Please note, this is recovered content from the former website of the New York Obesity Research Center website.


Dopamine rewards of different foods, pleasure and addiction

Date: May 10th, 2007

Speaker’s Name, Affiliation, and Seminar Title:

  • Roy Wise, NIDA. “Conditioned and unconditioned reward-related input to the dopamine system”
  • Allen Levine, University of Minnesota. “Opiods and Food Reward”
  • Leonard Epstein, Univ. Buffalo. “Behavioral Economics of Eating”
  • Kenneth Carr, NYU. “Eating Disorders and Addictions”
  • Dana Small, Yale University. “Application of the Incentive-Sensitization Theory of Drug Addiction in Overeating”
  • Mary Boggiano, University of Alabama at Birmingham, Birmingham, AL..”Factors in Animal Models that Change Feeding from a Homeostastic to Allostatic Function”
  • Richard Foltin. “Substance Abuse and Eating Disorders”

Symposium Title:“Neural Pathways for Natural and Artificial Reward”


  • Roy Wise, “Conditioned and unconditioned reward-related input to the dopamine system”

Dr. Halls Dr. Halls
This is important stuff. A great discussion.


Conditioning by dopamine rewards, also cocaine, heroin acetylcholine, glutamate.


The talk had two parts. First, I discussed evidence establishing that the importance of dopamine for the motivation to seek a reward is secondary to the importance of dopamine for stamping in the memory of prior experience with that reward. The importance of dopamine for “wanting” a reward in the future is secondary to the importance of dopamine for “liking” the reward in the past (JCPP 1978, 92, 661-71; Science, 1978, 201, 262-4; Psychopharmacol, 1984, 84, 446-51; Brain Res, 1986, 368, 62-8).

Thus alley running for intravenous heroin is not affected by haloperidol on the first trial when the animal experieinces heroin under haloperidol treatment, but the running for heroin on the next trial is degraded (Behav Neurosci, 112, 630-5).

Neuroleptics attenuate motivation for a reward only after degrading the remembered strength of the reward in question.

Second, I discussed recent microdialysis experiments in which we identified two excitatory inputs to the dopaminergic reward mechanism of the ventral tegmental area. The first was glutamate, which is released in trained animals as they initiate sessions of intravenous cocaine self-administration. Similar release is seen in initial extinction sessions, indicating that the glutamate release is associated with cues that trigger the expectancy of cocaine and not with the receipt of cocaine itself.

As the expectancy of cocaine is extinguished by repeated extinction trials, so is the glutamate response extinguished. The glutamate appears to contribute to the total reward experience; blocking glutamate receptors causes rats to respond as if the rewarding dose of cocaine had been lowered and it causes immediate response cessation in extinction trials (when responding is normally maintained for a time by conditioned reinforcers).

The second input was acetylcholine. Like glutamate, acetylcholine is released when trained rats expect to receive cocaine; unlike glutamate, acetylcholine shows a phasic response to cocaine itself.


Q. When you showed the effects of replacing saline with cocaine, and showed the glutamate peaks that resulted, are those time sensitive?
A. We didn’t measure that response across different days, so we don’t know.

Q. Is there any parallel with sweet taste, or with fat taste?
A. We are working to determine that.


  • Allen Levine, “Interactions between opioids and food reward”


Morphine increases fat intake in diet


Eating is fueled by a number of environmental, behavioral, and social cues, but one of the most important promoters of eating is taste AND REWARD. Dr. Levine presented data on the role of peptides, including opioids, in stimulating intake of fat and sweet. Morphine has been shown to preferentially stimulate fat intake in animals, and these effects are increased after food deprivation. However, the preference a rat has for fat or carbohydrate also affects the morphine-related feeding. Blocking opiods, with administration of naltrexone, can reduce intake of preferred foods, and of sweet tasting foods, suggesting that the opioid system is also involved in liking and intake of these foods. However, naltrexone fails to appreciably alter either the acquisition or the learning of a flavor preference conditioned by sucrose, indicating that opiods might have a role in inherent palatability, but not in learning of preferences. Opioids are likely involved along the preference continuum, and in maintenance of meal size intake, because of their ability to reduce the aversion associated with a very large meaL, and increase the palatability.

