ACCEPTANCE AND REJECTION

ACCEPTANCE AND REJECTION. Foods vary along a hedonic dimension, that is, in their ability to evoke pleasure. A food's hedonic value can differ significantly between individuals and among cultures. In developed countries at least, pleasure is probably the strongest determinant of diet. For most of us, most of the time, a global emotional response to the taste of a food determines whether it is consumed. Underlying this seemingly simple decision is a remarkable range of emotions—from blissful appreciation of haute cuisine to a profound rejection elicited by feelings of disgust. As with many other complex human behaviors, the development of food likes and dislikes reflects the operation of multiple influences—genetic inheritance, maternal diet, child raising practices, learning, cognition, and culture. In fact, the development of food preferences may be an ideal model of the interplay of these influences during our life span.

Foods may be selected or rejected for a variety of reasons, including their anticipated effects on health, their perceived ethical or environmental appropriateness, or practical considerations as price, availability, and convenience. However, it is our responses to the sensory properties of a food—its odor, taste, flavor, and texture—that provide the underlying basis of food acceptance. This article will focus on some of the influences that shape hedonic responses to foods, their flavors, and other sensory qualities.

Tastes

Despite evidence of innate hedonic responses to basic tastes, the vast majority of specific food likes and dislikes are not predetermined—no one is born liking blue cheese, for example. This is not to suggest that basic sensory qualities are unimportant. On the contrary, relatively fixed hedonic responses to sweet, salty, bitter, and umami (glutamate taste) tastes, and almost certainly fat, are present at or shortly after birth, and continue to exert an influence on food preferences. The strong affinity that children show for very sweet foods, and the persistence of the early development of liking for the taste of salt and salty foods throughout life appear to be universal. A majority in many Western societies also choose a diet that is high in fat.

However, innate responses do not account for the broad range of food likes and dislikes that develop beyond infancy. For instance, humans and many other mammals can detect bitterness at low levels and find it unpalatable because it is a potential sign of toxicity. Yet, while coffee and beer are typically rejected on first tasting, they are ultimately the strongest contenders for being the global beverages. The pungency of spicy foods is also initially rejected. Worldwide, though, chili is second only to salt as a food spice. Thus, although innate influences are clearly important in food selection, these are modified by our experience with foods (although both physiological makeup and culture will partly determine the extent to which experience is allowed to operate). What is more important than our innate preferences is the fact that we are predisposed to learn to like (and sometimes, dislike) foods. Some other preferences do appear to be common across cultures whose diets are very different. However, examples such as the widespread liking for vanilla and chocolate flavor are likely to reflect some degree of common experience.

Texture

Texture is a crucial criterion for sensory acceptance and rejection. Certain textures do seem to be universally liked, crispness, for example—perhaps through its association with freshness. Of course, to some extent, we will always prefer textures that are compatible with our dentition, and thus we would not expect infants to like hard foods. Foods that are difficult to manipulate in the mouth—such as soggy foods—are commonly disliked, as are foods that require excessive saliva and effort to swallow, such as dry, tough meat. While food texture is often cited as a reason for rejecting food, for example raw oysters, it is likely that such preferences are also a function of our prior expectations for specific foods.

Color

Food color is also undoubtedly a strong influence on acceptability, but again this is likely to reflect prior expectations. Whether we prefer white (U.S.) or yellow (U.K.) butter depends on what we have eaten in the past. Some colors have been thought to be inappropriate for food. The color blue, for instance, has been suggested as a candidate for a universally inappropriate food color—after all, very few foods are naturally blue. But recent marketing of brightly and "inappropriately" colored foods for children tends to undermine this notion, since the children appear receptive to unusual colors. Removing color from common foods does reliably reduce liking for those foods, perhaps by undermining our ability to identify their flavor, thus making them seem less familiar.

