Abstract

Food preferences are a type of feeding behavior: the ordered likes and dislikes that decide, among the foods available, which a person will choose and eat. Cognitive psychology treats them as the product of three interacting forces—a small set of unlearned hedonic biases present at birth, a powerful learning system that reshapes those biases through experience, and stable individual differences, part genetic, in how foods are perceived in the first place. This article develops that account: why newborns accept sweet and reject bitter, how mere exposure and flavor–nutrient association convert neutral or disliked foods into preferred ones, and why a single bitter vegetable can taste sharply different to two children because of variation in one taste-receptor gene. Three demonstrations make the newborn hedonic response, the exposure–preference curve, and the genetics of bitter perception concrete.

Keywords: food preferences, flavor learning, taste perception

Food Preferences is the heading the National Library of Medicine uses for the selection of one food over another, and it names something so ordinary it is easy to overlook as a psychological problem: everyone has foods they like and foods they do not, and those likes govern much of what they actually eat. Yet the ordering is neither fixed nor arbitrary. A short list of taste responses is present before any food has been tasted; almost everything else—a liking for coffee, for chili heat, for a regional cuisine—is learned, often against an initial rejection. And the same food is not perceived identically by everyone, because the sensory apparatus that preferences are built on varies from person to person. Those three threads—an innate starting point, a learning process that rewrites it, and individual variation in perception—are what make food preferences a problem for cognitive psychology rather than only for nutrition, and they are the structure this article follows.

Key Takeaways

  • Food preferences are the MeSH heading for the selection of one food over another; cognitive psychology studies them as the joint output of unlearned taste biases, learning, and individual perceptual differences.
  • A few hedonic responses are innate—acceptance of sweet and savory (umami), rejection of bitter and sour—and present at birth, before any food experience.
  • Most preferences are learned: repeated exposure raises liking, and pairing a flavor with calories or with an already-liked taste conditions a durable preference for it.
  • Flavor learning begins before birth and in breast milk, as flavors from the maternal diet reach the fetus and infant and raise later acceptance of those flavors.
  • People differ in the taste world they inhabit: variation in the bitter-receptor gene TAS2R38 makes some perceive certain vegetables as intensely bitter and others as nearly tasteless.

What Food Preferences Are

Within the Medical Subject Headings, Food Preferences denotes the selection of one food in preference to others—the ranked set of likes and dislikes an individual brings to the table. It is a facet of feeding behavior, the broader category of how organisms select, accept, and consume food, and it is what translates the physiological pressure of hunger into the specific act of eating this food rather than that one. Preference is not the same as intake—availability, price, and habit intervene—but across the choices a person is free to make, measured liking is among the strongest predictors of what they actually eat (Drewnowski, 1997).

A preference, in this literature, is a learned or unlearned hedonic ordering: a food is assigned a position on a like–dislike scale, and that position guides approach or avoidance. Three questions organize the study of how such orderings arise. First, what is present at the start, before experience—the unlearned biases a newborn brings? Second, how does experience change them—the learning that turns a rejected food into a staple? Third, why do two people exposed to the same foods end up with different preferences—the individual differences, partly perceptual and partly genetic, that make the same carrot taste different to different children? The omnivore faces these questions acutely: a diet open to almost anything is also open to poison, so an omnivore needs both caution toward the unfamiliar and a way to learn, from experience, what is safe and worth eating (Rozin & Fallon, 1987). The sections that follow take the innate starting point, the learning process, and the individual differences in turn (Figure 1).

Figure 1

The three forces through which cognitive psychology accounts for food preferences.

