Child characteristic correlates of food rejection in preschool children: A narrative review

.

One of the reasons for the high prevalence of children's food rejection is that infants are born with a natural preference for sweet taste and an aversion to sour and bitter tastes (Cowart, 1981;Steiner, 1973).Notably, sweet (and salty) taste preference is markedly elevated during childhood (Mennella, Finkbeiner, Lipchock, Hwang, & Reed, 2014;Mennella & Bobowski, 2015).As such, relatively sour and bitter foods (e.g., oranges or kale) are often rejected (Brown, 2010;Wardle et al., 2005).Apart from these taste preferences, children from around 2 years of age are predisposed to reject novel foods, a reaction called food neophobia.Food neophobia and picky eating are two closely related, yet distinct manifestations of food rejection differentiated by the food's novelty status (Rioux et al., 2017(Rioux et al., , 2019)).While food neophobia can contribute to picky eating behaviours, picky eating encompasses a wider range of selective eating behaviours beyond the rejection of novel foods (Dovey et al., 2008).Food neophobia is thought to be functional as it protects the young child from ingesting harmful substances (Pliner & Hobden, 1992).Although food neophobia might have been an advantageous adaptation in the more dangerous environment of hunter-gatherers, this tendency can nowadays, in an environment with an abundance of (safe) food options, hinder the acquisition of a healthy, varied diet.
In the last couple of years, there has been an increase in the number of publications regarding picky eating, with a search for 'picky eating' on PubMed showing an average of 60 new publications per year since 2015 -six times more than the yearly average of the decade before.Relatively little is known about the emergence of food rejection behaviour.Longitudinal evidence suggests that early life factors such as the delayed introduction of lumpy foods (Emmett, Hays, & Taylor, 2018) or greater maternal age (Taylor, Wernimont, Northstone, & Emmett, 2015) may lead to the development of food rejection behaviour.Prenatal or infantile exposure to food flavours through the amniotic fluid or breastmilk may also be implicated in early food preferences (Nicklaus, Schwartz, Monnery-Patris, & Issanchou, 2019).Studies have further identified several social and environmental factors that influence a child's food choice and intake later in life, showing, for instance, that food exposure (e.g., Birch & Marlin, 1982;Rioux, Lafraire, & Picard, 2018b;Wardle et al., 2003) and peer and parental modelling of healthy eating (e.g., Addessi, Galloway, Visalberghi, & Birch, 2005;Birch, 1980, pp. 489-496;Wind et al., 2006) decrease food rejections, whereas parental pressure to eat and an authoritarian or permissive parenting style have been related to more food rejection behaviour (e.g., Galloway, Fiorito, Francis, & Birch, 2006;Podlesak, Mozer, Smith-Simpson, Lee, & Donovan, 2017).While these external influences may all cause or maintain food rejection behaviour, eating behaviour has been shown to be partially determined by genetics (Cooke, Haworth, & Wardle, 2007), and there are considerable differences in eating behaviour between siblings growing up in the same environment (Webber, Cooke, & Wardle, 2010), suggesting that child characteristics may also play a key role in food rejection behaviour.Therefore, it is important to identify child characteristics that may be associated with food rejections and determine for whom a behavioural intervention is warranted and for whom food rejection behaviour may be persistent (Taylor & Emmett, 2019).Moreover, finding out which characteristics are related to the development and persistence of food rejections may facilitate the creation of new, child-centred interventions.
A child may perform several cognitive processes when it decides to accept or reject a food.For instance, they may perceive the food, recognize it or identify it as novel food, imagine how it will probably taste and decide if the reward of eating tasty food outweighs the risk of encountering a foul taste.Individual differences in characteristics implicated in these processes, such as risk-taking, anxiety and disgust sensitivity, might all influence the willingness to try a food.There is still a paucity of studies examining these child characteristics, but these few studies have already revealed interesting findings, identifying distinct reasons for children's food rejections.In the present review, we will provide an overview of these studies, focusing particularly on the relationship between food rejections and cognitive development, chemosensory perception and affective evaluation, food knowledge, decision-making strategies, anxiety and disgust sensitivity in preschoolaged children.To locate relevant articles, Google Scholar and Pubmed were utilized as primary search platforms.We focus on preschoolers since this is the age group where food rejections are most common (Brown, 2010;Cardona Cano, Tiemeier, et al., 2015a, 2015b), but include studies in older children and adults where evidence in preschoolers is still limited.Furthermore, we aim to identify the current gaps in our knowledge regarding characteristics of picky children to outline a future research agenda.