Fear of the New

The fact that humans are omnivores creates a paradox. On the one hand, we have access to a large range of potential nutrients; conversely (in nature at least), we are much more likely to be exposed to toxic substances. In the first two to three years of our lives, we exist in a highly protected environment, first in the context of breast or bottle feeding, and then through parental food selection and supervision. It is therefore adaptive for young infants to accept a wide variety of foods as the risk of exposure to potentially toxic nonfoods is low.

In later infancy, greater independence is typical, both in terms of the wider variety of other people encountered and also of the potential to come into contact with edible substances, which may be unsuitable for health or other reasons, outside direct parental influence. At this point, food neophobia often becomes apparent. Reluctance to consume novel foods at this age is most obviously reflected in statements of "I don't like it" to foods that have never been tried. The rejection of unfamiliar foods can now be seen as adaptive, given the wider risk of ingestion of potentially toxic substances. Food neophobia is found not just in humans, but also in a variety of non-human species, including rats, dogs, birds, and fish. Hence, it may be a universal safeguard against potential toxics.

The trait of food neophobia has been investigated in different age groups, as has the nature of the "fear" and how it can be modified. Even in adults, there often remain strong vestiges of childhood neophobia. While many welcome the chance to sample exotic foods or novel flavors, others remain unable to even consider consumption of foods beyond their usual repertoire.

Such reluctance is especially strong for foods of animal origin (unfamiliar meats, dairy products, or eggs), the same foods that elicit reactions of disgust, also thought to be a protective mechanism. Why this food-related personality trait varies so much among adults is unclear, but it might reflect the breadth of experience with different foods in childhood.

Interestingly, in both children and adults, food neophobia appears to be mediated less by any conscious awareness of the potential for danger, than by the much more immediate fear that foods will taste unpleasant. Consistent with this, willingness to try a novel food can be increased by strategies that reduce this anxiety, including providing information about the food's flavor or indicating that others have enjoyed it since. Highly neophobic individuals are more likely to choose an unfamiliar food after they have seen others select it. Specific nutritional information (such as the fact that a food is low in fat) also encourages selection of novel foods, but only for those for whom nutrition is important. In each case, the net effect is to assure the taster that the food is acceptable in terms of flavor and perhaps safety. Neophobia is a major issue for many parents concerned about the narrow range of foods that their children are willing to consume. A common strategy is to use one food as a reward for eating another food—one that the adult wants the child to eat. Unfortunately, these attempts frequently fail because the relative value of the foods is quite apparent. Rewarding the consumption of spinach by giving ice cream presents a message simple enough for any young child: ice cream is a good food (at least in terms of taste), otherwise why use it as a reward; spinach is bad, else why do I need to be rewarded for eating it? The unfortunate, if predictable, consequences of such strategies are increased liking for the reward and a decrease in liking for the target food.

Learning to Like

What does reduce neophobia and encourage consumption? In both children and adults, repeated exposure has been found to lead to increased acceptability of novel foods, with greater exposure producing greater liking. For example, three-and four-year-old children have been found to accept initially rejected cheese and fruits following ten exposures. It is possible that individuals who receive repeated exposure to a wide variety of foods as infants and children are least likely to be highly neophobic as adults, although this has yet to be established. That is, the more we experience different foods, the more we are willing to experience different foods.

Exposure appears to be the one mechanism that is necessary for liking to increase. With novel foods or flavors, repeated consumption might lead to increased liking via a reduction in neophobia—effectively a relief from the anxiety associated with novelty. It certainly produces an increase in familiarity, an important aspect of children's likes and dislikes, and it has been recognized for some time that sensations of recognition are in themselves positive. However, changes in liking for food ingredients or ingredient levels in already familiar foods strongly suggest that exposure per se produces liking, and that a food or flavor does not need to be completely novel. There are many commonplace examples of this, including the gradual increase in liking that accompanies changing from regular to low-fat milk or low-salt soup, or reducing sugar in tea or coffee.

