Parents and caregivers profoundly influence the eating environment in which children's preferences and intake regulation patterns develop: they determine the availability and composition of the child's diet, provide a model of eating behavior, and guide the child's eating through feeding practices ³². As such, the family eating environment constitutes the first and primary context in which children's eating patterns emerge.

Composition of the family diet. Humans begin life consuming a single food that provides all nutrients and move to a more complex diet in which a variety of foods must be consumed to meet nutrient requirements ³³. In the earliest studies on the topic, Davis ^{20, 34} observed that infants and children could adequately self-regulate energy intake to support growth and health when allowed to self-select food intake from an array of nutritious foods. She noted that the "trick" of the study was to use foods of good nutritional quality. In essence, the dietary profile of foods consumed by infants and young children reflected the food selection choices of the caregiver. Similarly, research with older children revealed that the availability and accessibility of foods in the home, specifically fruits and vegetables, are related to increased preferences for and intake of those foods³⁵. By selecting the foods that come into the home, parents have direct control over the foods to which children are repeatedly exposed. This point is particularly important given that familiarity and repeated exposure to foods facilitate the acquisition of food preferences¹⁹.

Observing the eating behavior of family members. A second role that parents play in the development of children's eating behavior is serving as models, both in terms of what and how much is consumed. One recent study found that mothers' own beverage choices were related to the trade-off between milk and soft drinks in their daughters' diets; mothers who consumed more milk had 5-year-old daughters who consumed more milk and less soft drinks³⁶. A subsequent analysis revealed that mothers who drank milk more frequently, and made milk available at meals and snacks, had 7-year-old daughters who drank more milk than girls whose mothers did not, themselves, consume milk frequently³⁷. A recent study provides evidence that modeling effects may be most significant for high energy containing foods³⁸. The persuasive nature of parental modeling on children's diets is demonstrated by the fact that watching older individuals consume "hot" or spicy foods induced Mexican children to taste and eat small quantities of the foods³⁹. Modeling and peer pressure are sufficiently rewarding to encourage children to consume foods that are initially unpleasant to ingest.

Whether children learn how much to eat from observing their parents is not as well understood. One study found that mothers who reported overeating had preschool-aged daughters who tended to eat in the absence of hunger⁴⁰.

Finally, children learn manners and adopt cuisine rules from observing their parents, such as whether foods are eaten with a spoon or with their hands and at what type eating occasions foods are normally consumed. For instance, children tend to prefer particular foods at the times of the day when those foods are most commonly consumed within a given culture⁴¹.

"... feeding practices that are particularly controlling are not effective, but rather may be counterproductive in promoting healthful patterns of eating. "

Child feeding practices. At any point in development, large differences may exist among parents in the extent to which they allow the child to control eating, including the timing of meals, as well as what and how much is eaten. A growing body of literature indicates that feeding practices that are particularly controlling are not effective, but rather may be counterproductive in promoting healthful patterns of eating. Specifically, the use of pressure and restriction in child feeding appears to have paradoxical effects on children's preferences. Experimental studies have shown that children's preferences decrease for foods they are pressured or rewarded to eat⁴².

The relationship between using reward contingencies to encourage intake, however, is not straightforward. A series of experiments have shown that rewards can promote the consumption of fruits and vegetables when paired with modeling and praise^{43, 44}. On the other hand, restricting access to foods draws children's attention to and increases their preferences and selection of those foods⁴⁵. In one study, Fisher and Birch⁴⁵ evaluated the effects of restricting access to cheese crackers on subsequent selection and intake. Compared to those snack periods in which cheese crackers were provided ad libitum, restriction resulted in an increased attention to and selection and intake of the restricted crackers when they were made freely available (Figure 6).
|

Figure 6. After restriction, children showed increased interest (behavioral response) in restricted foods as well as increased intake of restricted foods.

Because children readily form preferences for foods high in fat and sugar^{24, 46}, parents' restriction of such foods may only enhance liking and intake of readily preferred foods-even in the absence of hunger.

