RESEARCH, ANIMAL MODEL

The articles in this section describe studies of the effects of drugs on animals in the laboratory. These studies are important because many of our current beliefs about the nature of drug dependence involve concepts of learning and reinforcement, and many recently developed treatments are founded on these beliefs. The section contains An Overview of Drug Abuse research using animal models and detailed articles on various research concepts beings explored in this way: Conditioned Place Preference; Conditioned Withdrawal; Drug Discrimination Studies; Drug Self-Administration; Environmental Influences on Drug Effects; Learning, Conditioning, and Drug Effects—An Overview; Learning Modifies Drug Effects; Learning Modifier Drug Effects; Operant Learning Is Affected by Drugs.

See also Aggression and Drugs: Research Issues; Motivation and Incentives ; and the articles in the section entitled Research.

An Overview of Drug Abuse

A great deal of biomedical research is based on the belief that only through careful scientific analysis will we achieve a sound understanding of the problem of drug abuse and how to control it. Animal models of a human condition are an integral part of that analysis. Animal models were developed to help us understand the factors that control drug abuse. Under laboratory conditions it is possible to control many factors, such as the environment, genetics, drug history, and behavioral history, that cannot easily be controlled outside the laboratory. When these factors can be controlled, their influence on drug abuse can be precisely determined. As always, the use of animals involves the assumption that the behavior of animals is a valid model of a human disorder. The drug abuse research that has been conducted to this point makes it clear that this is a valid assumption.

There are three major animal models of aspects of drug abuse to consider: Physical Dependence, drug self-administration, and drug discrimination. Each of these has provided basic information about the fundamental processes that control drug abuse. In addition, each has provided practical information about the abuse potential of new drugs. Information on both of these topics represents an important contribution of animal research to solving the problems of drug abuse.

PHYSICAL DEPENDENCE

Often when a drug is administered repeatedly, Tolerance develops to its effects. That is, the dose of drug that is taken must be increased to achieve the same effect. With prolonged exposure to high doses, physical (or physiological) dependence may develop. That is, the person is dependent on the drug for normal physiological functioning. The existence of physical dependence is revealed when drug administration is stopped. When the drug is no longer administered, various physical changes begin to appear. Depending on the specific drug, these could include autonomic signs (e.g., diarrhea and vomiting), somatomotor signs (e.g., exaggerated reflexes, convulsions), and behavioral signs (e.g., decreases in food and water intake). These effects have been called withdrawal but in the literature are also known as abstinence syndrome.

Historically, it was believed that physical dependence was the cause of drug addiction. That is, it was felt that one had to become physically dependent on a drug before abuse would occur and that the drug dependence or addiction was motivated by the need to relieve the abstinence syndrome. One of the major contributions of modern drug abuse research has been to make it clear that this is not true. In fact, much drug abuse occurs in people who are not physically dependent. Nevertheless, since the need to avoid the abstinence syndrome can increase the likelihood that a person will continue to abuse a drug, it is important that we understand physical dependence. Also, it would be desirable for new drugs that are developed not to produce physical dependence.

The development of physical dependence is most common with the Opioids (morphine and morphine-like drugs) and central nervous system (CNS) depressants (e.g., Barbiturates and Alcohol). Since opioids are very valuable painkillers but produce physical dependence when used repeatedly, there has been great interest in the development of drugs that can kill pain but do not produce physical dependence. Standard approaches to testing new opioids in animals for their potential for inducing physical dependence have been developed. In the early stages of testing, a new drug that has been found to be an effective Analgesic is given to an animal that is physically dependent on morphine (mice, rats, dogs, and monkeys have been used). After giving the drug, a trained observer scores the occurrence, intensity, and duration of abstinence signs such as shivering, restlessness, irritability, abdominal cramps, vomiting, diarrhea, and decreased eating and drinking. The drug may not affect the abstinence syndrome; it may relieve it or it may make the syndrome worse. A drug that relieves morphine abstinence probably will produce morphine-like physical dependence and may not be considered for further development on this basis. On the other hand, a drug that has no effect on abstinence, or even makes it worse, probably will not produce morphine-like physical dependence and may be worth pursuing. Often such a drug will be evaluated for its ability to produce physical dependence when it is administered repeatedly. A drug that produces no physical dependence of its own is clearly a candidate for further development.