Although it is a necessary precondition, by itself, exposure is insufficient to explain why we end up liking some foods more than others. There appears to be a variety of other processes that operate during repeated food experiences, producing preferences for the diverse range of food odors and flavors that we encounter. Whether sniffed as aromas, or as characteristic flavor qualities in the mouth, food odors reliably inform us whether we have previously experienced a food. Odors are thus most likely to be the source of neophobic responses. However, there is nothing intrinsic to the odor or flavor of any food that means we will develop a strong like or dislike for it. During our early infancy (up to about three years old), we appear to be neutral to most if not all odors, except for those that also produce nasal irritation, such as ammonia. In contrast to those for tastes, odor preferences are almost certainly all learned, and rely upon our ability to form associations with other liked qualities. Pairing a novel flavor with a sweet taste, for example, reliably increases subsequent liking for that flavor, even when the sweetness is not present. This process, known as classical conditioning or associative learning, was first described scientifically by Ivan Pavlov. He famously demonstrated that the sound of a bell, previously associated with the presentation of food, would elicit gastric secretions in his dogs. While the principles of Pavlovian conditioning were developed using animal (especially rat) models, they appear equally applicable to explaining aspects of human food likes and dislikes.

The universal high palatability of sweetness and fat is a reflection of the ability of substances associated with these qualities to provide energy to the body. Our bodies find the provision of energy inherently rewarding. Consequently, repeatedly pairing flavors with ingested carbohydrates or fats produces increases in liking for associated flavors. Other postingestional consequences have also been described, including enhanced liking for flavors paired with the alerting effects of caffeine—a plausible mechanism, together with the energy provided by the sugar and milk fat sometimes added, for the enormous popularity of coffee.

The effects of conditioning by positive association and the absorption of energy-rich foods are broad enough mechanisms to account for very many food likes. One implication of this process and the body's response to energy is that we end up showing a liking for foods that are high in sugar and fat. Clearly, this has implications for health. We may know that high-fat foods present us with a risk in the long term, but what drives our behavior primarily is the fact that we like the fat—it gives the food a pleasant mouthfeel, it carries flavor well, and its provides the body with energy. The body's response is to promote liking for flavor associated with the fat. Eventually, it is not just the fat or sugar content that we find palatable, but the specific flavor of the food as well.

Food dislikes may also result from Pavlovian conditioning. Associating a characteristic food flavor with nausea, as sometimes occurs with food poisoning or a coincidental illness, will promote a rapid, often irreversible, "taste" aversion that actually seems to make the flavor become unpleasant. The development of aversions can be seen as highly adaptive—it makes sense to avoid foods previously associated with gastric illness. Consequently, the conditioned association tends to be very strong. In humans, taste aversions are typically both long lasting and robust enough to persist even if it is known that the food was not the source of the illness. As with neophobic responses, meat seems to be a common target when aversions do occur. An unfortunate consequence of the nausea associated with cancer chemotherapy is the development of taste aversions. Close to three-quarters of children aged two to fifteen years old undergoing treatment are reported to have at least one aversion. Taste aversions are not common enough to account for the majority of our food dislikes, since they appear to occur in only about 30 percent of people. However, they are a powerful indicator of the role that consequences of food ingestion can play in shaping our responses to a food's sensory qualities.

Odors are not the only sensory qualities in foods for which preferences are shaped by learning. Our most primitive sense is the detection of pain—unsurprisingly, since pain avoidance is the simplest key to survival. How then to explain the fact that at least a quarter of the world's population each day consume (and presumably enjoy) a meal containing an extremely potent irritant, capsaicin, which is present in chilies? Whatever the source of our increasing preference for pungency in foods, it must be a potent mechanism. Apart from the warning signals for pain, our bodies possess a built-in response to high levels of irritation. This defensive reflex, as it is known, consists of increased blood flow to the head, profuse sweating, tearing, and nasal discharge—physiological changes that are thought to have evolved as a means of rapidly eliminating toxins. Although frequent consumers of spicy foods experience somewhat less intense physiological responses and burn than infrequent users, there is no doubt that the burning sensations are actually part of the reason these foods are consumed, not something to be tolerated for other reasons.