An innate hedonic starting point, reshaped by learning, filtered through individual perceptual differences A left-to-right flow. On the left, an innate taste bias box lists acceptance of sweet and umami and rejection of bitter and sour, the unlearned starting point at birth. A central arrow labelled learning, covering mere exposure and flavor-nutrient association, leads to an acquired preference box on the right. Above the flow, a box labelled individual differences, including the TAS2R38 bitter-receptor gene, modulates how each food is perceived before learning acts on it. Innate taste bias accept sweet, umami reject bitter, sour Acquired preference learned liking guides food choice learning exposure · flavor–nutrient Individual differences TAS2R38 · perception varies
Note. A small set of unlearned hedonic responses is the starting point; learning—repeated exposure and the association of flavors with calories or with liked tastes—rewrites it into the acquired preferences that guide choice; and individual differences in perception, such as variation in the bitter-receptor gene TAS2R38, change how each food is experienced before learning acts on it. The three forces are the subject of the three sections and three demonstrations that follow.

The Innate Palate: Unlearned Taste Biases

Preferences do not begin from zero. Within hours of birth, before any food has been tasted, infants show organized hedonic responses to the basic tastes: sweet solutions are accepted and elicit relaxed, positive facial expressions, while bitter and sour solutions provoke rejection—gaping, grimacing, head-turning—and salty and savory (umami) tastes are accepted at moderate concentrations (Mennella, 2014). These responses are not preferences for particular foods; they are biases over tastes, and they make adaptive sense. Sweetness signals readily available energy and the sugars of ripe fruit and mother's milk; savory signals protein; bitterness is the taste of many plant alkaloids and is therefore a built-in warning, and sourness can signal unripeness or spoilage (Drewnowski, 1997).

The strength of these biases is itself developmental. Children prefer higher concentrations of sweet and salt than adults do, and this heightened liking recedes over adolescence—a trajectory tied to growth rather than to culture alone (Mennella, 2014). Sourness is the clearest case of a response that is not simply aversive: although newborns reject strong sour tastes, a substantial minority of children come to prefer intensely sour flavors, more than adults do, showing that even the unlearned responses are a starting point open to revision rather than a fixed verdict (Liem & Mennella, 2002). The affective grammar of these responses—the way distaste and, later, disgust organize what is rejected—is the same one that structures food rejection throughout life (Rozin & Fallon, 1987). The first demonstration varies the concentration of each basic taste and shows the modeled newborn hedonic response, making the shape of the innate biases—monotonic liking for sweet, rejection rising with bitterness—concrete.

The innate palate: how a newborn responds to each basic taste

Pick a basic taste and vary its concentration to read the modeled newborn hedonic response, scaled from strong rejection (−1) through neutral (0) to strong acceptance (+1). Sweet and umami are accepted and climb with concentration; salt is accepted only in a moderate band; sour and bitter are rejected, bitter the most steeply. Shapes after the developmental taste literature; illustrative.

neutral+1+0.50-0.5-10255075100reject ← hedonic → acceptTaste concentration (%) →
At 50% concentration, the modeled hedonic response to Sweet is 0.74 — the taste is accepted. Acceptance rises with concentration, the built-in signal for energy and protein.

Learning What to Like: Exposure and Association

If the innate palate is a short list, the adult preference ordering is long and specific, and the difference is made by learning. The simplest mechanism is familiarity. In a classic experiment, two-year-olds offered novel foods came to prefer the ones they had tasted most often (Birch & Marlin, 1982), an instance of the mere exposure effect—Zajonc's finding that repeated, unreinforced encounters with a stimulus reliably increase liking for it (Zajonc, 1968)—operating on flavor. The practical dose is larger than caregivers expect: systematic reviews of repeated-exposure studies in infants and toddlers find that acceptance of an initially rejected food typically rises only after on the order of eight to fifteen tastings (Spill et al., 2019).