Literally hundreds of new opioid drugs have been evaluated in animals for their capacity to produce physical dependence, and, in the process, we have learned much about physical dependence. It is clear that the higher the drug dose and the more frequent the exposure, the more intensive the physical dependence that develops. But recent research with human subjects has strongly suggested that even a single dose of an opioid may produce some level of physical dependence. Research has also shown that drugs that suppress the signs of morphine abstinence in a dependent animal generally have morphine-like effects themselves. That is, they suppress respiration and cough, kill pain, and have the potential to be abused and produce physical dependence. These drugs are known as opioid agonists. Other drugs, known as opioid Antagonists, may cause abstinence signs and symptoms to appear. Opioid antagonists do not have morphine-like effects themselves but are capable of blocking or reversing the effects of morphine and morphine-like drugs. Still other drugs, called mixed Agonist-Antagonists, can have either type of effect, depending on dose and whether the animal is physically dependent. This group of drugs has proven particularly interesting in terms of its contribution to our understanding of how opioids work. In addition, many of them are effective analgesics with apparently low potential to produce physical dependence.

Other classes of drugs besides opioids produce physical dependence in animals as well. Many of the basic findings about physical dependence on CNS depressants (e.g., dose and frequency) are similar to what has been found with opioids. However, the abstinence syndrome can be even more severe than that seen with opioids. Hallucinations and even life-threatening convulsions can develop when long-term abuse of a barbiturate or alcohol is stopped. Abstinence syndromes have also been found after long-term exposure to Tetrahydrocannabinol (THC), the active ingredient in Marijuana, and Phencyclidine (PCP). On the other hand, the abstinence syndrome that follows long-term exposure to such CNS stimulants as Amphetamine or Cocaine is, by comparison, mild.

DRUG SELF-ADMINISTRATION

The distinguishing characteristic of drug abuse is the behavior of drug self-administration. When that behavior becomes excessive and has adverse consequences for the individual or society, the individual is considered to be a drug abuser. Therefore, the development of animal models for studying drug self-administration was the essential first step toward identifying factors that control the behavior. Humans consume drugs by several different routes of administration, including oral (e.g., alcohol), intravenous (e.g., cocaine and heroin), and inhalation (e.g., nicotine and crack cocaine). Although some of the factors that control drug abuse may be independent of the route of administration, others may not. Therefore, it has been important to develop models in which animals self-administer drugs by each of these routes.

Early attempts to study drug self-administration in animals involved oral self-administration. Oral self-administration of drugs has proven difficult to establish in laboratory animals, probably because most drug solutions have a bitter taste. Also, when consumed orally, the onset of the drug effect is relatively slow, making it difficult for the animal to make the association between drinking and drug effect. For these reasons, when first given a choice between water and a drug solution, most animals choose the water. However, conditions can be arranged so that the animal drinks large amounts of the drug solution in relatively short periods, by making the drug solution available when food is available, either as a meal or delivered repeatedly as small pellets of food. After a period of drug consumption in association with food, food can be removed from the experiment and the animal will continue to consume the drug orally. When given a choice between the drug solution and water, the animal will prefer the drug solution. This approach has been particularly important for research with alcohol, since humans abuse this drug orally.

To study intravenous self-administration, an animal is surgically implanted with a chronic intravenous catheter through which a drug can be administered. The animal wears a backpack and tether that protect the catheter and attach to a wall of the cage. The cage usually has levers that the animal can press to receive a drug injection and lights that can be turned on to signal that a drug injection is available. At that time, a lever press turns on an electric pump and injects a drug solution through the catheter into the vein. In this way, the animal model mimics intravenous drug injection by humans using a syringe. Since taste is not a factor and onset of drug action is rapid, conditioning animals to inject drugs by the intravenous route has proven relatively straightforward.