Dr. Levine also mentioned several experiments addressing the question of how we get addicted to foods? During withdrawal from drugs, the release of dopamine decreases, while release of acetylcholine increases. Injection of nalrexone to rats given chronic episodic exposure to sucrose solutions results in a similar ration of dopamine to acetylcholine that is observed in rats withdrawing from morphine. Rats that are prefer sweet tastes also tend to self-administer more cocaine, morphine, and alcohol. Rats that prefer saccharin self-administer more ethanol than those that do not prefer saccharin. These and other data suggest some similarity between the pathways of both drug and food “addiction” or dependence.


Q. The graph on the left, is the saccharin used as an alternative reinforcer?
A. Yes.


  • Leonard H. Epstein, “Behavioral Economics of Eating”


Variations in Dopamine receptor phonotype, Taq 1 A1 allele


Behavioral economics involves understanding factors that influence choice, between two or more alternatives. Choices can be current (eg., eating healthy food or junk food now), or concurrent (eg., eating healthy food now to lose weight later). Studying participants’ choices, between alternatives, can help researchers understand the relative reinforcing values of foods and/or drugs.

While behavioral economics has long been used to study the relative reinforcing value of drugs of abuse, Dr. Epstein has applied this science to determine conditions where fruits and vegetables might be substituted in place of higher fat and energy food choices. In theory, reducing consumption of a high-energy, palatable food can be done by either decreasing its reward value, or increasing the value of competing activities. In this way, people might be encouraged to eat more fresh fruits and vegetables by adapting the environment, such that they cost less, or are easier to get. To close his talk, Dr. Epstein stated the possible connections between food reinforcement and obesity. High levels of food reinforcement might be expected in someone who has higher levels (or risk for) obesity. In fact, research from Dr. Epstein’s lab has indeed shown that obese subjects find food more reinforcing than do leans. The highest levels of food reinforcement is associated with variations in the dopamine (D2) receptor genotype, specifically at the Taq 1 A1 allele.

Given that obese and overweight subjects might have genetic variations that make foods more reinforcing to them, they might be less affected by changes that reduce the appeal of reinforcing foods (such as changes in price).


Q. In the model of pricing, do you know how people cognitively determine this breakpoint, where the price of the reward becomes high enough to change their preferred choice?
A. That is a great question, but I don’t know anyone who’s thinking about the problem that systematically.

Q. In the experiment with mothers, what was the simulation you did? Do you think that different groups of people will respond to that situation differently?
A. It’s possible. There is some indication that in different settings, the correlation is higher.

Q. Schacter found the opposite of what you did with obese participants. He found that the didn’t work as hard for choices. Can you explain the difference?
A. The Schacter experiments were more about externality. We do are studies when subjects are not food deprived, so that could have made a differenece.

Q. What is the relationship between variations in the dopamine receptors and BMI?
A. We seed a positive relationship between the D2 dopamine receptor and BMI.

Q. I wasn’t clear on what the food reinforcement activity was.
A. Subjects had the choice to eat palatable foods or read the paper (which is not a very palatable reinforcing stimulus).

Q. What would your advice be to people who binge repeatedly?
A. My advice would be to use alternative and environmental reinforces to increase the preferences for healthy foods, while decreasing the temptation to eat unhealthy foods.


  • Kenneth A. Carr, “Food restriction: enhancing effects on drug reward and striatal cell signaling”

Dopamine D1


Dr. Carr’s laboratory studies the modulation of brain reward circuitry and behavior by food restriction. Chronic food restriction increases the unconditioned rewarding effects of numerous drugs of abuse, whether injected systemically, intracerebroventricularly, or directly into nucleus accumbens. The food-restricted subjects are also unique in displaying a high and escalating rate of wheel-running in a brief access (30 min) protocol.

