When Your Toddler Won’t Eat Dinner – Food Takeaways from Cribsheet by Emily Oster

Sometimes it feels like no matter what you try, your toddler won’t eat dinner . . . Below is an in-depth review of toddler food strategies that Emily Oster discusses in her book Cribsheet.

SKIP TO: Action Items for Parents (Overview of 4 Basic Strategies)

SKIP TO: Key Takeaways (More in-Depth Discussion)

You can buy Cribsheet by Emily Oster HERE.

Parents of infants and toddlers. Late-term expecting mothers.

I’m a big fan of Emily Oster’s books. She is great at summarizing and evaluating research on a given topic and explaining the relevance of this research to a parent’s decision-making tree, acknowledging that different families have different needs and priorities. She is not prescriptive. I really like that, and I think lots of other parents similarly enjoy being treated like they can make their own decisions.

I’m writing a blurb on this book because my cranky pants toddler won’t eat dinner and sometimes refuses to eat anything but milk. (Do I sound bitter?) Oster addresses this topic head on in a few pages, so I wanted to dig into her advice a bit by reviewing some of the studies she references in more depth. I thought perhaps other parents might enjoy coming along for the ride.

If you’re short on time, here are some food strategies both from Cribsheet and from some of the studies that Oster cites in the book. I’ll dive into more details on each strategy and the rationales below in the Takeaways section.

There are two big things happening around age 2. First, kids’ caloric needs change because they are entering a period of slower growth. So keep that in the back of your mind. They literally don’t need as much food as they did just a few months back. If you’re really concerned, talk to your pediatrician. Just keep in mind that many parents (like 25-35%) think their kids aren’t eating enough around this age. Even though they are!

Second, kids are developing a sense of autonomy. Yay! One fun way to exercise that is to say no to food. Therefore, don’t try to coerce them into eating — this can backfire. Instead, find ways to encourage eating that support their developing autonomy. And keep in mind that it’s quite normal for some meals to be tiny and some to be massive.

Repeated exposure to a food over the course of 8 days will increase a child’s willingness to eat that food. An exposure means at least 1 teaspoon.

Many parents give up too early in trying to get a child to like a food. Researchers have found that toddlers may continue making negative facial expressions around a food, even when they are willing to eat more of it. This effect shows up in those same studies that show repeated exposures leads to toddlers being more accepting of the food. When parents in those studies ranked how much their children liked the food before and after exposure, they indicated that they thought there was no change.

Rotating exposure to of different types of foods within a food group may make a child more willing to try and eat new foods in that same food group. For instance, if a child tried a new fruit every day, the child should subsequently be more willing to eat a new, different fruit.

Further, offering a variety of foods within the food group within a single meal may be more important than offering the food at two different meals during the day. It seems like we need more research in this area to pin this down, but still may be worth trying. In other words, if you want to try this with vegetables, try giving your kid two vegetables at dinner, rather than one vegetable at lunch and one at dinner. Maybe offer two fruits at lunch then.

Kids are smart. If you make something else when they refuse to eat, they learn from that experience.

Encouraging a child to eat or try something new in a neutral tone of voice, eating with them and expressing your enjoyment of a food, explaining why a food is good for them, all increase the likelihood a child will accept a food. In addition, asking a child if she wants more, and only giving her more if she actually expresses a yes is more likely to promote a “yes” to trying something new. These are all examples of prompts / dinnertime behavior that supports a child’s budding independence. Pressuring, bribing, or coercing a child to eat can backfire (studies support this).

I’m focusing here on the narrow piece of Cribsheet that is relevant if your toddler won’t eat dinner (or any other meal / food). In other words, I’m not going to delve into most of what Oster covers in this book. (She also covers postpartum recovery, breastfeeding, co-sleeping, vaccines, the decision to stay home versus work, etc.)

In her chapter “Beyond the Boobs: Introducing Solid Foods,” Oster mentions that this is all very normal. Kids become very picky around age 2 and then slowly grow out of it as they hit elementary school.¹ She also notes that because many (admittedly not all) children are great eaters before this, parents are often taken aback and convinced that their child is not eating enough and is rapidly becoming malnourished.