Reliable methods for administering drugs to animals by inhalation are important for studying the abuse of drugs that are inhaled, such as Tobacco, Solvents, or Crack. Methods for studying solvent inhalation have been available for several years. Usually an animal is given the opportunity to press a lever to deliver a brief bolus of solvent vapor to the area around its nose. Methods for studying crack cocaine smoking in monkeys have only recently been developed. Monkeys are first trained to suck on a drinking tube; then the apparatus is arranged so that sucking on the tube delivers crack smoke to the monkey. Although the technique is new, it shows promise for the study of smoking in laboratory animals.

Research using these animal models has shown that, with few exceptions, animals self-administer the same drugs that humans abuse and show similar patterns of intake. For example, when given unlimited access to stimulants like amphetamine, both humans and animals alternate periods of high drug intake with periods of no drug intake. In the case of heroin, both animals and humans gradually increase drug intake to levels that are then stable for months and even years. In addition, animals do not self-administer drugs that humans do not abuse (e.g., aspirin) and even avoid those that humans report to be unpleasant (e.g., Antipsychotic Drugs). These basic findings validate this as an excellent animal model of drug abuse by humans. The exceptions are the hallucinogens and marijuana, which animals do not readily self-administer.

Research using the self-administration model has increased our understanding of drug abuse in several different areas. It has become clear that drug self-administration is controlled by events that are initiated inside (e.g., a drug-induced change in brain chemistry) or outside (e.g., stress) the organism. With regard to events initiated inside the organism, we have begun to learn about the Neurotransmitter systems in the brain that are activated when drugs are self-administered. These changes are probably responsible for producing the drug effect that people find desirable and that maintains their self-administration (the reinforcing effect). A substantial amount of recent research has focused on the neurotransmitter changes that are involved in the reinforcing effect of cocaine. It has been known for some time that cocaine increases the concentration of certain neurotransmitters in synapses. Research indicates that it is this effect on certain synapses in the CNS that use the neurotransmitter Dopamine in the brain that almost certainly plays the primary role in cocaine's reinforcing effect. Similar research suggests that the neurotransmitter Serotonin may play a primary role in the effects of alcohol.

Even though neurotransmitter changes occur when an individual self-administers a drug, they are not always sufficient to maintain drug self-administration or to make it excessive. Events initiated outside the organism—that is, environmental events—can increase or decrease drug self-administration. In the case of alcohol, for example, consumption can be increased in animals simply by presenting other things of value (e.g., food pellets) every few minutes. Although it is not known exactly why this occurs, analogous conditions may increase the consumption of alcohol and other drugs by some humans. Drug self-administration can also be decreased by environmental conditions. For example, increasing the cost of a drug or the effort required to obtain it decreases consumption. Drug self-administration can also be decreased by imposing punishment or by making valuable alternatives to drug self-administration available.

Animal research has also made it clear that certain individuals may, because of their genetic makeup, be more susceptible to the effects of alcohol or other drugs. For example, genetically different strains of rats can differ in their sensitivity to the effects of codeine, morphine, or alcohol. Also, animals can be selectively bred to be more or less sensitive to the effects of a drug. These findings clearly demonstrate a genetic component to drug sensitivity. Research suggests that these animals differ in the amounts of these drugs that they will self-administer. How broadly this conclusion cuts across drugs of abuse is unknown but is an active area of research.

In short, drug abuse research with animals has made it clear that whether drug self-administration occurs depends on an interaction between a drug, an organism, and an environment. A susceptible individual in an environment in which a drug is available and in which conditions encourage drug self-administration is more likely to be a drug abuser than one in which environmental conditions discourage drug abuse.

DRUG DISCRIMINATION

When a person takes a drug of abuse, it has effects that the person feels and can describe. These effects are called subjective effects (versus objective effects that can be seen by an observer), and they play an important role in drug abuse. A person is more likely to abuse a drug that has effects that the person describes as pleasant than one that the person describes as unpleasant.

The subjective effects of drugs of abuse have been studied in humans for many years and in several different ways. Early research involved administering drugs, usually morphine-like drugs, to former heroin addicts who then answered questionnaires that were designed to detect and classify the subjective effects of the drug. The single-dose opiate questionnaire asks the subject whether he or she can feel the drug, to identify the drug, to describe the symptoms, and to rate how much he or she likes it. The Addiction Research Center Inventory consists of a series of true-false statements that describe internal states that might be produced by drugs. The Profile of Mood States is a list of adjectives that can be used to describe mood. Responses to these questionnaires depend on variables such as type of drug and drug dose. Recent research has examined the subjective effects of a wider variety of drugs (including stimulants and depressants) not only in experienced but also inexperienced subjects. The purpose of this research is to understand the factors that can influence a person's subjective response to drugs of abuse.