Both regular exposure, commencing during breast-feeding, and postingestional energy conditioning are likely to play a part in the development of liking for hot foods, particularly in countries whose staple diet includes high levels of spiciness. To explain the recent increase in liking for hot foods in Western countries, though, a number of other interesting mechanisms have also been proposed. These include the hypothesis that the painful experience may activate the brain's natural opioid (morphinelike) biochemical systems, dampening pain and producing a chili eater's "high." Alternatively, it has been suggested that we derive pleasure from the "thrill" of the benign but highly stimulating experience of consuming hot foods.

Where Do Differences in Food Likes Come From?

If exposure, together with resultant learning processes, can substantially explain food preference development, what accounts for the differences in which foods we come to like? Exposure to flavors is now known to begin even prior to birth. Amniotic fluid, which comes into contact with the taste and odor receptors in the mouth and nose of the fetus, carries both taste and odor qualities. There is good evidence that the maternal diet during pregnancy can influence food preferences of the child following birth. Thus, it has been shown that infants whose mothers consumed carrot juice during pregnancy showed a greater liking for carrot-flavored cereal at six months of age than did a control group of children whose mothers consumed only water. Following birth, a wide range of flavors derived from the maternal diet is carried in breast milk, and this also influences an infant's later food preferences, including greater acceptance of novel flavors. In other words, the variety of a mother's diet can promote a varied set of food preferences in the infant. As a result, breast-fed babies are more likely to develop preferences following exposure to novel foods as infants. Whether this reflects early exposure to particular flavors, or a general effect of previous maternal dietary variety, is uncertain.

Social Influences

From childhood on, social interactions, whether within the family or with other groups, provide the context within which the majority of food experiences occur, and hence by which learning of food likes is facilitated. The pleasure associated with such interactions—the conviviality of a meal shared with friends, for example—may represent just as positive a conditioning stimulus for a new food flavor as sweetness. Thus, it may be that our estimation of the food at a restaurant has as much to do with the social environment as it does with the chef's skills. In children, pairing foods with the presence of friends, a liked celebrity, or attention by adults all increase liking for those foods, no doubt reflecting the positive hedonic value of each of these groups to the child.

This process is strongly evident in the relative impact of different social interactions on the food preferences of children. Surprisingly, despite the enormous opportunities in a family for exposing children to the foods eaten by the parents, parental preferences are poor predictors of child food preferences; in fact, they are no better predictors than the preferences of other adults. This suggests that the extent to which these sets of preferences are related has more to do with the wider culture than with any specific food habits within the family. A child's food likes and dislikes are much more likely to be associated with those of peers, especially specific friends, than those of its parents. Peers may also be as effective as families at helping to overcome neophobia, since the food choices of both friends or well-known adults strongly influence a child's food choices. The ultimate impact of social facilitation of food choice is that the liking eventually becomes internalized. That is, foods chosen because others do so become liked for their own sensory properties.

The Cultural Context

Dietary differences between cultures are almost always more pronounced than individual differences within a culture. The relatively limited amount of research that has been conducted on cross-cultural perceptions of sensory qualities finds fewer differences than are needed to explain the often markedly different preferences for foods. More plausibly, it is likely that differences in preferences reflect experiences with different foods. In addition to facilitating liking through exposure and the action of social influences, cultures act to define what substances are considered foods.

Foods that are unfamiliar to a culture may initially be seen as entirely unsuitable for consumption, while certain flavors may be regarded as inappropriate for specific foods. For example, bean paste is often used as a sweet filling in Japanese cakes, whereas in many Western countries, beans are expected to inhabit savory, not sweet, products. Again, porridge is either sweet or savory, depending on your heritage. In other cases, because of different histories of exposure, a preferred flavor in one culture may be perceived as unpleasant in another. The odor and flavor of lamb and mutton are highly liked in the West but rejected in the many parts of Asia that do not have the history of consuming sheep meat. Foods may of course be the subject of religious or cultural taboos, or even not be defined as food at all. In Western countries, we are unlikely to ever develop a taste for dog meat or snake blood.