Cognitive development
Food neophobia and picky eating are common in preschool years, with most children experiencing a food neophobic or picky stage during this time (Chao, 2018;Dovey et al., 2008).The age of onset of food neophobia and picky eating is around 2 years old, although picky eating may also arise later in childhood (Mascola et al., 2010).The change in prevalence of food rejections over time remains somewhat unclear, with some studies suggesting a peak around 3-4 years old (Brown, 2010;Cardona Cano, Tiemeier, et al., 2015a, 2015b;Dovey et al., 2008) and others suggesting a plateau in childhood prevalence after their onset (Cooke, Wardle, & Gibson, 2003;Koivisto & Sjödén, 1996;Mascola et al., 2010;Taylor et al., 2015).The onset of food neophobia is thought to be related to the child becoming more independent in their feeding, and a need to protect itself from ingesting toxins, which explains why food neophobia tends to arise around the time the child has developed the strength and locomotor skills for further exploring its surrounding (Cooke et al., 2003;Pliner & Hobden, 1992).As children grow older, they are exposed to many different types of food.This allows them to learn what is food and what is not, thereby gradually increasing their food acceptance (e.g., Bovet, Vauclair, & Blaye, 2005;Lafraire, Rioux, Giboreau, & Picard, 2016;Shutts, Kinzler, McKee, & Spelke, 2009).
In addition to these food exposure effects, researchers have theorized that children's rapid cognitive development during the preschool and early school years may be responsible for the decrease in food rejection behaviour later in childhood, including the development of executive functions (Anderson & Reidy, 2012).Indeed, impairments in emotional control (Zohar, Lev-Ari, & Bachner-Melman, 2019) and cognitive flexibility (Foinant, Lafraire, & Thibaut, 2022a) have been related to food rejections in children.Furthermore, studies have found that children who show higher levels of food rejections have impairments in language development (Qazaryan & Karim, 2019), learning ability (Chao & Chang, 2017), and immature abilities based on food categories (including categorization, cross-classification, inductive reasoning; e.g., Lafraire et al., 2020;Pickard, Lafraire, et al., 2021;Rioux, Picard, & Lafraire, 2016).All of these cognitive abilities show great development during the preschool years, possibly explaining why some studies have shown school-aged children to be less picky eaters than preschoolers (e. g., Cardona Cano, Tiemeier, et al., 2015a, 2015b, but see Mascola et al., 2010, who showed stability of picky eating prevalence until 11 years of age).Adding to this evidence, autism spectrum disordera neurodevelopmental disorder characterized by various cognitive impairments is often accompanied by picky eating behaviour (e.g., Dovey et al., 2019;Harris et al., 2023;Machado, Dias, Lima, Campos, & Gonçalves, 2016;Williams & Seiverling, 2010).Taken together, the prevalence of food rejections may increase in the early preschool years as a result of increasing independence in feeding and may decrease later in childhood as a result of increasing experience with food and cognitive development.

Perception and affective evaluation
The act of eating starts with the perception of food.Given that food varies in a great number of sensations (its colour, texture, flavour, etc.), and individuals differ in the way they perceive these sensations, it is perhaps unsurprising that differences in the perception of sensations have been related to picky eating behaviour (e.g., Coulthard & Blissett, 2009;Steinsbekk, Bonneville-Roussy, Fildes, Llewellyn, & Wichstrøm, 2017).Determinants of these individual differences in perception are a child's sensory threshold and sensory sensitivity.Sensory threshold refers to the point or concentration at which a child is able to detect a sensation, such as taste, touch or smell.When a child has a lower perceptual threshold, it has a higher sensory sensitivity (i.e., is more sensitive to sensations) and is more likely to dislike and avoid them to prevent overstimulation (Dunn, 1997).Food rejections may thus be an avoidance response caused by a low perceptual threshold and a negative affective response to sensations.Studies that have examined this theory for the sensations of taste, touch, andto a lesser extentsmell and vision will be discussed in the following sections.

Taste
Research into the relationship between taste perception and food rejection has focused primarily on individual differences in bitter taste ability.A commonly used method to determine someone's ability to taste bitter is the so-called PROP taste test, in which participants taste a paper strip or solution with the bitter tastant 6-n-propylthiouracil (PROP).Some participants taste the substance (tasters) but some do not (non-tasters; e.g., Bartoshuk, Duffy, & Miller, 1994;Keller, Steinmann, Nurse, & Tepper, 2002).PROP is chemically related to the bitter substance phenylthiocarbamide, a thiourea compound similar to those found in cruciferous vegetables (Bartoshuk et al., 1994;Keller & Adise, 2016).As such, the ability to taste PROP may be related to bitterness perception in vegetables, and in turn influence vegetable intake and preference.Indeed, research generally finds a negative relationship between the ability to taste PROP and vegetable consumption in adults: adults who can taste the bitter substance consume fewer vegetables than those who cannot (e.g., Dinehart, Hayes, Bartoshuk, Lanier, & Duffy, 2006;Duffy et al., 2010;Duffy, Hayes, & Sharafi, 2020; but see Catanzaro, Chesbro, & Velkey, 2013; Shen, Kennedy, & Methven, 2016 for null findings).However, in children, this relationship remains somewhat less clear, with some studies showing that PROP tasters like and consume fewer vegetables than non-tasters do (e.g., Bell & Tepper, 2006;Keller et al., 2002;Turnbull & Matisoo-Smith, 2002) and other studies failing to find this association (e.g., Hoppu, Laitinen, Jaakkola, & Sandell, 2015;Lumeng, Cardinal, Sitto, & Kannan, 2008;Nor, Houston-Price, Harvey, & Methven, 2021).Interestingly, a study by Cole, Wang, Donovan, Lee, and Teran-Garcia (2017) has revealed a relationship between picky eating and variants in the PROP-taster gene TAS2R38 in preschool-aged children, suggesting that children who are genetically pre-disposed to be bitter tasters are more likely to be picky eaters.
Studies examining the relationship between bitter sensitivity and food preferences have utilized the bitter substance quinine to identify differences in bitterness sensitivity in children.For example, Hartvig, Hausner, Wendin, and Bredie (2014) found that children who are sensitive to quinine drank less bitter grapefruit juice (but not a bitter aronia juice) than children who were less sensitive to quinine.A negative relationship has also been found between quinine sensitivity and vegetable liking in adolescents (Hald, Hald, Stankovic, Niklassen, & Ovesen, 2021) and vegetable preference and consumption in adults (Dinehart et al., 2006), with quinine-sensitive individuals liking and consuming fewer vegetables.However, interestingly, Vennerød, Nicklaus, Lien, and Almli (2018) found that preschoolers' sensitivity to quinine was positively related to preference for bitterness in chocolateshowing that quinine sensitive children preferred more bitter chocolatebut was unrelated to bitterness preference in drinks.This study further showed that children who were more sensitive to sweetness preferred more bitter and less sweet chocolates, but preferred less sweet drinks, although an earlier study did not find an association with chocolate preferences (Vennerød, Almli, Berget, & Lien, 2017).Further complicating the relationship between taste sensitivity and food intake is the finding that bitterness sensitivity in children is positively related to preference for and consumption of sweet and fatty foods (e.g., Hoppu et al., 2015;Keller et al., 2002;Keller & Tepper, 2004), suggesting a link between bitter taste sensitivity, sweetness perception and unhealthy food intake.
Together, studies suggest the involvement of sensitivity to bitter tastes in food intake, although the results in children are somewhat mixed and require further research.Nonetheless, interventions aimed at increasing vegetable intake in preschool children might benefit from differentially targeting bitter-sensitive children.For example, bittersensitive children have been shown to consume more broccoli when the broccoli is combined with a dip (Fisher et al., 2012).In addition, Carney et al. (2018) recently found that PROP tasters are more receptive to the use of spices as consumption enhancers than non-tasters.The use of dips and spices may thus provide an interesting starting point for increasing vegetable intake in this bitter-sensitive group.