There are two parts of this struggle that are worthwhile to tease apart. First, around this age, kids just eat less. Second, around this age, kids start developing a sense of independence, which encourages them to become more picky about what goes into their mouths. These two things feed into each other, but let’s address them one at a time.

First, the appetite decrease.

Oster points to a review article² which explains that the decrease in appetite is normal for children between ages 2 and 5, because children enter a slower growth period around this time. Just think about how much bigger your 2-year-old is now versus a year ago! Typically, toddlers grow about 4-5 inches from age 1 to age 2, and they gain 4-6 pounds. On the other hand, from age 2 to age 3, they grow only about 2-3 inches and gain 3-5 pounds.

Approximately 25% to 35% of parents describe their 2-5-year-old toddlers as poor or picky eaters. But the article asserts that parents are the ones with unrealistic expectations regarding how much their kids need to eat. Most of these kids are actually getting enough calories, especially when you consider the slowed growth curve. The article recommends that parents just focus on offering nutritious foods to the children – and leave the child responsible for how much he or she eats, without coaxing.

They also suggest offering small amounts (1 tablespoon of a given food for each year old the child is). I hadn’t heard this last one before, although we intuitively discovered it when our son made it a habit to refuse plates with large servings of foods we know he really likes. Maybe smaller portions feels more manageable? It certainly creates an opportunity to encourage your kid to try other things on the plate, especially if they want more of whatever they finished.

The article goes on to note that any snacks offered to children should be at the midpoint between meals. They should be dense in nutrients (we do lots of nuts, nut butters, and whole-milk yogurts in our home). Parents shouldn’t offer snacks if they will interfere with the upcoming meal. Further complicating things for parents, it’s completely normal for toddler appetites to be “erratic” during the 2 to 5 year-old age range. Intake can vary considerably between meals and even days.³

Of course, knowing that this is normal doesn’t address how parents are often not just worried about toddler starvation – they’re worried about hangry toddler mood swings. Therefore, even though articles suggest just offering the healthy stuff and letting the kid decide how much to eat, I certainly have had days where I just want to give my daughter whatever unhealthy food she’s currently begging for so that she’ll eat something, feel a bit better, and stop tantruming. Hangry toddlers are no joke, and, the article isn’t very helpful on that front. Just keep the toddler in a perpetual state of moderate hanger and eventually he’ll eat a green bean. Great for his digestion, not so great for parent-child bonding.

On a more serious note, I suppose it does take some stress out of the equation to realize that a number of meals are predestined failures. And that has nothing to do with you as a parent or your parenting skills! It’s just how toddlers are – sometimes they’re not hungry. If they aren’t, it sounds like parents should just leave it and sit tight for the next meal. Which actually may contribute to why toddlers often refuse to try new things. If they’re just not hungry, you’re already fighting an uphill battle.

Second, the budding sense of independence.

The article also explains that, psychologically, toddlers are developing a sense of autonomy. They therefore naturally become selective during this period. If parents pressure a child to eat, this budding autonomy may actually cause a child to resist eating. To avoid this reaction, the article recommends that parents focus on maintaining a positive meal environment. They can also model the eating behavior they want to see in their children (i.e., you eat the green beans and experience nirvana before your toddler’s very eyes). Toddlers are super into copying. More on how to successfully prompt toddlers to accept foods in the Family Meal Study below.

So its good to know all this is normal. Now let’s turn to the specific strategies to manage your toddler who won’t eat dinner (or other meals).

Oster discusses several studies briefly and then gives two general recommendations on how to get your toddler to eat well. First, if your kid won’t eat something, just keep trying. Children who try a new food several times are more likely to eat it going forward. Second, if your kid won’t eat something, don’t offer an alternative. These recommendations are discussed below. But first, let’s digress a bit into some studies that show how early experiences may shape a child’s receptiveness to certain foods.

