to a higher acceptance of the specific food the infant was exposed to in mother’s milk when this food is offered at the beginning of complementary feeding [20] and to a positive effect of breastfeeding on acceptance of a new food, even in the absence of a specific flavor exposure [22]. This work suggests that the variety of flavor exposure in breast milk may be key and not only the specific exposure to a given flavor.
To expand this building knowledge on the effect of early taste and flavor exposure (i.e., dietary exposure to a variety of tastes in foods), in the OPALINE cohort we wondered about the effect of exposure to specific tastes in the context of milk feeding. It has been described that breast milk contains much more glutamate than formula milk, a compound that is associated with the umami taste. In this context, we observed that infants who were breastfed longer had a higher preference for the umami taste at the age of 6 months [23]. Moreover, concomitant with the introduction of complementary foods which are characterized by a variety of tastes, taste differential reactivity (within-subject variability across tastes; the higher the score, the greater is the difference in reaction of the infant to the tastes) clearly increased [14].
In order to develop our understanding of infant dietary taste exposure, we developed a method to evaluate the taste properties of the diet by describing extensively the intensities of the tastes of all the foods consumed monthly by infants relying on data from “food taste databases” [24]. The application of this method showed that the exposure of French infants to tastes was dominated by exposure to sweet taste over the first year, as illustrated in Figure 1 (in relation to the sweet taste of milk and a number of foods introduced after the initiation of complementary feeding) [25]. We expanded these findings by comparing the dietary exposure to sweet taste and fattiness [26]. They both increased during the first year in relation to the introduction of complementary foods in the child’s diet, but exposure to sweetness increases more rapidly than exposure to fattiness.
However, we have not shown any specific associations between dietary exposure to each taste and acceptance of each taste measured in water, as explained above [27]. The longer-term associations between early exposure to taste and further taste preferences are currently being explored in the OPALINE cohort.
Concerning olfactory exposure, we have found that at 8 months of age only, positive correlations were found between liking of some unpleasant odors and early exposure to these odors through the mother’s diet. However, no correlations were found between infants’ liking of the pleasant odors and early exposure to the foods bearing these odors. This study highlights that early exposure to unpleasant food odors may increase subsequent liking (or reduce subsequent dislike) of these food odors at least until the age of 8 months [28].
Fig. 1. Dietary exposure of infants to the 5 primary tastes over the first year based on dietary data collected in the French OPALINE cohort (adapted from Schwartz et al. [25] with permission).
The originality of the OPALINE study was to evaluate the effect of early odor exposure on the infant through the mother’s spontaneous consumption of a wide spectrum of foods without asking mothers to consume one target food bearing a specific odor quality in sizeable amounts. Such “ecological” research regarding the influence of prenatal and dietary taste and flavor exposure on the establishment of preferences for foods with similar tastes and flavors is still in its beginning. More studies in different cultural contexts were the flavor experience is likely to be contrasted are needed to more completely understand these early imprinting phenomena.
The Influence of Taste and Odor Preferences on Infants’ Eating Behavior
To further explore the question of the association between taste or flavor preferences and food preferences, we further explored data from the OPALINE cohort. This was evaluated separately for taste and flavor.
Concerning taste, we characterized the taste intensities of all foods introduced at the beginning of the complementary feeding period. In parallel, mothers were asked to report their infant’s acceptance of these first foods. We applied classification methods to the taste profiles of the foods, which showed that 15 groups of foods with similar taste profiles had been given to children (e.g., salty foods, sour foods, and sour and sweet fruits) [29]. Then, we compared the average acceptance of all food groups and showed that the acceptance of new foods varied according to their taste profiles. More specifically, we showed that the acceptance of salty vegetables was higher than the acceptance of bitter vegetables, or of sweet and bitter vegetables [29]. Furthermore, we showed that a higher preference for sweet, sour, and umami tastes was associated to a higher acceptance of some sweet-, sour- and umami-tasting foods, respectively [29]. This supports the hypothesis that the preference for some foods was partly related to the specific preference for their taste properties. Finally, an exploration of the factors related to vegetable acceptance along the first 2 years showed that a higher bitterness acceptance was associated with a higher acceptance of vegetables at the age of 1 year [30].
Concerning flavors, we similarly explored the acceptance of foods bearing specific odors and evaluated whether their acceptance was related to the acceptance of the corresponding odors (presented alone in scented bottles). We found that at 12 months rejection of the odor of trimethylamine and dimethyl disulfide was related to the rejection of fish and sulfurous cheeses between 12 and 15 months, respectively [31]. Therefore, in the case of olfaction, the flavor-food acceptance associations concerned foods with strong, unpleasant flavors at 12 months only, suggesting that the olfactory system acts as an “alarm” system during this period of food transition.
Eating behavior in children is characterized by its evolution as far as food neophobia is concerned. By the end of the second year, neophobic reactions start happening and can also be designated as picky eating behavior. This developmental phase concerns most children, but we were interested in evaluating whether neophobic reactions could be related to differences in taste and smell acceptance. To evaluate this aspect, we considered taste differential reactivity by computing within-subject variability across tastes as well as flavor differential reactivity by computing within-subject variability across flavors. We found that at 20 months, food neophobia was associated to flavor differential reactivity but not to taste differential reactivity [32], stressing the importance of olfaction in the development of neophobic reactions.
Finally, we have developed a working model of chemosensory, experiential, and environmental factors likely to influence food likes at the age of 2 years (Fig. 2). In this model, we included parental feeding practices and feeding style in the environmental and experiential factors likely to influence likes. The evaluation of this model showed that most of the factors