The notion of culturally specific flavor principles has been proposed as a way of categorizing cultural differences in cuisines. Flavor principles are unique combinations of specific ingredients used in a variety of foods within a culture. This combination provides a characteristic flavor that foods within the culture share, and identifies them as originating from that culture. For example, a characteristic combination of ingredients in Japanese cooking is soy sauce, mirin (sweet rice wine) and dashi (a stock made from flakes of the bonito fish, which is high in umami taste). While Korea is geographically close to Japan, its flavor principle could not be less similar, with the intense flavors of garlic, chili, sesame, and soy dominating many dishes. Flavor principles not only define the national cuisine, they also perform a social role by acting as an expression of the individuality of the culture.

Flavor principles may help to provide a solution to the "omnivore's paradox" and the consequent neophobic response that novelty can elicit, thus limiting the foods available for consumption within a culture. A familiar flavor can provide a safe context for new foods, thus maximizing the breadth of the diet. On the individual level, recent findings suggest that a familiar sauce could increase the willingness of children to consume a novel food. A characteristic combination of flavorings may also provide variety and interest in diets dominated by bland staples such as corn or rice. Although a flavor principle might contain only a small set of characteristic seasonings, these can be combined in different ways. Moreover, what may appear to be a single ingredient or spice to an outsider may in fact have many subtle variations. Different chili varieties, for instance, vary considerably in the flavor and degree of heat that they impart to foods.

Increasingly, the food industry operates in a global setting. This is likely to mean that those foods that are purchased in your local supermarket are, or soon will be, also available on the other side of the world, perhaps within a culture whose cuisine is vastly different from your own. Whether this means that national flavor principles will ultimately be diluted or replaced is uncertain. Some evidence suggests they will not. Japanese urban populations have, for many years, enjoyed wide access to foods from other parts of the world, particularly Europe and the United States. Yet, while rice consumption has fallen and red meat and dairy food consumption has increased in recent years, there is little evidence that more traditional foods are disappearing. Moreover, Western food companies wishing to export to those cultures whose cuisines are substantially different are learning that incorporating aspects of the flavor principles of those cultures is essential for producing acceptable foods.

Food Choice: The Broader Context

Although a food's sensory properties may substantially determine what we like, they are only part of why we choose a particular food on a particular occasion. The determinants of our diet include factors that are both internal and external to the individual. Food choices are influenced by appetite, which in turn reflects when and what we last ate, and our overall state of physical and psychological health. In some extreme cases, these internal influences can render eating itself a pathological process, as in disorders such as anorexia and bulimia nervosa. Even in nonpathological circumstances, though, choosing a high-fat or -carbohydrate food may have more to do with our mood than anything else.

Liking is also heavily dependent on context. At its simplest level, cultural practices will determine whether or not we eat cooked meat or toast for breakfast. The extent to which either of these foods is acceptable will depend considerably on time of day. The same food can also vary in acceptability depending on where we experience it. Due to the influence of prior expectations, the same meal served in a restaurant is likely to be judged as more acceptable than if it is served in a student cafeteria.

Clearly, also, the reason why we first choose a food must be based on factors other than direct experience of, and therefore liking for, the sensory properties of the food. Food manufacturers and marketers rely on advertising and labeling to create a positive image for products, and attempt to create high (but not unrealistic) expectations for the product's sensory properties. If the food meets those expectations following purchase, then the consumer is likely to try the product again. Repeat consumption and the consequent associative and post-ingestive processes will then act to promote increased liking for the product.

See also Anorexia, Bulimia; Appetite; Aversion to Food; Disgust; Sensation and the Senses;Taboos.

BIBLIOGRAPHY

Bernstein, Ilene L. "Development of Taste Preferences." In The Hedonics of Taste, edited by Robert C. Bolles, pp. 143–157. Hillsdale, N.J.: Erlbaum, 1991.

Birch, Leann L., and D. W. Marlin. "I Don't Like It; I Never Tried It: Effects of Exposure on Two-Year-Old Children's Food Preferences." Appetite 3 (1982): 353–360.

