In moving from the sensory input to the action output within the human brain flavor system, it is natural to begin with emotions. The word emotions is derived from “to move.” Just as movements of the mouth and tongue make flavor an active sense, so is it also an active sense in that we must be motivated to acquire the food and liquid we put in our mouths. As indicated in the previous chapter, these systems have a close relation to the part of the brain called the amygdala, and their activity can be seen as beginning with the motivation for wanting a food or liquid, which may be learned as a liking for it and then become a craving for it.
What kind of brain activity represents our motivation to desire a flavor, our emotion that makes us prefer it and want to obtain it? And if we become too highly motivated, too desiring, how does this brain activity pass over to craving it? These questions have stimulated much research that it is hoped will help us understand normal liking for food and abnormal craving for it.
Images of Desire
In the human, investigators have used functional brain imaging to answer these questions. Among the first to do so were Marcia Pelchat and her collaborators at the Monell Chemical Senses Institute in Philadelphia in 2004. It will be useful to describe this pioneering work in detail as an example of how perceptions interact with emotions and how this interaction is studied in human subjects.
BOX 19.1
Brain Regions Activated by Pleasant Food Smells
The authors begin by pointing out that craving a favorite food is experienced by most people, particularly young people, and may play a significant role in excessive snacking, eating disorders, and obesity. There is no strong evidence that cravings are for nutritional types of food, but there is evidence that a dull, boring diet stimulates strong cravings for more flavorful foods.
The authors noted that studies up to then had shown that food flavors activate certain brain areas. These areas were largely consistent with our discussion in the previous chapter and are summarized in box 19.1. These previous studies used hunger to stimulate the desire for the food. Hunger affects the whole body in a way that may obscure pure cravings for a desired food. The authors therefore decided instead to use a monotonous diet as a baseline for judging the activation of brain regions caused by craving.
At the time no one had done this kind of study. One hypothesis was that the areas activated by craved foods might be the same ones activated by pleasant foods. However, there are many foods we like without craving them. In addition, the authors were well aware that craving is also a characteristic of drug addiction. A great many people had been interested in the brain mechanisms underlying drug addiction and had worked out experimental procedures for bringing out drug cravings in patients, which could be studied using brain imaging. These activities had produced evidence that cravings for drugs, including alcohol, activated specific areas in the brain, which are summarized in box 19.2.
The overlap of the first four items in each table was intriguing. So the authors decided to focus on possible similarities in brain circuits when subjects were stimulated by craved foods and by craved drugs: “We hypothesize that individuals receiving a monotonous diet will have greater food cravings and related activation in these candidate regions than individuals maintained on their normal diet.” The hypothesis was bold in suggesting that a strong desire for particular foods we like—an exaggeration of otherwise normal feeding habits—might activate the same brain circuits as an abnormal addiction to substances of abuse, including social drugs such as tobacco and alcohol as well as illegal drugs such as cocaine. It was the kind of hypothesis that drives science, because it enabled the investigators to structure their study in an effective manner that not only was well focused on the subject at hand, but also allowed direct comparisons with the other studies.
BOX 19.2
Brain Regions Activated by Drugs of Abuse
The subjects were healthy college volunteers, some of whom were maintained for a couple of days on a normal diet and some on a monotonous diet. The monotonous diet consisted of a vanilla-flavored drink containing 240 kcalories plus protein and vitamins. The subjects consumed, on average, nine 8-ounce (226-gram) cans of the stuff a day, for a total of around 2,200 calories per day.
At the end of the second day, the subjects were tested in the brain scanner. They were first tested at rest. Then names of two foods they had selected that they really liked were presented to them visually, and they were asked to imagine the smell, taste, and texture of their favorite dish of the food while the brain scanner recorded their activity. Presenting the names instead of the pictures of the food avoided showing non-optimal versions of the food to different subjects; with the names, the subjects could themselves imagine their favorite version. The functional images of what is called the BOLD signal were obtained in a strong magnet (rated at 4 Tesla) that enabled activity in small brain regions to be seen.