Exposure is not the only route. Flavors acquire value by association, through two forms of classical conditioning. In flavor–flavor learning, a new flavor paired with an already-liked taste (sweetness, for instance) becomes better liked itself. In flavor–nutrient learning, a flavor reliably followed by the post-ingestive consequences of calories—energy, satiety—gains preference, so that energy-dense foods come to be liked for their flavor as well as their effect (Birch, 1999). This conditioning is a form of reinforcement: the nutrient is the reward that raises the flavor's value. It also explains a counterintuitive clinical fact—pressuring or rewarding a child to eat a food tends to lower long-run liking, because it reframes the food as a cost rather than letting its own flavor and consequences do the conditioning (Scaglioni et al., 2018). Association also builds the aversive side of the ledger. A flavor followed by nausea acquires a strong, durable dislike in a single pairing—conditioned taste aversion, the one-trial learning Garcia and Koelling demonstrated when rats associated a novel taste, but not an audiovisual cue, with later illness (Garcia & Koelling, 1966). Its biological preparedness (taste pairs with sickness far more readily than with pain) and its resistance to extinction make it the mirror image of appetitive flavor–nutrient learning, and a lifelong food dislike traceable to one episode of illness is its everyday signature.

Remarkably, this learning starts before the first meal. Flavors from the maternal diet—garlic, carrot, vanilla—pass into amniotic fluid and later into breast milk, and infants exposed to a flavor by these routes accept it more readily when it appears in solid food (Mennella, Jagnow, & Beauchamp, 2001). Early exposure to a variety of flavors, in utero, in milk, and at weaning, is among the most consistent predictors of a child's later willingness to accept new and varied foods (Ventura & Worobey, 2013; Nicklaus & Schwartz, 2019). The second demonstration traces the exposure–preference curve: how modeled liking for a food climbs with repeated pairings, how the climb depends on where the food starts and how strong the association is, and why early exposures move liking faster than later ones.

The exposure–preference curve: how liking is learned

Modeled liking for a food after n repeated tastings: pref(n) = A − (A − s)·e−kn, with the asymptote fixed at A = 0.85. Set the starting liking s and the association strength k, then move the marker to read liking at any number of tastings. Early exposures move liking fastest. Illustrative model.

asymptote 0.85accepted 0.600.000.250.500.751.000481216Liking (0–1)Number of repeated tastings (n) →0.61
After 3 tastings the modeled liking is 0.61. Half the total gain arrives by 2.3 tastings, and liking crosses the 0.60 accepted threshold at 2.9 tastings. The curve is steepest at the start: the first few exposures do most of the work.

Individual Differences: Genes and the Bitter Taste

Two children can be offered the same vegetables, with the same exposure, and still diverge—because they do not taste the same food. The clearest instance is bitterness. Sensitivity to the bitter compounds phenylthiocarbamide (PTC) and propylthiouracil (PROP) is bimodal in the population: some people perceive them as intensely bitter, others barely at all. The variation was traced to a single gene, TAS2R38, which encodes a bitter taste receptor; common variants at a few positions define two main forms—a sensitive (“taster”) and an insensitive (“non-taster”) haplotype—and genotype at this one locus accounts for most of the difference in who tastes PTC (Kim et al., 2003).

Because the two haplotypes segregate like simple alleles, their combinations fall, to a first approximation, into Hardy–Weinberg proportions. With the sensitive and insensitive forms at roughly equal frequency, about one quarter of people carry two sensitive copies (most bitter-sensitive), about half carry one of each (intermediate), and about one quarter carry two insensitive copies (least sensitive). This matters for preference because the bitter compounds the receptor detects occur in brassica vegetables—broccoli, Brussels sprouts, kale—so a bitter-sensitive child meets a genuinely more aversive vegetable than an insensitive child does, and the perceptual difference shows up in liking and intake (Mennella, 2014). Genetics is not the whole of individual difference—sex and age differences in children's food preferences are reliable, with vegetables among the least liked and sweets among the most across groups (Cooke & Wardle, 2005)—but TAS2R38 is the textbook demonstration that a preference can rest on a perceptual difference written in the genome. The third demonstration sets the two haplotype frequencies, shows the resulting distribution of taster genotypes, and traces how a shift in perceived bitterness moves the modeled acceptance of a bitter vegetable.