Another unintended consequence of using pressure or restriction in child feeding is that it may guide children's eating by factors external to their own hunger and fullness. For instance, one study found that young children's ability to regulate energy intake was disrupted when they were encouraged to "clean their plates" by adults who offered them rewards for doing so and focused them on the amount of food remaining on the plate⁴⁷. Other work has found that parents who report using restriction in feeding have preschool-aged girls who are likely to eat restricted foods, even in the absence of hunger^{29, 45, 48}. In a recent study of non-Hispanic White girls followed from 5 to 7 years of age, Fisher and Birch⁴⁹ found that girls who ate in the absence of hunger at 5 and 7 years were more likely to be overweight at both ages. Further, they were likely to have parents who reported higher levels of restriction in feeding than girls who ate little in the absence of hunger. Feeding environments that are neglectful and characterized by the child having little social contact with and encouragement from their care provider produce equally undesirable outcomes, such as non-organic failure to thrive⁵⁰ and an increased risk of obesity in young adulthood⁵¹.

Whether controlling child feeding practices directly contributes to problems of overweight and underweight among children is unclear and will be best answered by longitudinal research. Several studies have not found evidence for an effect of parental control on weight status in young children^52-54. It is likely that parents' use of child feeding practices, in part, reflects their perceptions of the child's eating behaviors or weight status as being problematic³⁰. Costanzo and Woody⁵⁵ contend that excessive control is imposed in feeding when that area of child behavior is important to the parent and potentially problematic for either parent or child. This perspective is supported by empirical work showing that mothers' own restrictive eating behavior and perceptions of their daughters' risk for overweight are associated with the use of restriction in child feeding^{29, 30}.

Genes and the Environment

Parents provide their children with both genes and the environment in which eating occurs. Genetic influences on eating are apparent at birth, in the observation that infants possess an unlearned preference for the taste of sweet^{7, 56}. Research on the genetic basis for taste preferences has recently focused on taste perception of bitter substances⁵⁷. Individuals who perceive 6-n-propylthiouracil (PROP) to be an extremely bitter taste appear to be more sensitive to the intensity of sweet tastes^58,59 and the texture of fat⁶⁰. The consequences of these genetically programmed taste preferences for food intake selection patterns is yet unclear. Although the literature linking PROP status to intake patterns is small and somewhat conflicting, a central interest is whether an elevated sensitivity to bitter tastes translates into a reduced preference for, and intake of, foods with intensely bitter tastes, such as green leafy vegetables⁶¹.

A separate line of research attempts to discriminate genetic versus environmental contributions by comparing eating behaviors among individuals with varying degrees of genetic relatedness. Dietary patterns have been found to be similar among family members; spouses tend to have similar nutrient intakes^{62, 63} and children's nutrient and food intake reflect the diets of their parents^63-66. Twin and family studies focused on investigating genetic influences on energy intakes, intakes of specific foods, and food preferences have revealed modest genetic contributions in the range of 0-30%^67-69. For example, a recent study by Faith⁷⁰ found that approximately 24-33% of the variance in measured energy intake in a laboratory setting among monozygotic and dizygotic twins could be explained by genetic influences. There is some evidence that the manner in which food intake is regulated has a genetic basis as well: two studies have shown similarities among mothers and daughters to self-imposed eating restrictions^{71, 72}. The extent to which such similarities reflect genetic heritability, however, is poorly understood ^{67, 73}. Collectively, such research indicates that genetic variability in taste and food preferences as well as intake patterns has a potentially important role in explaining individual differences in eating behavior among children. Identifying genetic markers for eating will be critical to understanding how genetic predispositions interact with dietary experiences and environmental constraints to produce patterns of eating.