BOX 19.3
Brain Regions Activated by Craving a Food
The subjects all reported that they could easily imagine their favorite foods. The subjects on the monotonous diet in addition reported that they felt a craving for their liked food while imagining it, whereas craving was only reported by a portion of the subjects on a normal diet. No one reported a craving for the monotonous diet.
The brain scans were unequivocal. The subjects on the monotonous diet showed activation of specific brain regions while imagining their favorite foods, whereas those on a normal diet did not. The activated regions are shown in box 19.3. This result was significant, because all these regions fall within the group activated by drug cravings. Only the insula is shared with the pleasant food smells.
The authors thus consider that the results support their hypothesis of “a common circuitry for desire for natural and pathological rewards.”
The hippocampus, the insula, and the caudate nucleus are three areas that also merit further discussion in the interest of neurogastronomy.
The hippocampus has been shown to be involved in cocaine addiction, possibly by reinvoking memory mechanisms that drive this behavior. It may similarly provide the memory traces that drive food craving. In psychological terms, the memory functions as the reinforcer for the learned craving. When the subject sees the drug, the incoming visual image is reinforced by the memory trace in the brain. The sites where they meet would be equivalent to the buffer of a mental image, as discussed in the previous chapter.
We have already met the insula as a site of convergence of taste and smell inputs; it has also been shown to be involved in taste memories and in emotional behavior. Like the hippocampus, it may contain memories that act as reinforcers.
The caudate nucleus is a part of the system known as the striatum and plays multiple critical roles in sensorimotor coordination in the brain. It contains a high concentration of dopamine, the neurotransmitter released by fibers from the so-called substantia nigra (a region in the brain stem that appeared black to the early histologists who discovered it). This system is vulnerable to a range of disorders. Degeneration of the dopamine-containing cells in the substantia nigra and the terminals of their fibers in the striatum causes Parkinson’s disease, which is associated with tremor of the hands, slowing of movement, and a range of other disorders, including (surprisingly) a decline in the sense of smell. Disorders of dopamine are also believed to be involved in schizophrenia.
Among the normal functions of the striatum are its involvement in the formation of motor habits, which is of interest; food cravings can be regarded as habits that are hard to break. Dopamine is also involved in the reward system of the orbitofrontal cortex, which we will discuss in chapters 21 and 22. With regard to drug cravings, it has been suggested by Ingmar Franken, Jan Booij, and Wim van den Brink in the Netherlands that “in conditioned subjects dopamine has a role [in] the earlier, motivational phase, i.e., before the use of the drug and before the experience of pleasure per se. This motivational phase can be labeled as the desire phase of drug use: [in other words] craving.” This comment emphasizes the complex relations between the brain systems that produce the brain states we call motivation, leading to desire and craving, as well as to pleasure.
Finally, by subtracting the brain foci due to monotonous food cues from foci of liked food cues, the investigators were able to identify more regions that were activated when the subjects only thought about liking a food as well as craving it, as summarized in box 19.4.
Among these are the following that should be of interest for understanding the brain mechanisms for creating not only the perception of a food or flavor but the motivational and emotional states that make us like it.
The left fusiform gyrus on the bottom of the temporal lobe is interesting, because many studies have shown that activity in this area reflects our perception of an emotion we are expressing. It works together with the nearby amygdala and parahippocampal gyrus in the following way. When hungry subjects view pictures of food, functional brain scans show activity specifically in these areas, as reported by Marsel Mesulam and his colleagues at the Northwestern Medical School in Chicago in 2001. The authors interpret their results as follows: “These results support the hypothesis that the amygdala and associated inferotemporal regions [the lower parts of the temporal lobe] are involved in the integration of subjective interoceptive [inside the body] states with relevant sensory cues processed along the ventral visual stream [the parts of the visual system that are involved in identifying objects].”