Since subjective effects require a verbal description of an internal state, they can be directly studied only in humans. Over the past twenty to thirty years, however, it has become clear that animals can be trained to respond in a way that suggests they can detect the internal state produced by a drug. The behavioral paradigm is called Drug Discrimination, and the drug effect is called a discriminative stimulus effect. Although a number of drug-discrimination paradigms have been developed, the most common is a two-lever paradigm in which the animal is trained to press one lever after it has received a drug injection and the second lever after an injection of the drug vehicle or, in some cases, another drug. Responding on the lever that is appropriate to the injection is reinforced, usually by presenting a food pellet, while responding on the incorrect lever is not. If this is done repeatedly over a period of several weeks, the animal learns to respond almost exclusively on the lever associated with the injection. Although it is impossible to know what an animal feels, it seems as if the animal is reporting whether it feels the drug by the lever it presses. The animal can then be asked to "tell" us whether a new drug "feels" like the training drug. It will respond on the drug lever if the new drug is similar to the training drug and on the vehicle lever if it is not. It can also be "asked" whether a drug blocks the effects of the training drug. If the test drug blocks the effect of the training drug, it will respond on the vehicle lever when given both drugs.

There is a strong correspondence between the classification of drugs by humans based on their subjective effects and those by animals based on their discriminative stimulus effects. Research using the drug-discrimination model has increased our understanding of control of behavior by drugs in several different ways. First, this research has made it clear that behavior that is learned under the influence of a drug is more likely to occur again when the drug or a similar drug is taken again. This is a fundamental mechanism by which drugs control behavior. As with drug self-administration, a substantial amount of recent research has focused on the neurotransmitter changes that are involved in the discriminative stimulus effects of cocaine and alcohol. Again, dopamine seems to play a prominent role in this effect of cocaine, while serotonin may mediate the effects of alcohol. Environmental events, by contrast, do not seem to alter the discriminative stimulus effects of drugs substantially. However, little research has been done in this area.

ABUSE LIABILITY TESTING AND TREATMENT RESEARCH

One important application of animal models of drug abuse is the prediction of the likelihood that a new drug will be abused if it is made available to people. Clearly, the prevalence of abuse of a drug can be reduced by restricting its availability, and drugs with high potential for abuse should be the least available. All the models discussed here are used for predicting some aspect of the abuse liability of new drugs. However, the task is not simply a matter of detecting abuse liability and making the drug unavailable. Abuse liability must be considered in the context of any potential therapeutic use of the drug, and a cost-benefit analysis that weighs liability for abuse against therapeutic benefits should be made.

Opioids are an excellent example of these considerations. Morphine is often the only appropriate analgesic for intense Pain. However, it produces physical dependence and has a high potential for abuse. A drug that produces analgesia equivalent to or greater than that of morphine but does not produce physical dependence would be a highly desirable compound. Techniques for establishing this have been described in related articles. A new drug can be tested for its ability to suppress abstinence syndrome in monkeys that are dependent on morphine and for its ability to produce physical dependence of its own type in naive animals. A similar approach is taken with the drug in drug self-administration experiments. We may ask whether the drug maintains self-administration in experienced monkeys or whether naive monkeys will initiate self-administration. In addition, we can evaluate whether the drug is likely to be preferred to morphine by allowing an animal to choose between morphine and the new drug or determining how hard the animal will work to receive an injection of the drug relative to how hard it will work for morphine. Finally, we can ask whether the drug has discriminative stimulus effects that are similar to those of morphine or of any other drug of abuse. A drug that supports physical dependence, is self-administered, and has morphine-like discriminative stimulus effects is likely to have high potential for abuse in humans and unlikely to be a viable substitute for morphine. On the other hand, a drug that lacks one or more (preferably all) of these effects may be worth pursuing.

