The findings from the present set of experimental studies shed light on how different types of dietary experiences facilitate the acceptance of fruits and vegetables during early stages of weaning. Whether the dietary experience modified acceptance depended on the flavor or flavors of foods experienced, whether the experience occurred between or within meals or both, and whether the target food was a fruit or vegetable.
The first type of dietary experience involved repeated feedings of the same food at a target meal for 8 days. Such repeated dietary exposure to pears and green beans resulted in babies tending to eat more of that particular food, findings consistent with previous studies using pureed carrots, potatoes or green beans as the target food [9
]. The minimum number of exposures required to enhance acceptance appears to be more than two since infants in the BM Vegetable Variety Group had two exposures to carrots and spinach but did not show increased acceptance of either. Likewise, all infants in Study 1 were exposed to green beans on one occasion but neither group ate more green beans, although they did eat this food at a faster rate which may reflect that they were becoming more efficient feeders.
The preference that developed after repeated experience with a particular food did not generalize since repeated exposure to pears did not modify acceptance of green beans. Similarly, repeated exposure to pureed potatoes did not modify acceptance of carrots [9
]. To be sure, these foods differ from each other in color, texture and flavor. Whether experience with a food such as carrots would generalize to another that is more similar in these flavor properties (e.g., squash) is not known. However, we do know that the acceptance pattern that develops in infants after repeated exposure to a formula appears to be specific to the flavor profile experienced. Although two brands of hydrolysate formula (i.e., Nutramigen, Alimentum) are equally unpalatable to naïve infants and share similar flavor notes, infants who were repeatedly fed one brand of hydrolysate formula developed a preference for its flavor when compared to the alternative unfamiliar hydrolysate formula [19
]. How similar the flavor has to be for the baby to generalize remains unknown. However, as will be discussed below, experience with a variety of flavors does appear to generalize to foods within a particular category (e.g., fruits).
The second type of experience involved infants experiencing between-meal variety. That is, infants who ate a variety of fruits increased their intake of pears even though they had no direct experience with pears during the exposure period. Experience with fruit variety did not generalize to green bean acceptance, however. Neither did repeated exposure to a variety of vegetables between meals alter acceptance of green beans, carrots or spinach, although there was a tendency for increased acceptance of the green beans. In a previous study, we found that repeated exposure to between-meal variety to three starchy vegetables (i.e., peas, potatoes, and squash) enhanced the intake of carrots [9
]. Perhaps acceptance of green vegetables such as green beans or spinach is more difficult to enhance because they taste bitterer and thus more exposures may be needed. In addition, the exposure may require that some or all of the foods are bitter tasting [14
]. In other words, it is not just experience with between-meal variety that is important, but the flavor of the foods experienced and how it relates to the target food. How long the effect persists in the absence of continued exposure to particular foods or a variety of food remains unknown.
The third type of experience involved infants experiencing vegetable variety both within and between meals. Infants exposed to this type of variety ate more of all three of the target foods (i.e., green beans, carrots, spinach) after the exposure period. It is unclear what infants “learned” from the between- and within-meal varied diet. Studies on rats and humans [3
] indicated that the variety effect was more robust when there were pronounced sensory differences between the foods. Because the pairs of foods fed to the infants within each meal always included a green and orange vegetable, infants experienced contrasts across many sensory domains such as color, texture and flavor. Moreover, these infants also experienced different pairs of food and ate varying amounts of these from one day to the next, providing them with sensory diversity. Exposure to such multiple sensory contrasts might have accelerated the “transfer of diversity” effect [2
] by providing more varied flavor experience and more opportunities to condition flavor preferences based on the post-ingestive reinforcing effects of these nutritious foods [8
]. Because one member of the pair was experienced the day before, the infants were also provided with continuity in flavor experiences which may have eased the acceptance of the novel food in the pair [8
]. Such feeding strategies, providing novelty in the context of a familiar food, might prove to be an optimal combination to progressively accustom infants to a diversity of novel foods. Moreover, the sensory diversity within a meal might help to prevent sensory specific satiety, i.e. the decrease in liking of a specific food during the course of a meal [3
]. It should be emphasized that if experiencing within- and between-meal variety is effective not only in promoting intake within a food category but generalizes to other foods, this raises questions regarding the acquisition of overeating pattern as of an early age. Indeed, the availability of a variety of palatable foods is an important factor in the etiology of obesity, as shown in animal models [15
] and suggested by short-term human studies [24