Touch
Another aspect in which picky eaters have been found to differ from non-picky eaters is their sensitivity to tactile sensations.Texture is an important characteristic of foods, and has been shown to play a large role in determining food preferences in preschoolers (Zeinstra, Koelen, Kok, & De Graaf, 2007).Generally, children have been shown to prefer harder, crunchier vegetables over slimy or soft vegetables (Zeinstra, 2010), and tend to be sensitive to foods containing 'bits' or lumps in it (Wardle & Cooke, 2008;Werthmann, Jansen, Havermans, et al., 2015).However, individual differences in preferences for food texture exist, and these may be explained by differences in tactile processing.
Early research into the relationship between food rejections and tactile sensitivity revealed that tactile defensiveness (i.e., an aversive response to the experiences of touch) was related to lower vegetable preference and greater food rejections in children aged 3-10 years old (Smith, Roux, Naidoo, & Venter, 2005).Similarly, Coulthard and Blissett (2009) found that preschoolers who are more prone to dislike tactile sensations, measured using a parental questionnaire, consumed fewer fruits and vegetables.More recently, research further explored this relationship using behavioural measures of tactile sensitivity, and found that young children who disliked touching various food and non-food items (e.g., a velvet cloth, jelly, or mashed potatoes) had lower enjoyment of food (Nederkoorn, Jansen, & Havermans, 2015) and higher food neophobia scores (Coulthard & Thakker, 2015), although the studies did not find a relationship with parent reported variety of food intake (Nederkoorn et al., 2015) and parent reported fruit and vegetable intake (Coulthard & Thakker, 2015).A concurrent study by Werthmann, Jansen, Havermans, et al. (2015) investigated the effect of variations in colour, taste and texture on yoghurt acceptance, and found that yoghurt with a lumpy texturebut not yoghurt with a different colour or tastewas consumed less than a smooth, plain yoghurt.Finally, Coulthard and Sahota (2016) used a combination of (parental) questionnaires and a behavioural task to measure tactile sensitivity and found a relationship between both of these measures and food neophobia in both preschoolers and adults (see also Nederkoorn, Houben, & Havermans, 2019 for similar findings on picky eating in adults).A recent study confirmed these findings for picky eating in preschoolers using a parental questionnaire but did not find a relationship between picky eating and a tactile sensitivity measure combining child observations and caregiver interviews (Rendall, Harvey, Tavassoli, & Dodd, 2022).
Overall, the relationship between food rejections and tactile sensitivity seems relatively well-established, although it is yet unclear whether this is driven purely by an affective response (i.e., an experienced dislike for certain tactile sensations), or if differences in tactile function contribute to these findings.Regardless, studies have found that tactile-based exposure can be a useful strategy to increase food acceptance.An experimental study by Coulthard and Sealy (2017) revealed that preschool-aged children who engaged in sensory fruit and vegetable play consumed more fruit and vegetables than children who engaged in non-food sensory play or who were visually exposed to the fruits and vegetables.Another study found that children who played with a jelly-like texture with their hands consumed more strawberry jelly, but not smooth strawberry yoghurt or strawberry yoghurt with pieces, compared to children who did not play with the jelly (Nederkoorn, Theiβen, Tummers, & Roefs, 2018).These studies suggest that tactile exposure can be used to increase food acceptance through familiarization with textures.Future research needs to confirm these findings and examine whether they hold in samples of picky eaters and food neophobics.Moreover, research should examine whether the positive effects of tactile exposure-based interventions persist over a longer time and whether they are generalizable to other types of foods beyond those the children have been exposed to.

