in deciding which children survive to adulthood and pass on their genes.
In line with this, most of the vegetables on the dinner table today have been selectively bred to be far more nutritious, and far less toxic, than they would have been in the past. A child's reluctance to eat their greens might be a hangover from a time when those foods were less nutritionally beneficial, and more likely to be dangerous.
These ideas provide a neat and satisfying explanation for picky eating among toddlers. But are they actually supported by evidence?
Elizabeth Cashdan, an anthropologist at the University of Utah in Salt Lake City, explored the subject in a study published in 1998. At the time, she found hardly any formal scientific studies on whether infants really are willing to eat anything, before turning into recalcitrant toddlers. So she began collecting data herself.
The single greatest cause of poisonings among six- to 18-month-old children was eating houseplants
Her findings were, indeed, broadly in line with the ideas of the omnivore's dilemma. Food neophobia seems to kick in when children are between their second and third birthdays, and to get worse by the time children hit their fourth birthday. This is in keeping with the recent twin studies, which suggest genes begin to exert a greater influence on a child's eating habits sometime between the ages of 16 months and four years.
Cashdan also found that the threat plants pose to young children, even in today's developed societies, should not be underestimated.
She talked to the Utah Poison Control Center, a medical facility that offers advice to Utah residents about exposure to toxic substances. All sorts of potentially toxic cleaning products and medicines are kept in the typical home. However, Cashdan was told that the single greatest cause of poisonings among six- to 18-month-old children was eating houseplants.
Of course, if there really is an acute omnivore's dilemma early in childhood, it should be seen in other omnivorous and herbivorous species, not just humans.
Cashdan turned to the scientific literature. She found one report from the mid-1980s into the feeding behaviour of young gorillas that was particularly significant. Infant gorillas would try anything, the report suggested, but slightly older gorillas were far more wary of new foods.
If you put a Brussels sprout in front of a child, they'll pick it up and throw it around
Another study published at roughly the same time suggested the same is true of rats.
Even so, Cashdan says the evidence is still rather limited. The omnivore's dilemma makes intuitive sense and is supported by some evidence, but until researchers test the idea thoroughly, we cannot be sure it is the only factor controlling food fussiness.
Nicola Marples at Trinity College Dublin in Ireland thinks there is more going on.
She has studied a variety of animal species, including carnivores as well as omnivores and herbivores. Her findings suggest that all backboned animal species include some individuals that seem to want to stick to a narrow range of familiar foods throughout their lives.
"We have even seen it in fish," says Marples. "It does seem to be something that's utterly fundamental."
Animals like honeybees, which have painful stings or even lethal venom, often have bright and contrasting colours
Feeding a fussy toddler
A few tips from the experts
Is there anything parents can do to encourage their children to go against their instincts and try new foods? Clare Llewellyn at University College London has explored the problem through her research and suggests some pointers.
"Never, ever use food as a reward," she says – promising a child a slice of cake if they eat their broccoli, for instance. "That just makes the disliked food even more disliked, because the child thinks they need a reward in order to eat it."
It also has the unintended consequence of making the cake seem even more desirable.
"Parents can offer non-food rewards instead – stickers work well. Praise is very good too," she says.
There is also good reason to believe, from scientific studies, that eating when a toddler eats – and ideally sharing the same meal – might help the child overcome his or her reluctance to eat.
Llewellyn also urges parents to persevere if their toddler is particularly stubborn. If the child rejects a new food, Llewellyn recommends offering it again at the next meal. If it is again rejected, offer it at the next meal, and keep going.
"Our research indicates you need 15 exposures on consecutive days for a child to try a new food," she says.
She thinks such behaviour is often better described as "dietary conservatism" rather than food neophobia or fussiness. By definition, food neophobia should fade quickly: if a young animal sees the same novel food item over and over, it should lose its reluctance to eat it. But some animals – and some young children – maintain their reluctance to eat a new type of food long after it has lost its novelty factor.
What's more, "neophobia" implies fear, and Marples's studies – and her anecdotal experiences with children – suggest fear is not a factor in an animal's reluctance to eat certain foods.
"If you put a Brussels sprout in front of a child, they'll pick it up and throw it around," she says. "They're not scared of it. They're just not prepared to eat it."