Animal models of drug abuse have been used for the development of drugs that may be useful in the treatment of drug abuse. In some ways it seems unusual to suggest treating a drug abuse problem with another drug. However, in the case of opioids, Methadone, a morphine-like agonist, has proven to be quite useful in the treatment of opioid dependence. Although the drug is still self-administered and physical dependence is maintained, treatment with methadone allows the person to lead a relatively normal life that does not require the high-cost behaviors (e.g., crime, intravenous injection) associated with abuse of illicit opioids.

The animal models described here, particularly drug self-administration and drug discrimination, are now being applied to the development of drugs that may be useful in treatment. These approaches are based on the reasonable but as yet unvalidated assumption that blocking or mimicking the reinforcing and subjective effects of drugs will decrease drug abuse. In the case of cocaine, dopamine antagonists and, surprisingly, opioids have shown some promise in animal models as potential treatment compounds. It is not yet clear whether these compounds will be effective in humans. Nevertheless, this is an area of active research that shows promise for helping with treatment of drug abuse for development as treatment compounds.

(See also: Abuse Liability of Drugs ; Reinforcement ; Research )

BIBLIOGRAPHY

Brady, J.V., &Lukas, S. E. (1984). Testing drugs for physical dependence potential and abuse liability. NIDA research monograph 52. Washington, DC: U.S. Government Printing Office.

Colpaert, F. C., & Balster, R. L. (1988). Transduction mechanisms of drug stimuli. Berlin: Springer-Verlag.

Woolverton, W. L., & Nader, M. N. (1990). Experimental evaluation of the reinforcing effects of drugs. In Testing and evaluation of drugs of abuse. New York: Alan R. Liss.

William Woolverton

Conditioned Place Preference

A procedure called conditioned place preference has been used to study the "rewarding" effects of drugs. The procedure is designed to ask the question "When given a choice, will an animal prefer an environment in which it has experienced a drug to one in which it has not?" To answer this question, an animal is placed in an experimental chamber that is divided into two compartments that are different in some way. For example, they may have different floors and/or distinctive odors. Initially, the animal is placed in the chamber for several preconditioning trials and the time spent in each compartment is measured. Usually, a rat exhibits some preference for one or the other side in these trials. At this point, the experimenter can do one of two things—(1) modify the compartments in some way, perhaps by changing the lighting, so that equal time is spent in the two chambers before proceeding (balanced procedure), or (2) go ahead with the experiment with unequal preferences (un-balanced procedure). With either procedure, conditioning trials are conducted next.

To run conditioning trials, a barrier is placed in the middle of the chamber that does not allow the animal to switch sides. The drug of interest is then administered to the animal and it is confined to one compartment for usually fifteen to thirty minutes. If the unbalanced procedure is used, the animal is usually placed in the compartment that was initially avoided. A second group may be given a placebo (a substance that has no effect) under these same conditions or a placebo may be given to these same animals before placing them in the second compartment in alternating sessions. In this way, the effect of the drug is associated with a particular environment. After several—three to ten—conditioning sessions, the animal is placed in the chamber without being given the drug, and the door is removed so that the animal can spend time in either compartment. The length of time spent in each chamber is recorded and used as a measure of preference for that chamber.

The hypothesis underlying this sort of experiment is that the length of time spent in an environment should increase if that environment is associated with the effects of a drug of abuse. In fact, many studies have shown that this does happen with drugs such as Heroin, Cocaine, and Amphetamines. In the balanced procedure, animals spend more time in the drug-associated side than in the other side. In the unbalanced procedure, the animals spend more time in the drug-associated side than they did previously, but only rarely demonstrate an actual preference for it. As would be expected, preference is greater with higher doses of the drug and does not occur with placebo injections. In addition, it does not occur with drugs that are not typically abused, such as antipsychotic drugs, antidepressant drugs, and opioid antagonists. Thus, it seems likely that the technique measures a drug effect that is related to drug abuse.

Like other models for studying drug abuse, conditioned place preference has strengths and weaknesses. Among its strengths is that animals are tested in a drug-free state. Therefore, the measure of preference is not influenced by the direct effects of drugs. The procedure can be done with drug injections given by routes other than intravenous, therefore surgical preparation is not involved. Moreover, the procedure is rapid, with maximum effect usually evident within three conditioning sessions.

