Children’s reasoning on food properties and health relationships can contribute to healthier food choices. Food properties can either be positive (“gives strength”) or negative (“gives nausea”). One of the main challenges in public health is to foster children’s dietary variety, which contributes to a normal and healthy development. To face this challenge, it is essential to investigate how children generalize these positive and negative properties to other foods, including familiar and unfamiliar ones. In the present experiment, we hypothesized that children might rely on cues of food processing (e.g., signs of human intervention such as slicing) to convey information about item edibility. Furthermore, capitalizing on previous results showing that food rejections (i.e., food neophobia and picky eating) are a significant source of inter-individual variability to children’s inferences in the food domain, we followed an individual approach. We expected that children would generalize the positive properties to familiar foods and, in contrast, that they would generalize more often the negative properties to unfamiliar foods. However, we expected that children would generalize more positive and less negative properties to unfamiliar sliced foods than to whole unfamiliar foods. Finally, we expected that children displaying higher levels of food rejections would generalize more negative properties than children displaying lower levels of food rejections. One-hundred and twenty-six children, aged 3–6 years, performed an induction task in which they had to generalize positive or negative health-related properties to familiar or unfamiliar foods, whole or sliced. We measured children’s probability of generalization for positive and negative properties. The children’s food rejection score was assessed on a standardized scale. Results indicated that children evaluated positively familiar foods (regardless of processing), whereas they tend to view unfamiliar food negatively. In contrast, children were at chance for processed unfamiliar foods. Furthermore, children displaying higher levels of food rejections were more likely to generalize the negative properties to all kinds of foods than children displaying lower levels of food rejections. These findings entitle us to hypothesize that knowledge-based food education programs should take into account the valence of the properties taught to children, as well as the state of processing of the food presented. Furthermore, one should take children’s interindividual differences into account because they influence how the knowledge gained through these programs may be generalized.

Preschoolers’ neophobic dispositions mainly target fruits and vegetables. They received a great deal of attention in the past decades as these dispositions represent the main psychological barrier to dietary variety. Recently, children’s food neophobia has been found to be negatively correlated with their categorization performance (i.e., the accuracy to discriminate between food categories). We investigated categorization strategies among neophobic children, tendencies to favor one type of error over the other (misses over false alarms), in order to compensate for their poor categorization performance. To capture children’s categorization strategies, we used the Signal Detection Theory framework. A first experiment assessed 120 3-to-6-years old children’ sensitivity to discriminate between foods and nonfoods as well as their decision criterion (i.e., response strategy). In a second experiment, we manipulated the influence of food processing. The hypothesis was that food processing acts as a sign of human interventions that decreases uncertainty about edibility and thus promotes feelings of safety in the food domain. 137 children were tested on a food versus nonfood categorization task contrasting whole and sliced stimuli. In both experiments, increased levels of food neophobia were significantly associated with poorer categorization sensitivity and with a more conservative decision criterion (i.e., favoring “it is inedible” errors). Additionally, results from Experiment 2 revealed that food processing did not influence neophobic children, whereas their neophilic counterparts adopted a more liberal decision criterion for sliced stimuli than for whole stimuli. These findings are the first demonstration of a relationship between a decision criterion and food neophobia in young children. These results have strong implications for theories of food neophobia and laid the groundwork for designing novel types of food education interventions.

No abstract