As long ago as 1995, Marples and her colleagues had found evidence that dietary conservatism is heritable, suggesting it is governed to some degree by genes. The team selectively bred quail that were unwilling to eat an unfamiliar form of ladybird beetle. The birds' descendants became even more reluctant to eat the insects.
Marples says that dietary conservatism may even provide an explanation for one of the most enduring mysteries in evolution: why some plants and animals "advertise" their unpalatability with bright body colours.
Animals like honeybees, which have painful stings or even lethal venom, often have bright and contrasting colours. Yellow and black is a common combination. Predators are often wary of these creatures. This phenomenon of animals displaying warning colours is called "aposematism".
They're not scared of it. They're just not prepared to eat it
The trouble with aposematism is that it is difficult to work out how it can ever become established.
There seems to be a paradox at work. The first individual in a species that evolves a brightly-coloured body should be an easy target for a predator. Even if the brightly-coloured individual carries a toxin or an unpleasant taste, it is difficult to see how it could possibly survive and reproduce, let alone how the bright body colour trait could ultimately spread through the entire population.
Dietary conservatism offers one solution to the paradox. A predator might deliberately avoid eating the first brightly-coloured animal, precisely because it is so unlike its drab peers. As far as the predator is concerned, the brightly-coloured animal is something new, and eating something new is risky.
In a 2004 study, Marples and her colleagues performed lab experiments that show how dietary conservatism could indeed allow aposematism to become established.
They trained European robins to eat pastry "prey" that had been dyed brown; the colour of many of the robins' natural prey. Next, the researchers introduced a new, bright red form of the pastry prey. At first, the robins were presented with just one red pastry among a "population" of 20.
In any given population, it is only some animals that avoid eating new food
If the robins ignored the red form, Marples and her colleagues considered that it had survived and could "reproduce". So the next time the robins were fed, there would be more red and fewer brown forms among the 20 pastry prey.
Some robins so stubbornly refused to eat the red forms of the pastry that, over the course of several days, the red form gradually replaced the brown form. Eventually all 20 pastry prey items were red.
Follow-up experiments showed the same was true whatever colour combinations the team used. Even if the starting point was a population of brightly-coloured pastries containing a single brown one, the robins still shunned the solitary new colour form.
If something similar plays out in nature, aposematism might be easier to evolve than some biologists assume. "It's one explanation for how you get out of the paradox – although it's not the only one," says Marples.
Marples has also found that, while dietary conservatism is seen in every species she has studied, it is not seen in every individual. In any given population, it is only some animals that avoid eating new food. Other animals are far more adventurous.
It might be a bad thing for us to try to train out one of those traits
"We chose the name 'adventurous consumer' for these individuals, to contrast with 'dietary conservatism'," says Marples. "That meant we could talk about animals being 'AC' or 'DC'."
Most animal populations seem to contain both AC and DC individuals. ACs generally account for about two-thirds of animals in the population with DCs making up the other third, says Marples. She has no idea why both strategies co-exist, and why the ratio between the two is generally 2:1.
But the fact that these features seem to be universal makes her cautious about efforts to deliberately change a toddler's eating habits – to turn children, in her words, from DCs into ACs.
"Maybe it is more stable for populations to contain both DC and AC individuals," she says. "It might be a bad thing for us to try to train out one of those traits, until we've studied this and know more about why they co-occur. We might miss out on something important by insisting that everyone is AC."
It is worth bearing in mind that Marples's ideas have yet to turn mainstream. "This is still relatively young science," she says. "It takes a long time for other people to pick up on new ideas like this."
Many concerned parents might prefer to put her words of warning to one side and encourage their fussy child to broaden his or her diet. But how should they do so? Perhaps animal studies can offer some hints. After all, if the omnivore's dilemma – and food neophobia – is common in the animal kingdom, young animals must somehow extricate themselves from its effects.
Many researchers think that young animals must work out what is safe to eat through observation. Assuming they live in a social group that includes adults, a youngster might watch carefully to see which plants its parents eat, and then eat the same.
Cashdan looked through the literature for evidence that this actually happens in the wild. She found it: for instance, one 1984 study of mantled howler monkeys reported that a hungry juvenile will wait to see which branches its mother nibbles leaves from and then copy her behaviour.