There has been some research demonstrating that tastes can be formed early in life. Therefore, it may be important to introduce lots of flavors at a very young age. However, keep in mind that not all of the research in this area is great.

The Carrot Study

There is an Oster-blessed study⁴ that demonstrates that you can increase an infant’s receptivity to a particular flavor by introducing the flavor in the womb and/or through breast milk. Specifically, the “carrot study” encouraged one group of moms to eat a lot of carrots while pregnant and breastfeeding. Another group ate a normal amount of carrots. The children whose mothers had been in the high-carrot group preferred carrot-flavored cereal. The other children did not prefer the carrot-flavored cereal.

So if you’re pregnant or breastfeeding, try to eat the vegetables and other foods you want your kids to eat one day. You’re giving them their first taste of what meals should taste like. Keep in mind that you may need to target and then eat a lot of the specific item you want them to like. (I wish I had done this with tomatoes. What kind of tomato-hating monsters am I raising?)

The Food Habits Questionnaire

Oster flags a paper⁵ that received lots of media attention in 2017. The study found that children who ate a varied diet with lots of fruits and vegetables at 9 months old were likely to eat a varied diet with lots of fruits and vegetables at 6 months old. The researchers concluded that “tastes are formed early, and it is therefore important to expose children to a variety of foods early in life.”⁶

The study collected information from parents as follows. When infants were 9 months old, mothers received a questionnaire listing foods. They were asked to notate the foods and beverages offered to their infants on a daily or weekly basis. When the children were 6 years old, the mothers received a questionnaire with a wider variety of food options. They again reported what foods their children were eating.

The study found astonishing results (excuse my sarcasm). An infant’s dietary patterns at 9 months was predictive of their dietary patterns at 6 years old.  Those eating a variety of fruits and vegetables at 9 months were more likely to be eating a variety of fruits and vegetables at 6 years old.

Oster points out that the researchers’ conclusion that early exposure to foods affects the child’s preferences is one, but hardly the most likely, conclusion one might draw. A far more likely conclusion is that parents who offer a variety of fruits and vegetables to their 9-month-olds will continue to offer fruits and vegetables to their 6-year-olds. Maybe the researchers should have focused on how toddlers “eating” a variety of fruits and vegetables were in fact “being fed” a variety of fruits and vegetables. 6-year-olds aren’t exactly going out and doing the grocery shopping. One would expect a family’s eating patterns in year one to be predictive of their eating patterns in year five.

There is also some discussion in the study about how breastfeeding predicts more vegetable consumption. Oster doesn’t spend time on this, but it is an interesting correlation. My guess would be that this is explained by the fact that more educated, richer women are more likely to breastfeed,⁷ and are also more likely to eat more vegetables themselves and to offer their children more vegetables.⁸ In other words, the women more likely to breastfeed are also more likely to focus on offering their kids vegetables. The other explanation would be that women who eat a lot of vegetables may expose their children to these flavors through their breastmilk – and in fact, this is potentially supported by the carrot study discussed above. That said, I found another study (not discussed here) that demonstrates that breastfeeding confers an advantage in initial acceptance of foods, but only if the breastfeeding mother eats the particular food regularly.⁹ In other words, moms must pick their vegetable of choice.

Okay, so now let’s move on to some studies that are more useful to those of us with hangry toddlers.

Okay, so if you aren’t still breastfeeding, or you breastfed while eating cake and ice cream for for every single meal (one can dream), there is still hope for your child to eat well. There are two key strategies supported by this next study.

First, offer the child at least a teaspoon of the desirable food over and over again until she will accept more of it. Remember that the goal isn’t for your kid to eat tons of the given food in a sitting. Just for them to taste it enough times to develop a liking or at least an acceptance of it.

Second, offer the child at least a teaspoon of 2 foods in the food group you want your child to eat more of at a given meal, and try to rotate the foods each day. Again, you’re just aiming for them to taste the food during these initial exposures, not eat tons of it.