Birch, Leann L., Jennifer O. Fisher, and Karen Grimm-Thomas. "The Development of Children's Eating Habits." In Food Choice, Acceptance and Consumption, edited by Herbert L. Meiselman and Hal J. H. MacFie, pp. 161–206. London: Blackie, 1996.

Cabanac, Michel. "Physiological Role of Pleasure." Science 173 (1971): 1103–1107.

Cardello, Armand V. "The Role of the Human Senses in Food Acceptance." In Food Choice, Acceptance and Consumption, edited by Herbert L. Meiselman and Hal J. H. MacFie, pp. 1–82. London: Blackie, 1996.

Davis, Clara M. "Self-Selection of Diet by Newly Weaned Infants." American Journal of Diseases of Children 36 (1928): 651–679.

Meiselman, Herbert L. "The Contextual Basis for Food Acceptance, Food Choice and Food Intake: The Food, The Situation and The Individual." In Food Choice, Acceptance and Consumption, edited by Herbert L. Meiselman and Hal J. H. MacFie, pp. 239–263. London: Blackie, 1996.

Mennella, Julie A., and Gary K. Beauchamp "The Ontogeny of Human Flavor Perception." In Handbook of Perception and Cognition: Tasting and Smelling, edited by Gary K. Beau-champ and Linda M. Bartoshuk. San Diego, Calif.: Academic Press, 1997.

Pliner, Patricia. "The Effects of Mere Exposure on Liking for Edible Substances." Appetite 3 (1982): 283–290.

Pliner, Patricia, Marcie Pelchat, and M. Grabski "Reduction of Neophobia in Humans by Exposure to Novel Foods." Appetite 20 (1993): 111–123.

Pliner, Patricia, and Marcia L. Pelchat. "Neophobia in Humans and The Special Status of Foods of Animal Origin." Appetite 16 (1991): 205–218.

Pliner, Patricia and Catherine Stallberg-White "'Pass the ketchup, please': Familiar Flavors Increase Children's Willingness to Taste Novel Foods." Appetite 34 (2000): 95–103.

Pliner, Patricia, Paul Rozin, Myra Cooper, and George Woody. "Role of Specific Postingestional Effects and Medicinal Context in the Acquisition of Liking for Tastes." Appetite 6 (1985): 243–252.

Prescott, John, and Graham A. Bell. "Cross-Cultural Determinants of Food Acceptability: Recent Research on Sensory Perceptions and Preferences." Trends in Food Science and Technology 6 (1995): 201–205.

Rolls, Barbara J. "Sensory-Specific Satiety." Nutrition Reviews 44 (1986): 93–101.

Rozin, Elisabeth. Ethnic Cuisine: The Flavor-Principle Cookbook. Brattleboro, Vt.: The Stephen Greene Press, 1983.

Rozin, Elisabeth, and Paul Rozin. "Culinary Themes and Variations." Natural History 90 (1981): 6–14.

Rozin, Paul, and April E. Fallon. "A Perspective on Disgust." Psychological Review 94 (1987): 23–41.

Rozin, Paul. "Human Food Selection: The Interaction of Biology, Culture and Individual Experience." In The Psychobiology of Human Food Selection, edited by L. M. Barker, pp. 225–254. Westport, Conn.: AVI Publishing, 1982.

Rozin, Paul, and Theresa A. Vollmecke. "Food Likes and Dislikes." Annual Review of Nutrition 6 (1986): 433–456.

Rozin, Paul, and Debra Zellner. "The Role of Pavlovian Conditioning in the Acquisition of Food Likes and Dislikes." Annals of the New York Academy of Sciences 443 (1985): 189–202.

Sullivan, Susan A., and Leanne L. Birch. "Pass the sugar, pass the salt: Experience Dictates Preference." Developmental Psychology 26 (1990): 546–551.