Smell
A third important sensory determinant of food acceptance is smell.Food odour is strongly intertwined with food flavour, which is why people often report diminished food enjoyment when their sense of smell is impaired (Stevenson, 2010).However, many studies on picky eating to date have not differentiated between taste and smell sensitivity (e.g., Coulthard & Blissett, 2009;Farrow & Coulthard, 2012;Nederkoorn et al., 2015), likely because the questionnaire most commonly used to measure sensory sensitivity (i.e., Sensory Profile; Dunn, 1999) combines sensitivity to taste and smell within the same subscale.Smell and taste are chemical senses but research has shown that reactivity to taste and smellmeasured by ingestion of taste solutions and attraction behaviour towards scentsare not necessarily related and therefore should be examined separately (Monnery-Patris et al., 2015).
Food odour perception covers two routes: orthonasal and retronasal olfaction.Orthonasal olfaction is the perception of volatile (food) molecules from the air, which is important in anticipation of food, whereas retronasal olfaction is the perception of volatile (food) molecules through the mouth, which is important in food consumption and flavour (Boesveldt & de Graaf, 2017).Although food odour and its perception play an important role in appetite and food preference (for an overview, see Boesveldt & de Graaf, 2017;Ginieis et al., 2021), the effects of odour perception on food rejections have received little scientific attention, particularly in children.One study among toddlers found that reactivity to smell, as measured using odorized bottlesbut not tastewas positively related to levels of food neophobia, suggesting that food rejections in neophobic children might not just happen 'on sight', but also 'on smell' (Monnery-Patris et al., 2015).Furthermore, a study by Demattè et al. (2013) found that neophobic adults are worse at identifying odours than non-neophobic adults, possibly because neophobic individuals take smaller food sniffs (Raudenbush, Schroth, Reilley, & Frank, 1998).Since identification of odours relies heavily on previous experience with said odour, it seems likely that neophobic and picky individuals perform worse at odour identification tasks because they tend to avoid foods.Finally, a study by Stafford, Tsang, López, Severini, and Iacomini (2017) found no relationship between food neophobia and odour detection threshold among adults.Note that these findings are not necessarily contradicting the findings from Demattè et al. (2013), since the odour detection threshold only requires participants to note the presence or absence of a smell, rather than identifying the type of scent.Nonetheless, it remains unclear whether-and if so, which aspects ofsmell perception may play a role in food rejections.Future studies among preschoolers, when picky eating is most prevalent, may further elucidate this relationship.

Vision
The visual appearance of food may play an important role in determining whether a food is accepted or rejected (Lafraire, Rioux, Giboreau, & Picard, 2016).For example, green vegetables are especially likely to be rejected by children (Raggio & Gámbaro, 2018), and modifications in food colour have been shown to affect food preferences in picky adults (Annette & Stafford, 2023) but not children (Werthmann, Jansen, Havermans, et al., 2015).Recognition of foods also seems to play a role in food acceptance (see section 4), further highlighting the importance of visual presentation of foods.However, little is known about individual differences in visual sensitivity and its relation to food rejections.Circumstantial evidence does show heightened visual sensitivity in children with autism spectrum disorder (Grandgeorge & Masataka, 2016;Markram & Markram, 2010) -a disorder with high food rejection prevalence (e.g., Harris et al., 2023).Future research should directly examine the relationship between sensitivity to visual sensations and food rejections in typically developing children.
Taken together, it seems that children who display higher levels of food rejection possess an enhanced sensitivity to bitter tastes and tactile sensations, although findings on bitterness sensitivity remain somewhat inconclusive.Findings on the perception of odour and visual sensitivity are limited, but given the importance of odour perception in food flavour and the significance of food appearance, these aspects provide important avenues for future research.If children with higher levels of food rejection are indeed more sensitive to sensations, such heightened sensitivity could potentially extend to their ability to detect differences in food flavour, texture, odour, or appearance compared to what they are familiar with or anticipate, which may lead to difficulties identifying foods and, as will be discussed below, may result in their rejection.However, this has not been empirically tested yet, and further research is needed to validate this hypothesis.

