The obesity epidemic has led to increased scientific interest in how the brain controls human feeding behavior. Why do we get hungry? What biological mechanisms tell us what to eat and when to stop eating?

It’s long been assumed that two neurobiological mechanisms largely govern food intake: one that controls the need to eat and one that controls the desire to eat. The hypothalamus in the brain regulates the homeostatic control of food intake by receiving, coordinating and responding to metabolic cues and signals from the digestive system. By integrating these metabolic signals, the hypothalamus tells us when we need to eat to maintain a body weight “set point,” much like a thermostat set on a specific temperature. It is clear, however, that higher brain centers that control the desire to eat also substantially influence our food consumption. The dopamine reward system is one such brain center. (When you covet a bowl of chocolate ice cream after dinner, a food that you don’t need to eat for hunger but want to eat, it is your dopamine reward system that gets excited.) In many situations, this desire to eat can override the need to eat, leading people to consume tasty foods even when they’re not hungry. Our inability to forego these rewarding aspects of food intake override long-term homeostatic control, contributing to obesity.

Eating for Reward vs. Survival
Although the hypothalamus will direct intake based on the metabolic value of the food—when you’re very hungry, you seek out food with lots of calories—it remains to be determined whether the dopamine reward system can also sense a food’s energy content. In other words, does the dopamine system care about calories, or is it just concerned with taste and pleasure? Neuroscientist Ivan de Araujo and colleagues at Duke University (de Araujo is now at the John Pierce Laboratory, a research institute affiliated with Yale University) explored this question by using a line of mice genetically engineered to lack a functional receptor essential for detecting the taste of sweetness. In these mice, any change in reward behavior cannot be due to food palatability or the sensation of sweetness. If these mice prefer sweetness, thus, it is because sweeter foods have more calories, implying that there is something inherently rewarding about the consumption of calories.

In the first set of behavioral experiments, the authors showed that the genetically altered mice were completely insensitive to the “sweet” rewarding properties of sucrose (table sugar) and showed no preference for sucrose compared with water. In contrast, control mice without the genetic mutation strongly preferred the sucrose solution.

The scientists then exposed the different strains of mice to a “conditioning protocol” in which the rodents received alternating access to water or sucrose for six days. During these conditioning sessions, the genetically altered mice were able to associate the sweet solutions with caloric load post-ingestion, as the sugar water has more calories than plain water. Interestingly, both strains of mice now consumed significantly more sucrose. Although the genetically altered mice couldn’t taste the sweetness, they learned to prefer the sweeter water. This finding suggests that mice without functional sweet taste receptors were able to detect the reinforcing caloric properties of sucrose in the absence of sweet taste receptors. There seems to be something inherently pleasurable about ingesting food that contains calories.

As a critical control, the experiments were then repeated with sucralose (a.k.a. Splenda), an artificial sweetener that tastes sweet but contains no calories. Although normal mice consumed more sucralose than water during the conditioning period—they still preferred the sweet taste—the genetically altered mice did not.

Sweetness as Reward
These results indicate that sensing metabolic value can influence feeding behavior. It remained to be determined whether the dopamine reward system, known to respond to sweet taste, was also involved in calorie monitoring, however. To address this important question, the authors showed that calorie load increases dopamine levels in an area of the brain called the nucleus accumbens independent of taste in genetically altered mice using a technique known as in vivo microdialysis. Although both sucrose and sucralose increased dopamine above baseline in normal mice, genetically altered mice only showed an increase in dopamine with real sugar, indicating that caloric load (and not the taste of sweetness) was triggering their dopamine reward system. Even though these results undoubtedly show that calorie load affects the brain dopamine reward system independent of taste in the genetically altered mice, normal mice show no greater dopamine release for sucrose compared with sucralose. This discovery suggests that the presence of calories does not add more reinforcing power to the reward than taste alone. Future studies are needed to clarify whether this calorie load component can affect obesity independent of the taste of food.

One important caveat must be mentioned; all experiments involved food and water deprivation. Thus, the observed activation of the dopamine reward system by calorie load in this paper may be affected by the food-restricted state of the mice.

This study raises many intriguing future questions. How is calorie load sensed by the dopamine reward system? Do certain sugars (for instance, fructose) affect the dopamine reward system in different ways? Does the same phenomenon occur when calories come from different food types? For example, do calories from fat have a stronger affect? All of these questions are important to understand the underlying causes of human obesity. Understanding the rewarding properties of certain foods will help us to design effective ways to restrict the desire to eat once the need to eat is fulfilled. This study also adds to the growing body of information showing that metabolic cues are not solely the domain of the hypothalamus and that much more crosstalk occurs between metabolic cues and higher brain centers involved with the desire to eat than previously believed. Thus, categorizing food intake as hedonic versus homeostatic may not only be redundant but misleading. When it comes to eating, need and desire aren’t so separate after all.

Are you a scientist? Have you recently read a peer-reviewed paper that you want to write about? Then contact Mind Matters editor Jonah Lehrer, the science writer behind the blog The Frontal Cortex and the book Proust Was a Neuroscientist.