TAS2R38 genetics: why the same vegetable tastes different

Set the frequency of the sensitive TAS2R38 taster allele; Hardy–Weinberg gives the share of the population with each genotype (bars, left scale as a share of the population height). Set how bitter the vegetable is to see modeled acceptance per genotype (dots). At equal allele frequencies the split is the familiar 25/50/25. Illustrative model.

0255075100share / acceptance %25%25%PAV/PAVsensitive taster50%50%PAV/AVIintermediate25%75%AVI/AVIinsensitive non-tasterpopulation mean 50%
Bars: genotype share of population  •  Dots: modeled acceptance of the bitter vegetable
With the taster allele at 50%, about 25% of people are sensitive tasters, 50% intermediate, and 25% insensitive. At bitterness 1.00, the sensitive group accepts the vegetable only 25% of the time versus 75% for the insensitive group, for a population mean of 50%. The gap between genotypes widens as the vegetable gets more bitter — and vanishes when it is not bitter at all.

Worked Example

The exposure–preference curve in the second demonstration can be read off a single equation, and working one case by hand shows why caregivers routinely give up too soon. Model liking after n tastings as an exponential approach to an asymptote:

p(n) = pmax − (pmax − p0) × e−kn

where p0 is the initial liking before any exposure, pmax is the ceiling liking the food can reach, and k is the learning rate set by how reinforcing each tasting is. Take a food a toddler starts out disliking, p0 = 0.25 on a 0–1 liking scale, with a ceiling of pmax = 0.85 and a moderate learning rate k = 0.30 per tasting. At the first tasting, p(1) = 0.85 − 0.60 × e−0.30 = 0.41; by the third, p(3) = 0.85 − 0.60 × e−0.90 = 0.61. The half-gain point—where liking has closed half the distance from p0 to pmax—falls at n = ln 2 / k = 2.31 tastings, and the curve first crosses an “accepted” threshold of 0.60 at n = ln(0.60 / 0.25) / k, here about 2.9 tastings. Table 1 traces the full curve.

Table 1

Modeled liking across repeated tastings for a food starting at p0 = 0.25, with pmax = 0.85 and k = 0.30.

Tastings (n) Modeled liking p(n) Gain closed toward ceiling Status
0 0.25 0% rejected
1 0.41 26% rejected
3 0.61 59% just accepted
5 0.72 78% liked
8 0.80 92% liked
15 0.84 99% liked

Note. The curve is steepest at the start—the first three tastings close 59% of the possible gain—so the common practice of abandoning a food after one or two refusals discards most of the available learning. Acceptance here crosses 0.60 near the third tasting, consistent with the eight-to-fifteen-exposure figure from intervention trials once a lower starting point or smaller k is used (Spill et al., 2019).

Discussion

The three forces are not independent; they compose. The innate biases set the initial hedonic value of a food, individual differences in perception change what that value is for a given person, and learning moves it from there. A bitter-sensitive child starts a brassica lower on the like–dislike scale and may need more exposures to climb to acceptance, but the climb is still available—the perceptual handicap raises the learning cost, it does not lock the preference (Mennella, 2014; Spill et al., 2019). This composition is why food preferences are a cognitive problem and not merely a sensory one: the same sensory input yields different stable preferences depending on the history of exposure and association laid over it.

It also clarifies what a preference is. It is not a fixed property of a food, nor a free choice made afresh at each meal, but a learned hedonic memory—an ordering built up from experience and anchored to the body's unlearned responses and the individual's perceptual apparatus. Seen this way, the applied questions that dominate the nutrition literature—how to raise vegetable intake, how to prevent the entrenchment of preferences for energy-dense foods—are questions about how to arrange exposure and association so that the learning system builds the orderings one wants (Scaglioni et al., 2018; De Cosmi, Scaglioni, & Agostoni, 2017).