The Fruit & Vegetable Study

This paper¹⁰ describes two legs of a study – one with fruits, and one with vegetables – involving 74 pairs of mothers and infants between 4 and 9 months old (the infants, not the moms). Mothers were instructed to expose their babies to a particular food or a variety of foods over the course of 8 days (at least 5 grams, or approximately 1 teaspoon, to count in the study). The infants were evaluated for how much they were willing to eat of particular foods prior to and following the 8-day exposure period.

The fruit leg

In the first leg of the experiment, two groups of infants received different exposures to fruits over an 8-day period. The first group received pears at a target feeding at the same time each day. The second group received a fruit that rotated day to day (prunes, peaches, or apples, but not pears) at the target feeding. The groups were evaluated for their liking of pears and of green beans before and after the study. Both groups consumed more pears after the exposure period. Neither group consumed more green beans.

The vegetable leg

In the second leg of the experiment, three groups of infants received different exposures to vegetables over an 8-day period. The first group received green beans at a target feeding at the same time each day. The second group received a vegetable that rotated day to day (carrots, squash, peas, and spinach) at the target feeding. A third group received two vegetables at the target feeding each day, one green, one orange. One vegetable had also been offered the preceding day (pairings included carrots and peas, squash and peas, or squash and spinach). The groups were evaluated for their liking of green beans, carrots, and spinach before and after the exposure period. Afterward, the first two groups ate more green beans, but did not eat more carrots or spinach. The third group ate significantly more green beans, carrots, and spinach.

Lesson #1: Repeated Exposure Works

Repeated exposure to a specific food appears to increase the likelihood that a child will accept it. The researchers counted it as an “exposure” if the child ate at least 5 grams (about a teaspoon). The researchers note that infants in the second group of the vegetable leg of the study were exposed to carrots twice and spinach twice, but were not more likely to accept either at the end of the study. However, infants in the first group were more likely to accept green beans after 8 exposures. Based on this, the sweet spot appears to be somewhere between 3 and 8 exposures over the course of a week. Unclear how many times you’d need to expose a child if you’re not doing them so closely together in time, but apparently the rule-of-thumb in France is 10 exposures for a kid to learn to like a given food.

Lesson #2: Experiencing Variety Increases Preference for Similar Variety

Infants who experience a variety of foods will generally increase their preference for variety, and thus their willingness to accept new foods. This preference for variety is called the “variety effect” and has been studied in other contexts with various animals.¹¹

In this study, the researchers demonstrated that an infant’s preference for variety and willingness to try new things increased after exposure to a variety of fruit (i.e., one rotating fruit each day), because this exposure increased their liking of one particular fruit that they had not been offered every day: pears. The effect was similar to that seen in children who had been exposed to pears each day during the 8-day period.

The researchers also noted a limitation on the variety effect: it appears to be restricted to similar foods within a category (e.g., exposure to a variety of fruits will increase the child’s acceptance of other fruits, but not of vegetables). This limitation was seen when the children eating a variety of fruits increased their acceptance of pears, but not of green beans.

Lesson #3: Vegetables May Take an Extra Push

Third, there was a wrinkle in the study for which the researchers offered several potential explanations. Unlike in the fruit context, children offered one rotating vegetable each day were not more likely to accept other vegetables. Specifically, the children offered one rotating vegetable each day did not increase their liking for carrots and spinach, and showed only a slight increased preference for green beans after the exposure period. On the other hand, children offered two vegetables at one meal each day greatly increased their liking of green beans, spinach, and carrots. So somehow the variety effect was on show for the children fed two vegetables that rotated at the target meal each day, but it was not on show for the children fed one vegetable that rotated each day.

The researchers offer several theories for this discrepancy. First, when working with vegetables, which are not as sweet as fruit, children may simply require exposure to more vegetables that are similar to carrots, spinach, and green beans (which the third group received because they received at least one green and one orange veggie a day.)

Second, the variety effect may be more robust when children experience variety within the meal itself. That is, if they can experience differences in texture, color, and flavor between two different vegetables at a given meal. The researchers note that for rats, the variety effect is most pronounced when there are clear sensory differences between the foods.