A working hypothesis is that a common set of neuroadaptations account for increased drug reward magnitude and “compulsive” exercise during prolonged food restriction and maintenance of low body weight. In tissue preparations, a variety of pre- and postsynaptic neuroadaptations have been identified in nucleus accumbens and/or whole striatum of food-restricted rats. These include increased activation of ERK 1/2 MAP kinase, CaM kinase II, CREB, and increased c-fos, preprodynorphin and preprotachykinin gene expression in response to D-1 dopamine receptor stimulation.

Recent findings, using appropriate antagonists, confirm the involvement of the D-1 dopamine receptor in the enhanced drug reward magnitude and wheel-running of food-restricted rats, but fail to support involvement of increased MAP kinase signaling in these behavioral responses. In contrast, the increased MAP kinase signaling has been shown necessary for the downstream increase in CREB phosphorylation and c-fos expression in response to D-1 dopamine agonist stimulation.

Studies in progress are testing the hypothesis that the upregulated cell signaling and transcriptional responses are involved in experience-dependent neuroplastic changes that will be evident in associative learning paradigms. Overall, these studies may illuminate general motivational mechanisms as well as provide insight into the known relationship between food restriction/severe dieting and several forms of maladaptive goal-directed behavior in clinical populations including binge-eating, substance abuse, and compulsive exercise.


Q. Have you looked at the effect of the specific nutrient restriction on outcome measures?
A. I’d liked to, but have not got around to that yet.

Q. What is the effect of choice on individuals who are chronically restricting their food intake (like anorexics)?
A. That is a good question. I’m not sure if the long-term food restriction would increase vulnerability to alternative behaviors. It’s interesting, but I don’t know the answer.

Q. Your results and the previous two speakers remind me of the results we reported years ago. Food restriction in animals results in a decreased preference for sweets, no matter what the alternative is. Do you understand what is going on?

A. It doesn’t make sense to me, but it’s similar to what we’ve found.


  • Dana Small, “Application of Incentive-Sensitization Theory of Addiction to Overeating”

High BMI people rate foods as having less pleasure, so may need to eat more


The obesity epidemic is a recent phenomenon and is therefore not caused by a drift in the genome. Researchers have suggested that individual differences in food reward interact with environmental changes to predispose some individuals to overeat. Work in Dr. Small’s lab has shown that 1) there are overlapping regions encoding food and drug reward in humans and 2) there are separable networks encoding the anticipatory vs. the consummatory phase of food reward. Ongoing work in her lab is aimed at testing whether individual differences in anticipatory and/or consummatory food reward predict body mass index and to determine whether similar mechanisms account for compulsive eating and drug administration.

Preliminary data provide evidenced for heightened neural response (amygdala and OFC) and hedonic ratings of food but not floral aromas in high compared to low BMI subjects. These findings are in accordance with the incentive-sensitization theory of addiction proposed by Berridge and Robinson. Preliminary findings also show that high BMI subjects rate the perceived pleasure of eating food as lower than low BMI subjects. Accordingly, a negative relationship between BMI and response in medial OFC, a region known to encode the perceived pleasure of eating food, was observed. These findings support the reward deficiency model of compulsive behavior. Finally, strong correlations between BMI and response during food receipt were identified in the ventral pallidum, parahippocampal gyrus, midbrain and amygdala. Taken together these data indicate that there are multiple differences in food reward between individuals with high vs. low BMI and that heightened limbic responses may be observed during food consumption despite lower reported perceived pleasure.


Q. How did you avoid subjects becoming habituated to the stimulus?
A. We don’t see a lot of habituation, as there is a lot of time between presentation of the stimuli.