Human toddlers do seem to use parental or adult guidance in deciding what is and is not safe to eat
That same study also revealed that adult howler monkeys are careful to control their eating habits in front of infants. The adults never sampled new foods while infants were present.
Instead, they saved such gastronomic experiments until they were alone, perhaps to make sure that youngsters did not copy their elders and eat the new food, until the adult could confirm it was safe.
A similar sort of parental guidance might occur in chimpanzees. Cashdan came across reports of chimpanzee mothers pulling the hands of their infants away from toxic leaves.
But the behaviour may not be universal. A 2001 study of captive capuchin monkeys found that they could not be coaxed into trying a new food, of an unfamiliar colour, simply by watching their peers eating food of the same new colour.
Nevertheless, human toddlers do seem to use parental or adult guidance in deciding what is and is not safe to eat.
A 2005 study led by Leann Birch, then at Pennsylvania State University in University Park, used a similar procedure to the capuchin monkey study. Birch and her colleagues discovered that children ate and accepted more of a new food – semolina dyed an unfamiliar colour – if a nearby adult was eating semolina of the same colour. If the adult was eating a different colour of semolina, or not eating at all, the toddlers became far more reluctant to eat the new food.
Results like this make sense to Llewellyn, and they lead to an important tip for dealing with fussy children (for others, see "Feeding a fussy toddler"). Parents might like to try eating at the same time as their children – and ideally, the same food as the child. "My biggest bugbear is going to restaurants where there is a separate children's and adult's menu," she says.
Breastfed infants get experience with flavours in the maternal diet via breast milk
Adults and children probably ate together for much of human evolution, and they may continue to do so in the world's remaining foraging societies – something that Cashdan thinks might be worth researching. "I did a pilot project in a forager group, but didn't pursue it," she says. At the time – about 20 years ago – so few researchers seemed interested in the subject that she stopped working on it.
She suspects that children in these societies use social eating to deal with any food neophobia they might have, and that consequently such children are unlikely to be picky eaters.
However, the unfortunate reality is that many parents in developed societies would struggle to share mealtimes with their young children. "Some parents do arrive home late. A toddler can't eat an evening meal at 8pm," says Llewellyn.
Life in developed societies can have a detrimental impact on a child's eating habits in other ways.
Many mothers are pressured into returning to work soon after birth, and cannot necessarily find the time in their work schedules to pump breast milk for their baby. Consequently, such babies are often moved onto formula.
"Breastfed infants get experience with flavours in the maternal diet via breast milk," says Birch. "Formula is the same old stuff day after day, so it does not help the infant by providing flavour variety or repeated experience with specific flavours."
If you pull all of these strands together, the science begins to point towards a narrative that makes picky eating easier to understand.
Humans are, fundamentally, omnivorous foragers. In common with youngsters in other omnivorous foraging species, children have evolved to be wary of eating new foods to avoid accidentally poisoning themselves – the omnivore's dilemma.
With enough patience, parents can overcome a child's picky tendencies
Also, again in common with some other foragers, these youngsters learn to overcome their reticence by watching and copying their parents at meal times.
The fact that children will have experienced the flavours of key foods since birth, through their mother's milk, also helps them overcome any innate urge to reject new foods.
However, toddlers in developed societies often miss out on the cues that can help them escape the omnivore's dilemma. As a consequence, some of them become picky eaters. In extreme cases, a child's innate tendency towards food neophobia might be magnified and develop into full-blown food fussiness. They might retreat into a narrow diet that makes them feel comfortable, but that does their health little good.
Llewellyn has a final message of hope, though. Even where young children have developed into extremely picky eaters, it is still possible to help them broaden their diets.
With enough patience, she says, parents can overcome a child's picky tendencies. "Our genes are not our destiny. Just because this has some genetic basis, it doesn't mean it's not changeable."
Join over six million BBC Earth fans by liking us on Facebook, or follow us on Twitter and Instagram.
If you liked this story, sign up for the weekly bbc.com features newsletter called "If You Only Read 6 Things This Week". A handpicked selection of stories from BBC Capital, Culture, Earth Future and Travel, delivered to your inbox every Friday.