Current Directions

The active research front is increasingly intervention-focused and quantitative. A central question is how to translate the repeated-exposure finding into practice at scale: a systematic review and meta-analysis of strategies to raise vegetable consumption in preschool children found repeated taste exposure to be the most consistently effective single component, with reward and modeling adding smaller, less reliable gains (Nekitsing, Blundell-Birtill, Cockroft, & Hetherington, 2018). A parallel line sharpens the developmental timing, asking whether there are sensitive periods—weaning in particular—when exposure to variety and to vegetables is unusually effective, and whether early sensory experience durably biases later food choice (De Cosmi et al., 2017; Nicklaus & Schwartz, 2019).

A third direction is methodological: systematic reviews are replacing single studies as the unit of evidence, pooling the many small repeated-exposure experiments to estimate how many exposures are actually needed and for whom (Spill et al., 2019). Open questions remain about how robustly laboratory exposure effects transfer to the home, how large and durable gene–perception effects such as TAS2R38 are on real-world diets, and whether the early-exposure window can be exploited without coercion. The through-line is a shift from establishing that preferences are learned to quantifying how they are learned well enough to intervene.

Cognitive Implications

Food preferences are a compact case study in how an acquired value is built. They show a cognitive system taking a sparse set of innate signals, filtering them through a variable perceptual front end, and using associative learning to construct a detailed, durable ordering that then guides choice. The same ingredients—an unlearned prior, individual perceptual variation, and reinforcement-driven updating—recur wherever preferences are formed, which is why the study of food liking informs, and borrows from, the broader psychology of value and decision. It also connects directly to food fussiness, the selective-rejection pattern that is, in these terms, a preference ordering that has stayed narrow because the exposure and association that normally broaden it have not occurred. Understanding how preferences are learned is thus the general case of which overcoming picky eating is a particular, practically urgent instance.

Glossary

Conditioned taste aversion.
A strong, durable dislike for a flavor acquired in a single pairing with later nausea; a biologically prepared, extinction-resistant form of learning that is the aversive mirror of appetitive flavor–nutrient conditioning.
Flavor–flavor learning.
An increase in liking for a flavor that has been paired with an already-liked taste, such as sweetness; a form of associative conditioning that builds preferences without any nutritional consequence.
Flavor–nutrient learning.
A conditioned preference for a flavor reliably followed by the post-ingestive consequences of calories; the mechanism by which energy-dense foods come to be liked for their flavor.
Flavor.
The unified percept of a food combining taste, smell, and mouthfeel; the level at which most food preferences are actually formed and expressed, rather than at single basic tastes.
Food preference.
A learned or unlearned hedonic ordering of foods on a like–dislike scale; the selection of one food in preference to others, and among free choices a strong predictor of intake.
Hedonic response.
The pleasure or displeasure a taste or food evokes, observable in newborns as organized facial expressions of acceptance or rejection; the affective signal on which preferences are built.
Innate taste bias.
An unlearned hedonic response to a basic taste present at birth—acceptance of sweet and umami, rejection of bitter and sour—adaptive as a first approximation to nutrient and toxin.
Learned safety.
The process by which repeated ingestion of a food without ill effect teaches that it is safe, raising acceptance; the complement to the one-trial aversions a bad experience can produce.
Mere exposure effect.
The increase in liking for a stimulus that follows repeated, unreinforced encounters with it; applied to flavor, the mechanism by which repeated tasting raises preference.
Omnivore's dilemma.
The conflict faced by a species that can eat many foods: each unfamiliar item is both a possible nutrient and a possible toxin, making both caution and a way to learn safety adaptive.
Prenatal flavor learning.
Acquisition of flavor familiarity before birth, as flavors from the maternal diet reach the fetus through amniotic fluid; raises later acceptance of those flavors in infancy.
PROP / PTC taster status.
A person's sensitivity to the bitter compounds propylthiouracil and phenylthiocarbamide, dividing the population into tasters and non-tasters; largely set by the TAS2R38 genotype.
Repeated exposure.
The practice of offering a food many times without pressure; the best-supported way to raise acceptance of an initially disliked food, typically needing eight to fifteen tastings.
Sensory-specific satiety.
The decline in a food's momentary pleasantness as it is eaten to fullness, which shifts preference toward other, not-yet-eaten foods and promotes dietary variety within a meal.
TAS2R38.
A gene encoding a bitter taste receptor; common variants define sensitive and insensitive haplotypes and account for most individual variation in the perceived bitterness of PTC/PROP and related compounds.
Umami.
The savory basic taste, signalling the presence of glutamate and hence protein; accepted at moderate concentrations from birth and one of the innately preferred tastes.