The Contemporary Eating Environment

While the family is undoubtedly the first and most central environment in which children's eating develops, families do not parent their children in isolation of the larger societal and physical environments in which they live. Environmental constraints on parents' ability to promote healthful patterns of eating include time demands in family life, loss of the family meal, television, dining out, and childcare. In addition to these changes in the day-to-day aspects of family life surrounding eating, parents are also faced with the pressures of the broader conditions of health and eating conditions of our society. This climate includes poor diet quality and an ever increasing prevalence of overweight among parents and children at the same time that dieting and values on thinness are pervasive.

Time demands in family life. Understanding the climate in which children's eating develops requires examination of how family life has changed over the past several decades. An increasing number of Americans report always feeling rushed and never having enough time ⁷⁴. The number of dual-earner families has risen, with an increasing number of women in the workforce as well as an increase in the average number of hours worked by women^{75, 76}. At the same time, while men's participation in household activities has increased over the past two decades⁷⁶, women continue to be more responsible for household labor than men ⁷⁷: 87% of women indicate that they hold the primary responsibility for household meals⁷⁸. One study found that while married women's percent of paid hours nearly doubled from 1971-1991, the percent of time spent in household labor decreased by only 15% ⁷⁷.

The composition of families also has changed over the past decades, with an increase in single parent families from 12% in 1970 to 26% in 1996⁷⁹. Of concern is whether these trends may bear negatively on the frequency with which families rely on dining out and convenience foods of poor nutritional quality. It is possible that these increasing demands reduce the amount of time spent in all aspects of childcare by caregivers within the home-including food preparation and family meals ⁸⁰.

Figure 7. Parents should be encouraged to offer children a variety of foods, keeping in mind the developmentally appropriate portion size for individual children.

Overweight and dieting. The challenge to improve diet quality among children occurs at a time when epidemic levels of overweight among Americans are observed in conjunction with pervasive dieting, particularly among females, and body dissatisfaction and weight concerns among school-aged children. Currently, one of every four children ^{121, 122} and one of every two adults ¹²³ are overweight. While dieting behaviors tend to emerge in adolescence ¹²⁴, girls as young as 5 years of age appear to possess knowledge about dieting ¹²⁵. Further, dieting and body dissatisfaction may be prevalent among ¹²⁶, but are not limited to, overweight children. Data from the NHANES III survey revealed that roughly half of girls aged 12-16 years who considered themselves overweight were, in fact, of normal weight ¹²⁷. One recent multi-ethnic study of third-grade girls and boys found that one of every four boys and one of every three girls reported wanting to lose weight¹²⁸.

Promoting Healthy Eating Behaviors: Child Feeding Guidance for Parents

What does the climate surrounding eating signify for the capacity to nurture the development of healthful patterns of eating among children? Importantly, establishing feeding environments to foster healthful eating behaviors in children and support attainment of healthy weight and growth is an attainable goal. Using nutrition guidance within a developmental framework is a first and fundamental step to parents and caregivers cultivating those skills necessary to establish appropriately structured, pleasant, and interactive eating environments. Ellyn Satter, a well-known clinician who specializes in the practice of helping parents and children with feeding and eating challenges, maintains that there is a division of responsibility in child-feeding where the parent is responsible for the structure of the feeding environment and for the foods offered. She contends that children are responsible for how much and whether they choose to eat at a given eating occasion ¹²⁹. This stance is well supported by research from our laboratories and those of our colleagues.

Know what is appropriate for children. Knowing what children need to eat is requisite to translate effectively dietary goals into nurturing child-feeding behaviors. There are several important issues for most children. First, young children eat small amounts of food frequently; three meals and three snacks is a normal eating pattern until well into the school years. This is not to say that children should adopt a grazing pattern of eating or drinking energy-containing beverages-having discrete eating occasions is important. Second, the appearance of erratic intake patterns is not synonymous with poor eating habits; children's self-regulation of energy intake occurs across a number of meals-many of which will be supervised by an adult other than the parent. Parents should be encouraged to consider the adequacy of intake across the day and beyond, rather than to focus on "getting a child to eat" at a particular eating occasion. Finally, young children require fewer calories and accordingly, smaller portion sizes. Because increasing portion sizes may increase energy intakes, particular caution should be given against routinely offering pre-portioned adult-size servings of beverages and snacks. The United States Department of Agriculture's Food Guide Pyramid can be used as a reference for understanding an appropriate number of servings within each food group as well as the contribution of each food group to children's total dietary intake (Figure 7). Examples of reasonable portion sizes for small children are given in Table 1.