Food knowledge and inferences
As many parents know, children who are picky or food neophobic often refuse foods that are mixed, broken, or from a different brand than they are used to, and prefer packaged foods over fresh foods (e.g., Carruth, Ziegler, Gordon, & Barr, 2004;Dial & Musher-Eizenman, 2019;Harris, Ria-Searle, Jansen, & Thorpe, 2018).This is thought to stem from a child's need to be able to identify foods.Correct identification of foods is essential for determining their post-ingestive consequence (e.g., taste, toxicity, palatability) and thus preventing illness and distaste.To correctly identify foods, children need to acquire experience with and knowledge about them (Mura Paroche, Caton, Vereijken, Weenen, & Houston-Price, 2017).Researchers have posited that exposure to a variety of foods and food preparation methods may lead to an increase in children's food identification ability, and consequently, an increase in food acceptance (Lafraire, Rioux, Giboreau, & Picard, 2016).This increase in food acceptance could in turn lead to better identification of food by exposure to a wider array of foods.
Key to successful identification of foods is an efficient categorization system (Ross & Murphy, 1999).Since no two apples look exactly alike, it is important for children to learn which foods fit into the 'apple' category and which do not to prevent accidently ingesting toxins.There are many ways to cross-classify foods, such as taxonomy (an apple is a fruit), script (an apple is a breakfast food), evaluation (an apple is healthy), or colour (an apple is red; Nguyen & Murphy, 2003;Pickard, Thibaut, Philippe, & Lafraire, 2023.This categorization is based on a child's knowledge and experiences with food, as well as on perceptual characteristics of the food, such as shape and colour.Children differ notably in their ability to categorize foods (Tatlow-Golden, Hennessy, Dean, & Hollywood, 2013).On average, children as young as 3 years old have been shown to be able to discriminate foods from non-foods, even when these were matched on colour and shape (Lafraire, Rioux, Roque, et al., 2016).Children are also able to evaluate different food groups: most preschoolers know that fruits and vegetables make you healthy and strong, whereas junk food does not (Dial & Musher-Eizenman, 2019;Nguyen, 2007Nguyen, , 2008) ) and from 6 years of age children have been shown to differentiate between long-term and short-term consequences of food ingestion (Thibaut, Lafraire, & Foinant, 2020).Differentiation between more similar food groups such as fruits and vegetables also develops during the preschool years (Rioux et al., 2016).Furthermore, research has shown that preschoolers can use different categorization methods interchangeably and use them to make inductive inferences about novel foods (e.g., Nguyen, 2020;Nguyen & Murphy, 2003;Rioux et al., 2016).For instance, children as young as 4 years old know to use taxonomic categories when making inferences about biochemical properties of food (e.g., if an apple contains a specific ingredient, then a pear is more likely than cake to also contain this ingredient), and also know to use script categories when making inferences about situational foods (e.g., if cake is eaten on a holiday, then ice cream is more likely than soup to also be eaten on that holiday; Nguyen & Murphy, 2003).
As successful food categorization is key to correct identification of food, and thus correct inference of the post-ingestive consequences, researchers have proposed a link between children's food categorization performance and food rejection behaviour.Supporting this hypothesis, a series of studies by Rioux and colleagues repeatedly found food rejection to be negatively correlated to preschoolers' food categorization ability (Rioux et al., 2016(Rioux et al., , 2018a(Rioux et al., , 2018b)).For example, preschoolers who were pickier in their eating performed worse at a fruit/vegetable categorization task than children who were less picky (Rioux et al., 2016).Interestingly, recent studies suggest that neophobic or picky children also perform worse at thematic categorization tasks (Pickard, Lafraire, et al., 2021;Pickard, Thibaut, & Lafraire, 2021).In these studies, children were asked to make analogies based on an example given.Children were presented with pictures and descriptions of either a categorical (e. g., applemelon) or a thematic (e.g., breadbutter) food pair, and had to extend this relationship to another food (e.g., a burger patty) by choosing between a categorical match (e.g., chicken) and a thematic match (e.g., a burger bun).The authors found that more neophobic and picky children made more mistakes than less neophobic and picky children.A follow-up study by Pickard et al. (2023) sought to further disentangle this relationship by asking children to make thematic or script inferences about food when the correct match was pinned to either an irrelevant distractor only (study 2) or to an irrelevant distractor and a situationally inappropriate associate (i.e., a thematic match for script trials and a script match for thematic trials; study 3).Their study revealed that food neophobic children make more mistakes when the situationally inappropriate associate was included as a possible answer, but not when only an irrelevant distractor was used, suggesting that the relationship between food rejections and food categorization does not reflect a general lack of conceptual knowledge, but rather a decreased ability to appropriately use this knowledge.Thus, food rejections may be related to an immature categorization system.
Differences have also been found in the way picky eaters make inferences about novel foods.For example, Rioux et al. (2018a), Rioux et al. (2018b) and Lafraire et al. (2020) asked children to infer made-up properties of different vegetables (e.g., 'contains zuline') to another food, which was either the same colour or belonged to the same category.For example, children were shown a picture of a vegetable (e.g., a red bell pepper), and were told that it contains zuline.They were next asked to choose which food they think also contains zuline: a red apple (i.e., a food with the same colour), or a squash (i.e., a food from the same taxonomic category).More picky or neophobic children were more likely to make inferences based on colour similarity, whereas less picky or neophobic children were more likely to use category membership to make inferences.More recently, a study by Foinant, Lafraire, and Thibaut (2021a,b) showed that neophobic children were more likely to generalize negative properties of food (e.g., "gives nausea") to other foods than less neophobic children, even when foods were familiar.Together, these studies indicate that neophobic children may use more appearance-focused and harm-avoidant strategies when making inferences about foods.
In sum, children's abilities based on food categories seem to play a role in food rejection behaviour, specifically regarding food recognition, categorization performance and making inferences about food.Although past studies have provided valuable insights into the cognitive processes related to food rejection, little is known about the mechanism (s) responsible for these relationships.An underdeveloped categorization system as seen in picky eaters may cause these children to have an inaccurate expectation about the taste of the food, which has been related to decreased food liking (e.g., Yeomans, Chambers, Blumenthal, & Blake, 2008).Other studies suggest that a lack of cognitive flexibility could underlie the association between cognitive abilities and picky eating behaviour (Foinant et al., 2022a;Zickgraf, Richard, Zucker, & Wallace, 2020).This may further explain why picky eating is commonly seen in children with autism spectrum disordera disorder characterized by rigid behaviours (e.g., Dovey et al., 2019;Machado et al., 2016;Williams & Seiverling, 2010).Finally, it is important to keep in mind that the relationship between picky eating and food knowledge is thought to be cyclical, with limited food knowledge leading to more food rejections, and more food rejections, in turn, hindering the experience with different foods necessary to increase one's food knowledge.Future studies are necessary to elucidate the causal relationship between food knowledge and food rejections.