Key Researchers

Gary K. Beauchamp

American experimental psychologist and emeritus director of the Monell Chemical Senses Center; his work on the chemical senses and the early development of taste established how innate responses and early experience together shape flavor preference.
Wikipedia · Google Scholar · Faculty page

Leann L. Birch

American developmental psychologist (1946–2019) at Pennsylvania State University and later the University of Georgia; her experiments on repeated exposure and flavor–nutrient learning established the modern account of how children's food preferences are acquired.
Wikipedia · Wikidata · Google Scholar

Adam Drewnowski

Nutrition scientist at the University of Washington, where he directs the Center for Public Health Nutrition; his research links taste preferences, energy density, and the determinants of food intake across the lifespan.
Wikipedia · Google Scholar · Faculty page

Julie A. Mennella

Developmental psychobiologist at the Monell Chemical Senses Center; her studies of prenatal and postnatal flavor learning and the ontogeny of taste preferences demonstrated that flavor experience begins before birth and in breast milk.
Wikipedia · Faculty page

Paul Rozin

American psychologist (b. 1936) at the University of Pennsylvania; his analyses of disgust and of the acquisition of food likes and dislikes framed the psychology of what people will and will not eat.
Wikipedia · Google Scholar · Faculty page

Frequently Asked Questions

What is a food preference?

It is a hedonic ordering of foods—the ranked set of likes and dislikes that decides, among the foods available, which a person chooses to eat. The National Library of Medicine indexes it as the selection of one food in preference to others, and across free choices measured liking strongly predicts actual intake (Drewnowski, 1997).

Are any food preferences innate?

A few taste responses are. From birth, infants accept sweet and savory (umami) tastes and reject bitter and sour ones, biases that roughly track nutrients and toxins. These are responses to basic tastes, not to whole foods, and they are the starting point that experience then reshapes (Mennella, 2014).

How are most food preferences formed?

By learning. Repeated exposure to a food raises liking, and pairing a flavor with calories (flavor–nutrient learning) or with an already-liked taste (flavor–flavor learning) conditions a durable preference for it (Birch, 1999; Zajonc, 1968).

How many times does a child need to try a food to like it?

More than most caregivers expect. Systematic reviews of repeated-exposure studies in infants and toddlers find acceptance of an initially rejected food usually rises only after roughly eight to fifteen tastings (Spill et al., 2019; Birch & Marlin, 1982).

Can food preferences form before birth?

Yes. Flavors from the mother's diet pass into amniotic fluid and breast milk, and infants exposed to a flavor by these routes accept it more readily in solid food, so flavor learning begins in the womb (Mennella, Jagnow, & Beauchamp, 2001; Ventura & Worobey, 2013).

Why do some people hate vegetables like broccoli?

Partly genetics. Variation in the bitter-receptor gene TAS2R38 makes some people perceive the bitter compounds in brassica vegetables as intensely bitter and others as nearly tasteless, so the same vegetable is genuinely more aversive to some children (Kim et al., 2003; Mennella, 2014).

Does rewarding a child for eating a food make them like it?

Generally no. Pressure and contingent reward tend to lower long-run liking by reframing the food as a cost; letting the food's own flavor and nutritional consequences do the conditioning works better (Scaglioni et al., 2018).

What is the best way to broaden a child's food preferences?