Feeding Guide for Children*

Provide diverse eating experiences. A parent's understanding that neophobia and other childhood eating behaviors are a part of normal child development may reduce anxiety and decrease the intensity of power struggles. A persistent and patient approach on the part of the parent is critical to a child's acceptance of a variety of foods given that most children tend to reject new foods and go through stages of food "jags." Repeated exposure to novel foods induces "learned safety," or acceptance and trust of new foods and, over time, has a positive influence on dietary variety. However, it requires many exposures to a novel food before acceptance and liking are positively impacted. Parents seldom continue to offer foods in the face of rejection, perhaps believing that food preferences are "hard-wired," like preferences for the basic tastants sweet or bitter. More often, parents move on to other novel foods-which may also be rejected. Acceptance of new foods by the young child requires paced introduction-taking care to build food acceptance slowly and patiently.

Both pressuring children to eat and restricting their access to foods appear to have paradoxical effects on preference. The central goal of developing healthy preferences can be encouraged by supporting children's participation in food choices while providing diverse food experiences. As such, parents can be encouraged to adopt the perspective that all foods can be consumed in moderation. Combining familiar tastes with new foods, such as offering a familiar dip with a new vegetable may facilitate acceptance^130. Children's acceptance of novel foods can be encouraged by involving them in the purchase of foods as well as in recipe selection and preparation. It is critical for parents to accept that initial rejection does not mean a food will ultimately be rejected.

"Children learn both through interaction and by observing."

Help children learn to guide their eating according to internal cues. Caregiver strategies that are child-focused and developmentally appropriate with respect to how much children consume, that are autonomy-producing, and that center on internal signals of hunger and satiety engender greater sensitivity to the energy density of the diet and better self-regulation skills^{31, 47}. Children learn both through interaction and by observing. Therefore, parents and caregivers should first examine their own eating cues before they consider how to help children use hunger and fullness to guide their eating. It is likely that parents' own control over food intake is modeled by children-both with respect to when eating routinely begins and ends. For instance, a child may learn that it is normative for eating to continue beyond fullness by observing a parent who always chooses to eat dessert regardless of what was consumed at dinner. Further, parents may unknowingly project their own eating concerns on to their children. For instance, parents who restrict themselves to prevent overeating may feel that restriction is also necessary to achieve the same goal for their children. It is important to consider whether such directives result in the unintended outcome of focusing children on the parental instruction at the expense of paying attention to their own hunger and fullness. Finally, parents should encourage participation and dialogue that gives both parent and child the opportunity to talk about each other's hunger and fullness as guides to knowing when to begin and finish eating.

"Adult and peer modeling exert powerful influences on children's food acceptance patterns. Adult and peer modeling exert powerful influences on children's food acceptance patterns."

Serve as role models for children. Adult and peer modeling exert powerful influences on children's food acceptance patterns. Children learn and internalize messages regarding normative behavior by observing important others. Social settings and cues from the feeding environment provide necessary opportunities for young children to learn about food and eating. As such, mealtimes and the context in which foods are presented can have a dramatic impact upon children's reaction to them and their acquisition of preference or dislike for them. Praise, when delivered appropriately, profoundly influences children's eating behaviors. Foods that are paired with positive attention, like praise or reward, become highly sought-after and preferred.

In contrast, when children are coerced or bribed to ingest target foods (like green vegetables), often by the enticement of sweet, energy-dense foods (that they easily learn to prefer), the liking for the target food falls below initial ranking. Thus, the common practice of using (food) reward to encourage ingestion of less-preferred target foods may produce a short term gain in consumption but is counter-productive for the long term goal of establishing healthful eating habits and acceptance of a wide variety of foods.