Decision-making
When a child is confronted with a novel food, several choices may influence its decision to put it in its mouth.For example, a child may decide whether an item is food or non-food and whether the food is tasty or distasteful.According to the Signal Detection Theory (SDT; Macmillan & Creelman, 2004), this decision-making process is characterized by two components: sensitivity and strategy.Sensitivity refers to one's ability to discriminate stimuli and depends on the prior knowledge and experiences of the individual.Strategy, on the other hand, refers to one's bias when categorizing stimuli, which can be liberal (e.g., 'I don't know a yellow carrot, but I think it is edible/tasty'), or conservative (e.g., 'I don't know a yellow carrot, but I think it is inedible/distasteful').Within the food domain, the costs of incorrectly categorizing an inedible food as edible are much higher than the costs of incorrectly categorizing an edible food as inedible.Accordingly, adopting a conservative strategy towards novel foods may be beneficial for the child's survival.In general, though, children have been shown to adopt a rather liberal decision-making strategy (Foinant et al., 2021a(Foinant et al., , 2021b;;Lafraire, Rioux, Roque, et al., 2016), often incorrectly categorizing non-foods as foods.Nonetheless, individual differences in food decision-making are evident and may explain why some children are pickier than others.A recent study by Foinant et al. (2021a,b) demonstrated that neophobic children categorized fewer (food and non-food) items as foods than non-neophobic children did in a categorization task, and thus adopted a more conservative decision-making strategy.
Food decision-making strategies may be influenced by children's sensitivity to punishment and reward.When confronted with uncertainty about a novel food's edibility or tastiness, children who are sensitive to punishment may be more inclined to avoid novel foods due to the high subjective costs of experiencing an unpleasant taste.Conversely, children who are sensitive to reward may be more inclined to take a risk to try novel foods, as they value the rewarding properties of a tasty food more highly (Crane, Brown, Chivers, & Ferrari, 2020;Vandeweghe, Verbeken, Moens, Vervoort, & Braet, 2016;Vandeweghe, Vervoort, Verbeken, Moens, & Braet, 2016).Many studies have demonstrated a relationship between reward sensitivity and food intake (De Decker et al., 2016;Guerrieri, Nederkoorn, & Jansen, 2008;Nederkoorn, Braet, Van Eijs, Tanghe, & Jansen, 2006, but see Jonker et al., 2019 for null findings).For example, Guerrieri et al. (2008) found that reward-sensitive children consume more calories than less reward-sensitive children when presented with a wide variety of foods.Punishment sensitivity, on the other hand, has received less scientific attention, especially among children.Studies have found that punishment sensitivity and reward sensitivity were positively related to food avoidance and food approach behaviours in preschoolers, respectively (Vandeweghe, Verbeken, et al., 2016;Vandeweghe, Vervoort, et al., 2016).Furthermore, punishment sensitivity has been related to eating disorders (Harrison, O'Brien, Lopez, & Treasure, 2010;Matton, Goossens, Vervaet, & Braet, 2015) and higher sugar intake (Tapper, Baker, Jiga-Boy, Haddock, & Maio, 2015) in adolescents and adults.Thus, food rejections may be caused by a sensitivity to the potential 'punishment' of distasteful foods, although evidence of any causal effect is still lacking.
Another personality trait implicated in the food decision-making process is sensation seeking.Sensation seeking is the tendency to seek out new experiences, and as such, it is strongly related to the aforementioned constructs of reward and punishment sensitivity (Harden et al., 2018).Research has shown that sensation seeking is negatively related to food neophobia in adults (e.g., Alley, Willet, & Muth, 2006;Pliner & Melo, 1997;Sivrikaya & Pekers ¸en, 2020) and older children (Maiz, Balluerka, & Maganto, 2016).This relationship has been explained by sensation seekers' lower levels of general neophobia and higher levels of risk-taking (Alley & Potter, 2011;Pliner & Hobden, 1992).However, to our knowledge, no study has examined whether sensation seeking is related to food neophobia in a population of preschool-aged children.Nonetheless, considering the evidence supporting this relationship in adults and the evidence demonstrating the role of the related construct of reward sensitivity in food rejections in children, it seems likely that sensation seeking also plays a role in food rejection in preschoolers.
Overall, research has demonstrated the involvement of decisionmaking strategies and related personality traits in food rejections.Although there is considerable evidence supporting the relationship between reward sensitivity and food intake, the role of punishment sensitivity and sensation seeking in food rejection has not clearly been demonstrated yet and would thus be an interesting avenue for future research.Furthermore, subsequent studies should focus on adapting food acceptance interventions towards these personality characteristics.If picky eaters are indeed more sensitive to punishment, removing the 'costs' (i.e., the risk of inedibility or distaste) from healthy food consumption may be more effective than using a reward-focused strategy.For example, a recent study found that food processing (e.g., slicing) can be used as a safety cue to help children make decisions about the edibility of a food (Foinant et al., 2021a(Foinant et al., , 2021b)), although these findings do not seem to hold for food neophobic children (Foinant, Lafraire, & Thibaut, 2022b).Alternatively, herbs, spices, or sauces may be added to mask the (dis)taste of foods (Carney et al., 2018;Fisher et al., 2012).

Anxiety
Anxiety is characterized by a state of distress related to threatening thoughts, situations or objects (Barlow, 2004).It is thought to have arisen from evolution to allow for rapid processing of information through activation of the sympathetic nervous system, leading to increased vigilance and fast responding (e.g., through fight, flight, or freeze).Anxiety may be particularly relevant in the food domain, where vigilance about consumption can protect an individual from illness or death (Maratos & Sharpe, 2018).Indeed, research has revealed that preschool-aged children often refer to a fear of negative consequences (e.g., nausea, dying, falling ill) when asked why they reject novel foods (Johnson, Moding, Maloney, & Bellows, 2018).In addition, food neophobics exhibit a physiological response to novel foods similar to the fear response seen in anxious individuals (Raudenbush & Capiola, 2012), and have been shown to display heightened levels of overall anxiety compared to non-neophobics (e.g., Farrow & Coulthard, 2012;Maiz & Balluerka, 2018;Zucker et al., 2015).Moreover, similar techniques have been shown to be effective in the treatment of food neophobia and other phobias (i.e., repeated exposure; e.g., Birch & Marlin, 1982;Rioux et al., 2018b;Wardle et al., 2003).Together, these findings not only suggest that food neophobia can be considered a 'true' phobia but also suggest that further parallels may be drawn between anxiety and food rejections, some of which will be discussed below.

Attentional bias.
Attentional bias refers to the tendency for certain types of stimuli to capture or hold one's attention (Field et al., 2016).Because food rejections in picky eaters and food neophobics are often visually driven, it is perhaps unsurprising that attentional biases have been related to such food rejections (Maratos & Sharpe, 2018).Using a visual probe task, a study by Maratos and Staples (2015) found a heightened attentional bias for novel food stimuli in neophobic compared to non-neophobic children aged 8-11.Additionally, they found that a lower willingness to try the unfamiliar food stimuli was related to greater attentional bias for these foods.However, a subsequent study by Etuk and Forestell (2021) failed to find both of these effects in younger children.Furthermore, a study in adults recently found a larger attentional bias for unappealing or disliked vegetables compared to appealing vegetables, but no relationship was found with food neophobia scores (Agovi et al., 2022).Thus, evidence for the role of attentional biases in food rejections is mixed, and future research is warranted.