The major weakness relates to the drug effects that it is measuring. Since drug administration is not due to the behavior of the animal (i.e., self-administration), it is by definition not a reinforcing effect. Although many of the same drugs that are self-administered induce place preferences, it is not clear whether the drug effect studied in conditioned place preference is the same as that studied in procedures that directly measure reinforcing effects. Another weakness is that is it not known whether it is meaningful to compare drugs in terms of their ability to engender place preferences. That is, if drug X induces a greater place preference than drug Y, does it have more abuse potential? Finally, because the procedure involves the simple behavioral response of moving from one chamber to another, it is not known whether it can be used to study some of the complex behavioral variables that are known to be determinants of drug self-administration. Despite these ambiguities, however, the simplicity of the procedure makes it likely that it will continue to be useful for studying drug abuse.

(See also: Abuse Liability of Drugs ; Reinforcement )

BIBLIOGRAPHY

Bozarth, M. A. (1987). Conditioned place preference: A parametric analysis using systemic heroin injections. In Methods of assessing the reinforcing properties of abused drugs, pp. 241-273. New York. Springer-Verlag.

Hoffman, D. C. (1989). The use of place conditioning in studying the neuropharmacology of drug reinforcement. Brain Research Bulletin, 23, 373-387.

William Woolverton

Conditioned Withdrawal

Upon cessation from drug taking, many individuals experience unpleasant effects (i.e., Withdrawal), which can include both physiological and psychological symptoms. For example, for Opiate drugs such as Morphine and Heroin, withdrawal symptoms can include restlessness, anorexia, gooseflesh, irritability, nausea, and vomiting. Withdrawal symptoms are most pronounced following a long history of exposure to Alcohol and opiates, but a variety of withdrawal symptoms can occur after exposure to most psychoactive drugs.

As with most other drug effects, researchers have shown that these withdrawal symptoms can be conditioned or linked by learning to environmental cues. This research on conditioned withdrawal has included both human case reports and laboratory animal research. For example, Vaillant (1969) reported that individuals who had been abstinent from opiates for months would experience "acute craving and withdrawal symptoms" upon reexposure to situations previously associated with opiate use. Further, Goldberg and Schuster (1967) showed that withdrawal symptoms also can be conditioned in laboratory animals. In their experiment, rhesus monkeys were addicted to morphine by giving them the drug repeatedly. The monkeys were then given an occasional injection of nalorphine, an opiate antagonist, which immediately led to the monkeys exhibiting signs characteristic of withdrawal. The injection of nalorphine was always given in the presence of a specific environmental stimulus, in this case a tone. Following several exposures to the tone paired with nalorphine, Goldberg and Schuster found that presentation of the tone itself was sufficient to produce the withdrawal signs.

The behavioral mechanism most likely to account for the phenomenon of conditioned withdrawal is classical conditioning (also known as Pavlovian ). In Pavlov's classic experiments on this type of conditioning, a neutral stimulus such as a bell, is repeatedly paired with a nonneutral stimulus such as food. Eventually the bell itself elicited salivation, which was initially observed only to the food. In conditioned withdrawal, a neutral stimulus (e.g., a bell, a needle, a room, a friend, a street corner, or certain smells) is paired with the non-neutral stimulus of withdrawal until eventually those neutral stimuli will also elicit withdrawal symptoms.

The phenomenon of conditioned withdrawal can have important implications for drug-abuse treatment. The experience of drug withdrawal is often an important factor in the long-term maintenance of drug abuse. That is, as individuals experience withdrawal, they are likely to seek out a new drug supply to relieve withdrawal symptoms. An important aspect of drug-abuse treatment is relieving the symptoms of withdrawal during the period immediately following the cessation of drug use. Conditioned effects, however, are often long-lasting and do not depend on the continued presentation of the initial nonneutral stimulus (in this case withdrawal). Even after a patient has been withdrawn from a drug, stimuli that have been conditioned to elicit withdrawal symptoms may still be effective. Therefore, upon reexposure to those stimuli a patient may be much more likely to relapse to drug abuse. Thus, to be successful, any treatment regimen for drug abuse must deal with conditioned withdrawal.