Q. Did you account for binge eating disorder? In our studies, we’ve found differences in binge vs. non-binge eating, depending on response to food cues.
A. We are looking at that, but don’t have enough power yet.

Q. Does giving away a small sample of good tasting food differ when you have obese subjects? They might be suffering from increased, or a greater level, of food deprivation.
A. Yes, that might be occurring.


  • Mary Boggiano, “Factors in animals models that change feeding from a homeostatic to an allostatic function”

Stress is reduced by binge-eating.


Allostasis is defined as maintaining internal viability among changing conditions. While mechanisms are designed to adapt to environmental changes, there are individual changes in the ability to do so. Both binge-eating and obesity can be studying in the context of allostasis, in that they are examples of maintaining viability amongst radically changed environmental conditions. Broad and accessible availability to high energy fast foods and increased portion sizes at restaurants and in grocery stores are examples of how obesity is a response to this changed environment. Furthermore, the demands for physical activity are much lower than those of our ancestors, further encouraging a state of energy excess. Obesity and eating disorders are present, according to the speaker, because food no longer fulfills just a metabolic function, but rather provides an allostatic function. Understanding the process by which food fulfills an allostatic function can result in better treatments for both obesity and eating disorders.

Dr. Boggiano presented several experiments using animal models to study effects of stress, access to palatable foods, and history of dieting on food intake. Data suggest that binge-eating may be a mechanism by which animals reduce levels of stress, and thus be serving an allostatic function. Animals with a history of dieting tend to consume more palatable foods (cookies) during stressful conditions, compared to stress alone (without a history of dieting). This interaction is an example of allostasis, the animal maintaining internal viability among changing conditions.


Q. Is there a pattern to when animals will stop eating after stress?
A. The strongest effects are seen 4 hours post stress, but animals will respond weakly up to 24 hour post stress.

Q. How long do you stress animals, and are all the stressors equally effective?
A. Yes, although I don’t know about the tail pinch. But most of the stressors are the same.


  • Richard Foltin, “Substance Abuse and Eating Disorders”

Eating disorders, exercise and binge eating


Just as only a small number of people who try drugs become drug abusers, only a small number of people who eat develop eating disorders. It is unclear how informative studies that contain non-drug abusers, or people without eating disorders actually are. Just as in drug abuse, eating disorders can be seen along a continuum of eating behavior, from normal to disordered. Thus, in order to use normal individuals in studies, we must assume that studying normal behavior (both eating and drug related), will somehow inform us on the mechanisms behind these processes. These mechanisms can further be applied to understanding the processes of drug abuse and eating disorders.

Drugs are typically used for reasons of both positive reinforcement (to increase positive feelings or performance), or for negative reinforcement (to decrease physical symptoms of an illness, or avoid withdrawal symptoms). Initial feelings of positive or negative reinforcement may be followed quickly by feelings guilt or despair (negative effects). A question that follows is whether or not behaviors like eating or exercising can be reinforcing.

Research suggests that, in fact, exercise is reinforcing to anorexic patients, while binge eating can be reinforcing to bulimic patients. But, because bulimic patients will not binge if they cannot purge, perhaps purging is also a reinforcement to bulimic subjects. The continued identification of similarities between eating disorders and drug abuse will allow the large amount of data available from the drug abuse field to be applied to the study of eating disorders and obesity.


Q. You showed a nice relationship between the amount of drug and dopamine. Is there a way of showing this with food?
A. The effect with food is consistently much less (only about 15%) of what the effect with drugs are.

Q. Does it depend on animals liking of the food being offered?
A. At least some studies have shown that it’s the stimulus alone, and not the amount of stimulus being offered.

Q. Is it possible that both the appetizer and stimulant might increase animals focus on a task?
A. Amphetamines definitely increase animal’s focus on learning.

Q. Is binging & purging reinforcing? Are there any parallel behaviors in drug abuse?
A. Only vomiting after use of opiates might be a parallel.