Confirmation bias.
Confirmation bias refers to the search for information that confirms one's initial belief while ignoring disconfirming information (Nickerson, 1998).For example, when confronted with a food, picky eaters may search for information to confirm their initial belief that a food is distasteful.A recent study by Dibbets, Borger, and Nederkoorn (2021) examined the relationship between confirmation bias and food neophobia in young adults.In their study, participants were shown pictures of four unknown fruitstwo of which looked tasty and two of which looked distastefuland asked what further information they wanted to know about the fruits (e.g., "How much do you want to know if most people enjoy eating the food?").Results showed that participants requested more negative information for distasteful than for tasteful-looking fruits.However, food neophobia scores were unrelated to this biased information seeking.Thus, confirmation bias does not seem to be implicated in food neophobiaat least not in young adults.To our knowledge, there are no studies that have examined the relationship between food rejections and confirmation bias in children.Future research should strive to replicate findings from Dibbets et al. (2021) in this highly picky population.
6. 1.2.3. Other biases.Given the demonstrated relationship between food rejections and anxiety, other biases may be involved in food rejections as well, but to date have not been studied in this context.One of such biases which may be an interesting avenue for future research is the negative interpretation bias, which is the tendency to interpret ambiguous situations or stimuli in a negative way (Stuijfzand, Creswell, Field, Pearcey, & Dodd, 2018).For example, picky or neophobic children might be more likely to misinterpret their tummy ache as being the result of the carrots they ate rather than the flu, or they may misinterpret their mother's negative facial expression as resulting from her not liking the food rather than her having had a bad day.This would be in line with the previously discussed SDT perspective, which suggests that uncertainty about food (e.g., due to its atypical colour), can lead to its rejection (see section 5).Thus, a negative interpretation bias might show similarities to a conservative decision-making strategy.We speculate that the interpretation bias may be particularly relevant to food neophobia, since new foods are generally more ambiguous than familiar foods.Nevertheless, even familiar foods may be ambiguous when presented in an unfamiliar meal context; therefore, the interpretation bias may play a role in picky eating as well.Similarly, it would be interesting to examine whether memory biasthe tendency to selectively recall memories congruent with an emotional state (Muris & Field, 2008) plays a role in food pickiness.For example, picky children might remember food experiences better that evoke a negative emotional response (e.g., fear or disgust), which are involved in memory biases in non-picky populations (Muris & Field, 2008;Schienle, Potthoff, Schönthaler, & Schlintl, 2021).Finally, future research could focus on the role of the recognition heuristic, or the innate preference for recognised stimuli, in food rejection behaviour.Specifically, food neophobia may be a consequence of this heuristic evolved to prevent accidental ingestion of toxins (Goldstein, 1998;Goldstein & Gigerenzer, 1999).It would be interesting to examine if food neophobic and picky children rely on the recognition heuristic more often than neophillic and non-picky children do, and whether they do so only within the food domain or across multiple domains.This may explain picky children's preference for packaged and uniform foods (e.g., Carruth et al., 2004;Dial & Musher-Eizenman, 2019; see section 4).Overall, research into the role of information-processing biases is limited, and further research in these areas may shed light on the underlying and possibly differential causes of picky eating and food neophobia.

Disgust
Food rejections may also be driven by another response: disgust.Disgust is an emotional response to stimuli that are (potentially) contagious, and it is thought to have arisen from a fear of contamination (Rozin & Fallon, 1987).As such, it is highly relevant in the food domain, where individuals risk poisoning through the ingestion of contaminated foods.Individuals differ in their sensitivity to disgust, and this disgust sensitivity has repeatedly been related to picky eating behaviours in adults (e.g., Al-Shawaf, Lewis, Alley, & Buss, 2015;Ammann, Hartmann, & Siegrist, 2018;Egolf, Siegrist, & Hartmann, 2018;Kauer, Pelchat, Rozin, & Zickgraf, 2015;Çınar, Karinen, & Tybur, 2021).A heightened disgust sensitivity might thus explain why picky eaters reject so many (potentially contaminated) foods.
However, unlike adults, preschoolers have a limited understanding of contamination (e.g., Fallon, Rozin, & Pliner, 1984;Rozin & Fallon, 1987).Reasoning about contamination is thought to require an understanding of the presence, spread, and consequences of exposure to pathogens, which are often invisible and thus difficult for children to understand (Rozin & Fallon, 1987).For example, DeJesus, Shutts, and Kinzler (2015) found that 3-4-year-old children are equally willing to eat food that they believe has been sneezed upon or licked compared to a clean plate, whereas older children consumed less of the contaminated food and evaluated it more negatively.Likewise, preschool-aged children have been shown to willingly eat or drink contaminated substances once these contaminations (e.g., a bug or hair) have been removed (Fallon et al., 1984;Rozin et al., 1985Rozin et al., , 1986)).
Although young children seem to lack concern for pathogens, they have been shown to be concerned with other types of contamination.For instance, studies have shown that children's liking for a food decreases once it has touched a disliked food (Brown, 2010;Brown & Harris, 2012a, 2012b).However, no relationship was found in these studies between this disgust response and food neophobia.On the other hand, a study by Kutbi et al. (2019) did find a relationship between disgust sensitivity and food neophobia/pickiness in preschoolers, although in this study child disgust was measured using a parental measure, which relies on parental interpretation of a child's response and as such, may not be an accurate depiction of a child's actual experienced disgust.Furthermore, contamination-based food rejections in young children may occur when foods are touching (Rioux et al., 2018a,b), and children have been found to show disgust reactions to culturally inappropriate foods or food combinations (De Moura, 2007; DeJesus, Gerdin, Sullivan, & Kinzler, 2019, but see Rozin et al., 1986, who showed that young children generally accept inappropriate food combinations).Finally, an important source of information to determine potential food contamination is food texture (i.e., lumpy textures can reflect spoiled food; Werthmann, Jansen, Havermans, et al., 2015), which has emerged from the literature as one of the reasons for child food rejections (see section 3.2 Touch above).
In sum, young children's food acceptance is unaffected by pathogenlike forms of contamination, but they do show disgust reactions to foods 'contaminated' by disliked foods and food textures.Further research is needed to establish whether differences in disgust sensitivity explain differences in food rejection behaviours in preschoolers.The information-processing biases discussed earlier could provide a mechanism explaining this relationship by altering a child's processing of disgust-relevant information about foods (Knowles, Cox, Armstrong, & Olatunji, 2019).