Third, the researchers noted that there may be some benefit in there being some continuity between tastings – in this group, one vegetable offered had also been offered the day prior. The researchers note that this combination of novelty in the context of a familiar food may be an optimal way to introduce new foods.

The theories proposed by the researchers make an argument for attempting to offer your child two vegetables at a given meal – and if you can only offer two vegetables a day, offering them within the same meal may be more beneficial than spreading them out between meals. Of course, this may be a problem if your chosen meal is dinner, and your toddler refuses to eat dinner. As far as I’ve found in an initial search, there doesn’t seem to be a study that tests this point. It may also be helpful to introduce a new vegetable alongside a familiar vegetable.

(Inspired to look up some new toddler-friendly vegetable recipes?)

The one theory the researchers did not suggest was that the “vegetable” category was simply too broad. (Remember the limitation on the variety effect — it only applies to similar foods.) Personally, I think carrots and spinach have very different flavor profiles. I would argue that sweeter vegetables like carrots, squash, and even maybe peas belong in a different category from more bitter vegetables like spinach.

The final lesson from the Fruit & Vegetable Study was fascinating to me. The parents of infants who accepted more of a given food did not report any perceived increase in the child’s liking of the food.

Further, the researchers referenced another study analyzing infant expressions during feedings (as recorded and then objectively observed by researchers), revealing that 8 days of exposure to a given vegetable did not decrease the number of negative faces the infant made during feeding, even when the infant ate more of the food.¹² The authors of the study noted that if “mothers were apparently unaware of the changes in acceptance, they may give up too soon when introducing these foods.”

Therefore, a key takeaway here is that parents should continue to offer a vegetable even when the child appears not to like it. A better indicator of progress is whether the child is willing to eat the food — not her facial expressions during the feeding.

How parents offer food to their toddlers matters. If your toddler won’t eat dinner, don’t use coercive prompts, but also don’t offer alternatives.

The Family Dinner Study

Oster discusses a study¹³ of 60 families with 12-36 month-old children. The researchers in this study noticed that what people do often differs from what they say (and may genuinely believe) they do. So the study required that families, in addition to filling out questionnaires, videotape two interactions. First, families filmed a typical dinnertime at home where the children ate familiar foods. Second, families filmed a situation where the parents introduced a new fruit or vegetable. Before conducting the taping, the researchers asked the parents to complete a questionnaire about their feeding methods as well as how picky they perceived their kids to be.

There were several interesting takeaways from this study.

Lesson #1: Pickiness is Apparently Not Defined by Number of Food Refusals

After reviewing the video interactions, the researchers noted that there appeared to be no difference in the number of refusals between the children parents characterized on their surveys as picky and those characterized as non-picky. Rather, the difference was in the type of refusal. For those characterized as picky, the researchers observed more crying or covering of the mouth in the picky group when they refused to eat a food.

Lesson #2: Food Prompts Matter

The researchers found that the prompts most likely to be accepted were “autonomy supportive.” Coercive prompts were more likely to result in a refusal. Parents of the picky eaters tended to use more coercive prompts. Of course, as the researchers note, it is hard to know if the parents of the picky eaters used more coercive prompts because the children appeared to them to be picky, or if the coercive prompts caused the picky behavior.

This feeds into the point above about how children at this age are developing a sense of autonomy – therefore, finding ways to incorporate the child’s budding independence are more likely to succeed.

Lesson #3: Other Actions Associated with Food Acceptance

In addition to the prompts, there were other factors that the researchers observed to be associated with more acceptance of a food:

• Verbally offering a child additional servings and only giving more if the child said “yes”;
• Praise of the eating behavior or of the child in general; or
• Imposing restrictions on how much food, or of a particular food, that the child could eat.

Lesson #4: Other Actions Associated with Food Refusal

The researchers found that, based on parents’ self-reporting (not from their observations of the videos), the following factors were associated with an increased likelihood in a child refusing food:

• begging a child to eat; or
• making something else if the child refuses to eat.