Early, repeated, unpressured exposure to a variety of flavors—beginning in infancy and at weaning—is the most consistently effective strategy, with repeated taste exposure the single strongest component in intervention trials (Nekitsing et al., 2018; Nicklaus & Schwartz, 2019).

References

Birch, L. L. (1999). Development of food preferences. Annual Review of Nutrition, 19, 41-62. https://doi.org/10.1146/annurev.nutr.19.1.41

Birch, L. L., & Marlin, D. W. (1982). I don't like it; I never tried it: Effects of exposure on two-year-old children's food preferences. Appetite, 3(4), 353-360. https://doi.org/10.1016/S0195-6663(82)80053-6

Cooke, L. J., & Wardle, J. (2005). Age and gender differences in children's food preferences. British Journal of Nutrition, 93(5), 741-746. https://doi.org/10.1079/BJN20051389

De Cosmi, V., Scaglioni, S., & Agostoni, C. (2017). Early taste experiences and later food choices. Nutrients, 9(2), 107. https://doi.org/10.3390/nu9020107

Drewnowski, A. (1997). Taste preferences and food intake. Annual Review of Nutrition, 17, 237-253. https://doi.org/10.1146/annurev.nutr.17.1.237

Garcia, J., & Koelling, R. A. (1966). Relation of cue to consequence in avoidance learning. Psychonomic Science, 4(1), 123-124. https://doi.org/10.3758/BF03342209

Kim, U.-K., Jorgenson, E., Coon, H., Leppert, M., Risch, N., & Drayna, D. (2003). Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science, 299(5610), 1221-1225. https://doi.org/10.1126/science.1080190

Liem, D. G., & Mennella, J. A. (2002). Heightened sour preferences during childhood. Developmental Psychobiology, 41(4), 388-395. https://doi.org/10.1002/dev.10067

Mennella, J. A., Jagnow, C. P., & Beauchamp, G. K. (2001). Prenatal and postnatal flavor learning by human infants. Pediatrics, 107(6), e88. https://doi.org/10.1542/peds.107.6.e88

Mennella, J. A. (2014). Ontogeny of taste preferences: Basic biology and implications for health. American Journal of Clinical Nutrition, 99(3), 704S-711S. https://doi.org/10.3945/ajcn.113.067694

Nekitsing, C., Blundell-Birtill, P., Cockroft, J. E., & Hetherington, M. M. (2018). Systematic review and meta-analysis of strategies to increase vegetable consumption in preschool children aged 2-5 years. Current Obesity Reports, 7(1), 60-67. https://doi.org/10.1007/s13679-018-0297-8

Nicklaus, S., & Schwartz, C. (2019). Early influencing factors on the development of sensory and food preferences. Current Opinion in Clinical Nutrition and Metabolic Care, 22(3), 230-235. https://doi.org/10.1097/MCO.0000000000000554

Rozin, P., & Fallon, A. E. (1987). A perspective on disgust. Psychological Review, 94(1), 23-41. https://doi.org/10.1037/0033-295X.94.1.23

Scaglioni, S., De Cosmi, V., Ciappolino, V., Parazzini, F., Brambilla, P., & Agostoni, C. (2018). Factors influencing children's eating behaviours. Nutrients, 10(6), 706. https://doi.org/10.3390/nu10060706

Spill, M. K., Johns, K., Callahan, E. H., Shapiro, M. J., Wong, Y. P., Benjamin-Neelon, S. E., Birch, L., Black, M. M., Cook, J. T., Faith, M. S., Mennella, J. A., & Casavale, K. O. (2019). Repeated exposure to food and food acceptability in infants and toddlers: A systematic review. American Journal of Clinical Nutrition, 109(Suppl 7), 978S-989S. https://doi.org/10.1093/ajcn/nqy308

Ventura, A. K., & Worobey, J. (2013). Early influences on the development of food preferences. Current Biology, 23(9), R401-R408. https://doi.org/10.1016/j.cub.2013.02.037

Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9(2, Pt. 2), 1-27. https://doi.org/10.1037/h0025848