Discussion
In sum, food neophobia and picky eating are important forms of food rejection that often target fruits and vegetables.As described in this review, picky eating behaviour may be explained by several cognitive processes, including differences in taste and tactile perception and affective evaluation, categorization ability, sensitivity to punishment, anxiety, or disgust.Current evidence shows that taste or tactile-sensitive children are more likely to be picky eaters than less sensitive children, that picky eaters have an immature categorization and induction system compared to non-picky eaters, and that food neophobic children show increased anxiety compared to non-neophobic children.Preliminary evidence further suggests that odour and visual perception, punishment sensitivity, sensation seeking, information-processing biases and disgust sensitivity may be involved in food rejections, although future studies need to confirm and elucidate these findings.
The finding that cognitive processes play a role in food rejections has implications for our understanding and treatment of food rejection behaviour.Whereas most research on food rejections focussed on external explanations for the behaviour (e.g., peer and parental influences), these cognitive processes help us explain food rejections from the perspective of a child's mind.This can not only explain differences in pickiness between children growing up in the same environment, but also allows for the development of new, child-centred interventions.For example, if food rejections can be explained in terms of an immature categorization system (c.q.Chao & Chang, 2017;Qazaryan & Karim, 2019;Rioux et al., 2016), then teaching children to rely on taxonomic category membership (e.g., both foods are vegetables) rather than perceptual similarities (e.g., both foods are red) to draw inferences about food might lead to fewer food rejections, and in turn a more varied, healthier diet and fewer health problems later in life.Furthermore, these findings allow for identifying at-risk children and targeting interventions towards this specific at-risk group.Finally, understanding food rejections from a cognitive perspective may help relieve some stigma from parents and caregivers as they are now often held responsible for their child's picky eating behaviour (Miller, Miller, & Clark, 2018;Seshan et al., 2022).
There are several important limitations in the current body of literature about the role of child characteristics in food rejection behaviour.First, most studies described in this review are cross-sectional in nature.This limits the interpretation of the findings in terms of causality and does not reveal insight into the developmental trajectories of food rejections.It is important to define a timeframe to more fully understand food rejection behaviour, as some cognitive processes may be more important for the onset of food rejections, whereas others may be more important in food rejection persistence.Longitudinal studies are necessary to reveal which of these factors influence food rejections at which point in development.Moreover, future studies should strive to identify whether the child characteristics described in this review are predictors of food rejections, or whether they are a consequence of picky children's limited food repertoire.Furthermore, future studies should examine if the correlates discussed in this paper hold for both picky eating and food neophobia, or if they may differentially relate to these types of food rejection.
Finally, research should focus on whether and how we can target these domains to decrease childhood food rejection behaviour.For example, future studies could assess whether increasing children's knowledge about food or teaching them to use more abstract categorization strategies could increase food acceptance.Although previous research has already demonstrated a positive relationship between food literacy and healthy diets, effects from food literacy interventions are generally small and data on its long-term effectiveness is limited (e.g., Gripshover & Markman, 2013;Hersch, Perdue, Ambroz, & Boucher, 2014, pp. 2003-2014;Savoie-Roskos, Wengreen, & Durward, 2017).These interventions thus require further optimisation to ensure practically significant increases in food acceptance in young children.Furthermore, research could assess whether attentional bias modification techniques may be used to redirect picky children's attentional bias away from the fear-evoking properties of healthy foods.Such studies could, in turn, be used in the development of interventions and provide guidelines to caregivers aimed at managing child picky eating behaviour.It is important to note that child food rejections are complex and influenced by a multitude of factors with varying degrees of significance for individual children.For instance, one picky eater may highly value the ability to recognize food, whereas another may be sensitive to the distaste of foods.Consequently, interventions that are effective for one child may not be effective for another.Further, it is important for future studies to examine potential child-environment interactions to form a more complete understanding of food rejection behaviour.

Conclusion
In conclusion, the present review highlights current evidence and promising lines of research into child characteristics that could be implicated in young children's food rejection.Many factors, including differences in sensory perception and affective evaluation, categorization ability, sensitivity to punishment, anxiety, and disgust, appear to play a role in picky eating behaviour, which provides useful starting points for further research and development of interventions.It is unclear if there is a dominant factor explaining child food rejections or whether different types of food rejections in children can be clustered according to their emotional and sensory sensitivity or cognitive development.Since childhood is the optimal time to learn healthy food habits, further understanding food rejection behaviour is not only relevant for children and parents today, but may also lead to better health later in life.In a time where unhealthy diets and obesity have become a global health concern, understanding children's food rejections may be key to establishing a healthier society.