I recently came across a book called “French Kids Eat Everything” that has a lot of overlap with this one. It has strong “Bringing up Bebe” vibes, if you read that one. (Why are we North Americans so obsessed with French child-rearing?) This is one Canadian woman’s story of how she and her husband (who is French) decided to move back to his small home town in France. At the time, her young children seemed to be eating mostly plain pasta with Parmesan and snacks.

The book details many cultural differences regarding how the French introduce their children to food. One of which is the philosophy that if a child doesn’t like something, he or she just hasn’t tried it enough times yet. From her telling, it seems to be a generally accepted axiom in France that children need to try something around ten times in order to learn to like it. So parents aren’t at all phased by kids disliking a food, or persuaded by an initial rejection that the child just doesn’t like the food. They just tell the child that she probably hasn’t tried it enough times yet. The book also discusses some interesting ideas, like removing snacking, treating meals with more ceremony, and the role of peer pressure in encouraging children to eat new foods.

I’m still finishing this book up, but it seems to fit well with this topic and it is told in a narrative format, which I think makes it a compelling read (at least so far).

---

This blog contains affiliate links. So, if you purchase a book through one of the Amazon links, I may earn a small commission at no additional cost to you. Thank you for supporting this project!

1. Oster, E. (2019). Cribsheet: A data-driven guide to better, more relaxed parenting, from birth to preschool (p.197). Penguin Press. . ↩︎
2. Leung AK, Marchand V, Sauve RS; Canadian Paediatric Society, Nutrition and Gastroenterology Committee. The ‘picky eater’: The toddler or preschooler who does not eat. Paediatr Child Health. 2012 Oct;17(8):455-60. doi: 10.1093/pch/17.8.455. PMID: 24082809; PMCID: PMC3474391. ↩︎
3. For more info, see: https://www.nejm.org/doi/full/10.1056/NEJM199101243240405. ↩︎
4. Mennella JA, Trabulsi JC. Complementary foods and flavor experiences: setting the foundation. Ann Nutr Metab. 2012;60 Suppl 2(Suppl 2):40-50. doi: 10.1159/000335337. Epub 2012 Apr 27. PMID: 22555188; PMCID: PMC3363345. ↩︎
5. Rose CM, Birch LL, Savage JS. Dietary patterns in infancy are associated with child diet and weight outcomes at 6 years. Int J Obes (Lond). 2017 May;41(5):783-788. doi: 10.1038/ijo.2017.27. Epub 2017 Feb 28. PMID: 28133360. ↩︎
6. Oster, 2019, p. 195. ↩︎
7. Oster, 2019, p. 68. ↩︎
8. Nutrition gap between rich and poor is growing, but don’t blame food deserts, researchers say (chicagobooth.edu) ↩︎
9. Forestell CA, Mennella JA. Early determinants of fruit and vegetable acceptance. Pediatrics. 2007 Dec;120(6):1247-54. doi: 10.1542/peds.2007-0858. PMID: 18055673; PMCID: PMC2268898. ↩︎
10. Mennella, Julie & Nicklaus, Sophie & Jagolino, Amanda & Yourshaw, Lauren. (2008). Variety Is the Spice of Life: Strategies for Promoting Fruit and Vegetable Acceptance During Infancy. Physiology & behavior. 94. 29-38. 10.1016/j.physbeh.2007.11.014. ↩︎
11. The researchers describe several studies. In one, mice were given either plain water or water with three different flavors for twelve days. After the study, the mice given three different waters were more willing to try a new flavor of water than the mice only ever offered water. ↩︎
12. Forestell CA, Mennella JA. Early determinants of fruit and vegetable acceptance. Pediatrics. 2007 Dec;120(6):1247-54. doi: 10.1542/peds.2007-0858. PMID: 18055673; PMCID: PMC2268898. ↩︎
13. Fries LR, Martin N, van der Horst K. Parent-child mealtime interactions associated with toddlers’ refusals of novel and familiar foods. Physiol Behav. 2017 Jul 1;176:93-100. doi: 10.1016/j.physbeh.2017.03.001. Epub 2017 Mar 14. PMID: 28315360. ↩︎