Day-to-day fluctuations in the caloric intake of the fruits (Study 1, BM Fruit Variety Group) or vegetables (Study 2, BM and BM–WM Vegetable Variety Group) during home exposure may reflect that infants are regulating calories [27
]. However, it should be emphasized that the present study was not specifically designed to address the issue of caloric intake regulation during the home exposure period. Although mothers were trained on the feeding procedure during the first two experimental days at Monell, the data may reflect their own biases toward continuing to feed a food they deemed unpalatable or misinterpreting signs of satiation from their infants. During the Monell Test Sessions, the infants determined the amount and pacing of feeding and thus these data are more reliable in gauging their acceptance of a food. This underlines the importance of a controlled feeding procedure to assess infant's food acceptance and calls into question data obtained from maternal reports. Because the majority of infants were both breastfed and formula fed, which reflects current feeding practices [28
], we were not able to assess the influence of history of milk feeding on fruit and vegetable acceptance, which is however likely to impact solid foods acceptance as suggested by previous studies [10
]. Nor were we able to determine in the present study why mothers’ ratings of infants’ liking of the foods did not change after the exposure period, despite greater acceptance of the food. However, a recent study which analyzed the infants’ facial expressions during feeding revealed that 8 days of dietary exposure to a particular vegetable did not reduce the number of negative faces made by the infant during feeding [11
]. We hypothesize that facial expressions may have evolved to signal that infants are eating something harmful, and consequently, are more salient cues for the mothers.
The well-documented pediatric obesity epidemic and dramatic increases in its associated clinical diseases are major public health challenges facing most developed countries [30
]. One dietary strategy to prevent or treat obesity couples increasing consumption of foods that are nutrient and fiber rich, yet low in excess energy from added sugars and fats (e.g., fruits, vegetables), with decreasing consumption of energy-dense and less nutritionally desirable foods [31
]. Increased fruit and vegetable consumption early in life may have other long-term benefits. For example, a diet rich in fruits early in life may have a protective effect on cancer risk in adult [34
] and a diet rich in fruits, vegetables, and dairy products during preschool years may have beneficial effects on blood pressure during childhood and adolescence [35
Despite such benefits, recent surveys alarmingly revealed that the intakes of fruits and vegetables are quite low in US pediatric populations [28
]. Three out of ten toddlers do not consume even a single fruit and one in four a single vegetable in a given day; none of the top five vegetables was a dark green vegetable [28
]. Instead, toddlers were more likely to be eating fatty foods such as French fries and sweet-tasting snacks and beverages, flavors that are innately preferred by children [28
The findings from the present study and others conducted in our laboratory [9
] suggest that not only can infants clearly discriminate the flavors of different fruits and vegetables but repeated opportunities to experience a particular fruit or vegetable or a variety of these foods promote the willingness to eat these foods and hopefully, in the long term, preferences for the ‘tastes’ of these foods. Developing strategies to promote the liking of the taste of these foods is extremely important since the best predictor of fruit and vegetable intake is whether children like how it tastes [37
]. The Start Healthy Feeding Guidelines
which were recently developed by an expert panel of pediatric health and nutrition professionals in response to the need for practical, evidence-based advice on introducing solid foods to infants and toddlers concluded that there was no evidence that supports the need to restrict or avoid any foods for weaned infants who are not at risk for allergies [38
]. The present study suggests that by introducing a variety of fruits and vegetables, both within and between meals, infants might be more accepting of fruits and vegetables, the consumption of which is generally low in the pediatric population [28
], and the acceptance of which is difficult to enhance beyond toddlerhood [39
]. Because mothers were apparently unaware of changes in acceptance, they may give up too soon when introducing these foods [10
]. Instead, they should be encouraged to focus on their infants’ willingness to eat the food and to continue to provide their infants with repeated opportunities to taste it as well as others within that food category.
Research has shown that greater diversity of social stimulation during early infancy is associated with higher scores on measures of cognitive development in domains such as language, reading, and mathematics [41
]. The present findings suggest that the beneficial effects of varied experiences during early development are not limited to cognitive functioning and provide experimental evidence to support the claim that experience with diversity of flavors sets the pattern for a diversified diet [42
]. Such information may prove valuable to health care providers and parents in developing strategies for promoting fruit and vegetables. An appreciation of the complexity of early feeding and a greater understanding of the cultural and physiological mechanisms underlying the development of food preferences will aid in our development of evidence-based strategies to facilitate fruit and vegetable acceptance by children.