(See also: Causes of Substance Abuse ; Wekler's Pharmacologic Theory of Drug Addiction )

BIBLIOGRAPHY

Goldberg, S. R., & Schuster, C. R. (1967). Conditioned suppression by a stimulus associated with nalorphine in morphine-dependent monkeys. Journal of the Experimental Analysis of Behavior, 10, 235-242.

Vaillant, G. E. (1969). The natural history of urban narcotic drug addiction—Some determinants. In H. Steinburg (Ed.), Scientific basis of drug dependence. New York: Grune & Stratton.

Charles Schindler

Steven Goldberg

Drug Discrimination Studies

When a person takes a drug of abuse, it has effects that a person feels and can describe. These are termed subjective effects and they play an important role in drug abuse. People are more likely to abuse a drug that has effects they describe as pleasant than one they describe as unpleasant.

The subjective effects of drugs of abuse have been studied in humans for many years and in several different ways. Early research involved administering drugs, usually morphine-like drugs, to former Heroin addicts—who then answered questionnaires that were designed to detect and classify the subjective effects of the drug. The single-dose Opiate questionnaire asks subjects whether they can feel the drug, to identify the drug, to describe the symptoms, and to rate how much they like it. The Addiction Research Center Inventory consists of a series of true/false statements that describe internal states that might be produced by drugs. The Profile of Mood States is a list of adjectives that can be used to describe mood. Responses to these questionnaires depend on variables such as type of drug and drug dose. Recent research has examined the subjective effects of a wider variety of drugs (including Stimulants and Depressants) in both experienced and inexperienced subjects. The purpose of this research is to understand the factors that can influence a person's subjective response to drugs of abuse.

Since subjective effects require a verbal description of an internal state, they can only be studied directly in humans. Since the 1960s, however, it has become clear that animals can be trained to respond in a way that suggests they can detect the internal state produced by a drug. The behavioral paradigm is called Drug Discrimination, and the drug effect is called a discriminative stimulus effect. Although a number of drug-discrimination paradigms have been developed, the most common is a two-lever paradigm. Here the animal is trained to press one lever after it has received a drug injection and the second lever after an injection of the drug vehicle or, in some cases, another drug. Responding on the lever that is appropriate to the injection is reinforced, usually, by a food pellet; responding on the incorrect lever is not reinforced. If this is done repeatedly over a period of several weeks, the animal learns to respond almost exclusively on the lever associated with the injection.

Although it is difficult to know what an animal feels, it seems as if the animal is telling us whether it feels the drug or not by the lever it presses. The animal can then be asked to "tell" us whether a new drug "feels" like the training drug. It will respond on the drug lever if it does and on the vehicle lever if it does not. It can also be "asked" whether a drug blocks the effects of the training drug. If the test drug does block the effect of the training drug, the animal will respond on the vehicle lever when given both drugs.

CONCLUSIONS

What makes this area of research so exciting are the striking similarities between the classification of drugs by humans, based on their subjective effects, to those by animals, based on their discriminative stimulus effects. Therefore, this animal model can be used to investigate the influence of factors such as genetics, drug history, and behavioral history—factors that cannot be easily controlled in human subjects—on the subjective effects of drugs. It also allows us to predict whether a new drug is likely to have subjective effects, like a known drug of abuse, or is likely to block the subjective effects of the drug of abuse, without giving the drug to humans. If an animal is trained to discriminate a drug of abuse and presses the drug lever when given the new drug, then it is highly likely that the new drug will have subjective effects in humans similar to those of the drug of abuse. Its availability might then be restricted. If the animal responds on the vehicle lever when given the combination of the new drug and the drug of abuse, the new drug may block the subjective effects of the drug of abuse. Such a drug might then be useful for treating drug abuse.

(See also: Abuse Liability of Drugs ; Drug Types ; Sensation and Perception )

BIBLIOGRAPHY

Colpaert, F. C. (1986). Drug discrimination: Behavioral, pharmacological, and molecular mechanisms of discriminative drug effects. In Behavioral analysis of drug dependence. Orlando, FL: Academic.