BOX 19.4
Brain Regions Activated by Liking a Food
In plain language, this means that when we are hungry and view a picture of a food that we like, the picture sets up activity in the visual pathway that reaches cells in these regions to produce our personal internal “food image” of that food. This “interoceptive” image produces an emotion of liking those foods, as well as a motivation to acquire and consume them. It is an example of a mental image that is distributed among different regions and different modalities, a “multiregional multimodal image” that represents emotional and motivational states rather than the perception of what is pictured. The implication for the neurogastronome is that when you sit down to enjoy your meal, the hungrier you are the more active are your internal “emotional flavor images” of the food flavors you are perceiving. Pelchat and her colleagues call these images of desire. They are to the flavor action system what the flavor images are to the flavor sensory systems.
The amygdala is usually more involved than indicated here. As we have seen, it is sensitive to the intensity of taste stimuli, among other things. It also has an essential role in a wide range of emotional behavior in all mammals.
The cingulate gyrus is also of special interest to neurogastronomy. It is a ring of cortex just above the fibers of the corpus callosum that interconnect the two hemispheres. Several brain-scanning studies have shown that this is the brain region most consistently activated by stimuli that have a strong emotional quality and that figure in both food liking and food and drug craving. We have seen that it is part of several of the frontal lobe systems involved in those states.
Finally, the authors note that activity in the orbitofrontal cortex does not show up in their results. This may be due to the difficulty of imaging the BOLD signal in this part of the cortex, which is so close to the underlying bony braincase. But they note it may also be due to the fact that orbitofrontal activity may be involved in thinking about food in general, whether it is liked or disliked. This only reminds us that a scientific study combines biology, experimental methods, and interpretation, leaving unexplained results that require further study.
The overlap between cravings for food and drugs of addiction has become a major theme in modern studies, as is discussed in detail in chapter 22.
Chocolate: From Craving to Disgust!
A second perspective on brain mechanisms in food craving comes from a study by Dana Small and her colleagues in 2001 on craving for chocolate. They recruited nine subjects who described themselves as “chocolate lovers” and were high on a scale of “chocoholics.” Four and a half hours after breakfast, they began to be tested in a brain scanner with a square of chocolate. The diabolic strategy behind the study was not only to examine the brain areas activated by the chocolate at the start, but to continue the experiment with successive squares of chocolate until the subjects had been fed to satiety—until they could not stand to eat another square of chocolate.
The subjects varied, some of them quitting when they reached 16 squares (about a half bar of chocolate), whereas others lasted up to 74 squares (about two and a half bars). There was a five-minute rest period between each square. As the study progressed, from hunger to satiety, the subjects were asked with each square how it rated on a scale of “delicious, I really want another piece” to “awful, eating more would make me sick.”
BOX 19.5
Brain Regions Activated by Chocolate
The brain scans thus provided evidence regarding the brain areas that were active when subjects were hungry and highly motivated to eat a craved-for food and those that were active when the subjects forced themselves to eat the same food when it tasted “awful.” The results are summarized in box 19.5. At the beginning, the active brain areas were the subcallosal region [under the corpus callosum connecting the two hemispheres], the orbitofrontal cortex, the insula and operculum, the striatum, and the midbrain. We can call this the flavor image of chocolate when it is desired. At the end, the active areas were the orbitofrontal cortex, the parahippocampal gyrus, and the prefrontal regions. This can be called the flavor image of chocolate when it has been eaten to satiety. Neurogastronomes may use this as an indication of how the activity within our brains shifts with the consumption of even our favorite foods.
A subtle aspect of this research strategy was that what changed was the reward value of the chocolate. There is much current interest in brain science in the reward value of a stimulus, because it signals what an animal holds important, is motivated to work for, and will make decisions about acquiring or not acquiring in relation to alternatives. Wolfram Schultz in Switzerland introduced this notion with his studies of how the dopamine system in the brain is activated when a monkey is making such decisions about reward value. The chocolate study was an example of making decisions about reward value in relation to hunger or satiety. As discussed in chapter 12, reward value is one of the key functions of the orbitofrontal cortex, the neocortical end station of the olfactory pathway. This will be discussed further when we consider how the brain makes decisions about flavor (chapter 22).