John Prescott

Cravings for Food

At some time, most of us have experienced a craving for a specific food—something that we must have now, and which we will go out of our way to obtain. It is almost as if the body is insisting that we must have that food. There is much anecdotal information about craving and physiological needs, but less hard evidence for such specific appetites. It is clear that we get hungry and thirsty, but does the body really crave particular nutrients?

The one incontrovertible specific hunger that humans possess is for sodium chloride, common salt. Salt is metabolically essential and most of the time this need is both met and exceeded through diet. Clinical studies have demonstrated that in cases where the body is depleted of salt, humans develop strong appetites for the taste of salt, and its normal degree of palatability is increased. The same is true in experiments in which volunteers are fed low-salt diets—salty foods increase in palatability. Hence, it appears that a change in the hedonic value of the taste of salt is the mediator for increased intake when depleted.

Beyond salt appetite, however, there is little strong evidence that other specific appetites exist. There are reports suggesting an association between pica (the consumption of earth) and mineral (especially iron) deficiency. This practice appears to be most prevalent among pregnant women in poor rural communities. Pregnancy is well known to be associated with craving for foods, but it is not clear whether such "normal" cravings are related to metabolic needs.

The single most commonly craved food in Western societies is chocolate. Although chocolate contains phamacologically active compounds, there is no evidence these compounds are what is craved. Instead, the craving for chocolate is related to craving sweet foods generally and to chocolate's palatability, based on an optimal combination of sugar and fat. Chocolate craving is more common among women, and hormonal influences have been suggested as being important. The craving shows a peak around the time of menstruation and is also more common during pregnancy. While chocolate and sweet food cravings do occur among males, cravings for savory foods are more common.

A less extreme version of craving is the phenomenon of "moreishness." Again, wanting "just one more bite" appears to reflect the high palatability of certain foods, rather than a desire for any specific nutrient. Foods described as moreish also tend to be consumed in small amounts. Often their consumption is subject to a voluntary restraint determined by social mores; you may want another slice of cake, another piece of chocolate, or another potato chip, but will often hold back to avoid seeming intemperate. Because of the typically small portion sizes associated with moreish food, this may be an example of the appetizer effect, which occurs when the initial consumption of palatable foods increases appetite for further eating.

Explanations for craving, moreishness, and appetizer effects have recently focused on the brain's biochemistry, in particular those functions mediated by opioid (morphinelike) peptides. Interfering with the functioning of this biochemical system using opioid blocking drugs leads to reduced food consumption overall and also to attenuation of appetizer effects, apparently because the foods become less palatable. Conversely, it is possible that increased opioid levels may induce cravings by making foods more palatable. Such changes may occur in a variety of circumstances—dieting, stress, exercise, alcohol consumption—all of which are known to influence the brain's opioid systems.

Cravings thus tell us little about the body's nutritional needs, beyond the fact that highly palatable foods tend to be high in energy. Other evidence also points to strategies to maximize energy intake. At least in Western countries, given ample availability, we tend to consume a diet that contains 35 to 40 percent fat, well in excess of what we need to survive. Moreover, from early infancy onwards, we will attempt to compensate for reductions in calories at one meal with an increase at the next.

In addition to energy intake, we seem predisposed, as omnivores, to seek variety in our diet. As noted in the section on sensory-specific satiety, this may be one way of optimizing survival through ensuring adequate nutrient intake. Classic studies on dietary self-selection were carried out by Clara Davis in the 1920s and 1930s. She allowed recently weaned infants access to a varied selection of foods and found that they first tasted widely and then developed preferences for a selection of these foods. This research has been often misinterpreted to suggest that the body has an innate wisdom, in that the foods the infants selected represented a balanced nutrient intake. This was inevitable, however, given the range of foods available.

This is not to say that mechanisms responsive to our needs are not in operation. On the contrary, the palatability of energy and sodium sources, the avoidance of toxins through dislike of bitterness, the rapid formation of aversions to foods associated with gastric illness, and the maintenance of nutrient variety via sensory-specific satiety, are all innate predispositions that modulate the hedonic value of sensory properties of foods to help ensure survival.