Introduction

In another Cyberounds^®, we discussed the psychological aspects of poor or self-damaging health choices made by our patients. In this Cyberounds^®, we consider various neurobiological aspects of systems involved in reward and reinforcement and the control of rewards and reinforcing behaviors. A better understanding of these processes will deepen our understanding of impulsive behaviors with associated negative consequences -- such as overeating or drinking too much. The participant is advised that we can only introduce these topics in this conference.

Almost all of the negative health behaviors discussed were associated with the ingestion of a substance such as food, drugs, or alcohol. During the last three decades, it has become clear that the reward reinforcement system of our brains is involved in a wide variety of motivated behaviors, such as eating, drinking, and sexual behavior.(1) This system, however, not only mediates these natural behaviors which are critical for survival, but also mediates the use of drugs and alcohol and palatable foods which can (if abused) result in severe morbidity and mortality. Below we discuss briefly what is known about the anatomy and neurochemistry of that reward reinforcement system.

Discussion

When talking about addictive or impulsive behavior it is easy to overlook the fact that most of the time humans do exercise some sort of choice over whether or not to engage in these behaviors. Most reinforcing or rewarding behaviors are not persistently or addictively pursued. Rather, the healthy individual chooses a balanced variety of reinforcing behaviors. A part of this choice system which might be called the Reflectivity System (because it allows individuals to weigh, in both a purposeful and automatic manner, the potential negative consequences of an action) which is thought to involve frontal cortical and limbic connections. Serotonin is implicated as an important neurochemical in this control system.(2) These two systems, the reward reinforcement pathway and the reflectivity or control system, must be well integrated for good behavioral health choices. Let's turn to a consideration of these systems.

It has become increasingly clear that a final common pathway for the reinforcing effects from both natural and nonnatural reinforcers/rewards involves the dopaminergic path extending from the ventral tegmental area to the nucleus accumbens and forebrain.(3) Perhaps the most compelling and easily understandable set of experiments demonstrating the importance of dopamine involves placing a very small probe to sample intercellular synaptic fluid in the nucleus accumbens (NB: it is unclear if the fluid obtained is truly from the synapse). Once the probe is in place, an experimenter can measure changes in neurochemical levels in this area which are associated with various behaviors. If dopamine is critical in the reward or reinforcement activity of this area, one would expect an increase in dopamine levels with use of a reinforcing substance and a decrease in dopamine levels with use of a non-reinforcing or punishing substance.

Dr. Bartley Hoebel and his colleagues have published a series of papers over the past ten years which show that ingestion of a variety of substances of abuse, as well as food and water, results in an increase in dopamine levels in the intercellular fluid in the nucleus accumbens.(4),(5) Perhaps the most intriguing experiment done by this group, however, is a demonstration of how experience can affect the dopaminergic response in the nucleus accumbens.(6) Saccharine is a sweet substance which the rats would normally ingest voluntarily. Ingestion of saccharine was initially demonstrated by these investigators to cause an increase in dopamine levels in the nucleus accumbens dialysate.

The experimenters next administered lithium in associated with saccharine. Pairing lithium with saccharine has been shown to result in decreased ingestion and even avoidance of saccharine (i.e., a conditioned taste aversion occurs). After the pairing of lithium and saccharine, the saccharine no longer increases dopamine levels in the nucleus accumbens; in fact, dopamine levels decreased in response to saccharine. Thus the same stimulus, saccharine, results in a dramatically different dopaminergic response in the nucleus accumbens after the experience of pairing saccharine with sickness.

Importantly, these microdialysis findings regarding dopamine and reinforcement are supported by studies showing that microinjections of dopamine antagonists can also block reinforcement.(1),(3) While cocaine and amphetamine act directly on dopaminergic neurons, it appears that alcohol and opioidergic drugs affect neurons with opioid receptors which subsequently act to alter dopaminergic systems in the nucleus accumbens.

The discovery that this dopaminergic pathway from ventral to mental area to forebrain was critical to reward and reinforcement has led many investigators to speculate that there are abnormalities in this pathway in alcoholics and other patients troubled by excessive use of reinforcers. For example, Blum et al announced in 1990 that an allele for the dopamine receptor D2 (DRD2) appeared to be implicated in severe cases of alcoholism.(7)

While this hypothesis about DRD2 has not been clearly supported in a variety of human studies searching for genetic differences in dopaminergic systems, studies of animals genetically breed to self-administer alcohol do show differences in dopamine measures." Jack Modell and colleagues did test the hypothesis that alcohol self administration is mediated by dopaminergic mechanisms in humans when they had subjects self administer alcohol in a laboratory situation while taking either a low dose haloperidol, a dopamine receptor blocker, or placebo. In fact, haloperidol did decrease self administration of alcohol in this laboratory situation.(8) Unfortunately, thus far, there is not a dopamine blocker useable for treating drug addiction in humans because of unacceptable motor side effects such as tardive dyskinesia.

As noted above, some substances such as alcohol appear to have their reinforcing effects mediated by an opioid step prior to dopaminergic involvement. Therefore, opioid antagonists such as naltrexone have been tried in the treatment of alcohol dependent patients. When naltrexone or placebo is given to alcohol-dependent subjects who have completed an alcohol treatment program, the naltrexone-treated patients relapse significantly less frequently and report less craving for alcohol than the placebo-treated patients.(9),(10),(11) Moreover, the naltrexone-treated patients report less positive effects from sampling alcohol which would be consistent with the hypothesis that naltrexone blocked the reinforcing/rewarding effects of alcohol. Thus, our new understanding of the reinforcement/reward pathway is leading to new treatments.

Given the fact that both natural and nonnatural reinforcers or rewards activate the dopaminergic pathway, it was hypothesized that animals that liked highly palatable foods would also prefer substances of abuse. Rats were sorted by preferences for sweet solutions; those rats with high preferences for sweet solutions self administered more alcohol than rats with low preferences for sweet solutions.(12) Conversely, rats bred for high preferences for alcohol showed high preferences for sweet solutions while rats bred for low preferences for alcohol showed low preferences for sweet solutions.(13) This linkage of preferences for natural and nonnatural reinforcers extended to rats that were bred for high or low rates of electrical self stimulation of the reinforcement pathway. In fact, rats bred for high self stimulation frequencies in the reward system had high preferences for sweet solutions while those with low self stimulation frequencies had low preferences for sweet solutions.(14) Moreover, when a laboratory animal is deprived of a reinforcer such as a sweet solution it appears that the animal will compensate for this deprivation by an increase of use of an alternate reinforcer such as a drug.(15)

This substitution of one activator of the dopaminergic reward system for another may underlie some recent results in humans. In unpublished results, our group has shown that alcoholics early in abstinence report increased likelihood of binge eating and increased preferences for sweet foods. Kampov-Polevoy and colleagues (16) showed that during the early phases of abstinence, a higher number of alcoholics report maximum preferences for highly sweet solutions than do nonalcoholic people. Yung Gordis and colleagues (17) showed several years ago that alcoholics who had remained abstinent longest after treatment were those who used the most table sugar. It is possible that humans who are trying the stop the use of one reinforcer are turning to the use of another reinforcer which would activate the reward pathway. However, this remains to be proven as no one has demonstrated that dopaminergic mechanisms mediate these substitutions. Also, it remains to be shown whether these substitutions are beneficial to subjects in the long run.

One must wonder when reading the reinforcement literature why animals or humans who are using a positive reinforcer ever stop the use of that reinforcer. This brings up the topic of how animals and humans make choices about the use of one reinforcer versus another. It also raises the issue of how an animal chooses to stop using a reinforcer (i.e., when he or she registers satiation). It appears that serotonin is importantly involved in both reflection on the use of reinforcers and in satiety functions. There is now a compelling body of data which supports the hypothesis that patients who are obsessive (i.e., those who reflect excessively prior to performing a behavior which might be wrong) are characterized by high metabolic rates in their frontal areas on PET scans and by high serotonergic measures. Conversely, impulsivity has been linked to low serotonergic function. For example, patients who attempt impulsive suicides and impulsive homicidal behavior, and those with severe early-onset alcoholism as well as bulimia all have low levels of serotonin metabolites in their cerebrospinal fluid.(18)

The hypothesis that there is an important balance between serotonergic and dopaminergic pathways which determines the balance between acting impulsively and reflecting excessively remains to be proven. However, one might posit that an individual with an excessively active dopaminergic reward system and an underactive serotonergic control system might be prone to excessive use of reinforcers which could have negative effects on health. Conversely, if one is characterized by low dopaminergic and high serotonergic function in the relevant areas of the brain. Then excessive obsessing and too little action for optimal function might ensue.

In the future it is hoped that we have a much clearer understanding of these balances in underlying behavioral choices and that these can be modulated by a both psychotherapy and psychopharmocologic interventions so as to enhance the health of our patients. In the meantime, we can perhaps use the nascent knowledge we do have to view addiction and impulsive behaviors with more enlightened compassion.