Colpaert, F. C., & Balster, R. L. (1988). Transduction mechanisms of drug stimuli. Berlin: Springer-Verlag.

William Woolverton

Drug Self-Administration

One factor that distinguishes a drug of abuse from a drug that is not abused is that taking the drug of abuse increases the likelihood that it will be taken again. In such a case, we say that this drug has reinforced the drug self-administration response and that it has reinforcing effects. Factors that influence reinforcing effects, therefore, profoundly influence drug self-administration and drug abuse. Knowing the reinforcing effects of drugs is essential to understanding drug abuse.

Techniques developed on laboratory animals allow us to study the reinforcing effects of drugs, using the intravenous and oral routes as well as smoking. To study intravenous self-administration, the researcher surgically implants a chronic intravenous line (a catheter) through which a drug can be administered. Laboratory animals (rats, mice, monkeys, and so on) live in cages in which they can operate some device, usually a lever press, that turns on an electric pump to send some drug solution through the catheter. Oral self-administration is harder to establish, since drugs are usually bitter; however, by arranging conditions so that large amounts of drug solution are ingested in relatively short periods—usually by adding the drug to water when food is available—researchers can condition animals to self-administer drugs orally. Research on the smoking of Tobacco or Crack-Cocaine is important and this too needs conditioning for reliable study.

Animals used in research studies have shown that, with few exceptions, they abuse the same drugs that humans abuse and show similar patterns of intake. (Exceptions include Marijuana and Hallucinogens, such as LSD, which animals do not seem to find reinforcing.) Drug self-administration studies have been used to predict whether a new drug is likely to be abused by humans if it becomes easily available. More important, such research has allowed us to understand some factors that can increase or decrease the reinforcing effects of drugs that contribute to human drug abuse. Some of these factors relate to the drug itself; others to the environment. For example, drugs that increase the concentration of the Neurotransmitter Dopamine in the synapses of the brain (e.g., cocaine) are more likely to have abuse potential than those that do not.

Research has made it clear that even the most preferred drug—cocaine—will be self-administered differently depending on environmental conditions. If more lever presses are required to obtain it (it "costs" more), less is consumed. Drug self-administration can also be decreased by punishment or by making valuable alternatives available. In short, drug self-administration research has shown that whether a drug will be abused is determined by a complex interaction between the drug, the environment, and the organism. Current research is aimed at understanding the dynamics of that interaction in a quantitative way.

(See also: Abuse Liability of Drugs ; Adjunctive Drug Testing )

William Woolverton

Environmental Influences on Drug Effects

More than any other discipline, the field of behavioral Pharmacology has attempted to understand the influence of nonpharmacological, or environmental, factors on the effects of abused drugs. Since the classic demonstration by Dews (1955, 1958) showing that the effects of pentobarbital and Methamphetamine depend on the manner in which behavior is controlled by the schedule of Reinforcement, researchers have been interested in various environmental influences on the effects of drugs. Some of these effects are described elsewhere in this encyclopedia (and see Barrett, 1987, for a more detailed review). This article reviews additional influences to illustrate the overwhelming conclusion that the effects of a drug depend on complex environmental variables that may override the typical pharmacological effects of a compound. Indeed, the evidence for environmental influences on drug action is so compelling that when the effects of abused drugs are characterized, "susceptible to environmental modulation" should be a salient distinguishing description along with physiological features.

BEHAVIORAL CONSEQUENCES

The specific manner in which behavior is controlled by its consequences may often represent a strong influence on drug action. In research situations, this is apparent in the effects of Amphetamine or Cocaine on punished and nonpunished responses maintained by the presentation of food. Low rates of nonpunished responses are typically increased by these drugs (psychomotor stimulants), whereas comparable low rates on punished responses are not affected by these drugs or are only decreased further. In the Dews studies (1955, 1958), the effects of the drugs differed depending on whether behavior was maintained at relatively high response rates under a fixed-ratio schedule that provided food following every n th response or whether responses occurred at lower rates under a fixed-interval schedule that provided food for the first response after t minutes. Explanations of the differential effects of the drugs could not be based on different levels of motivation, since these schedule conditions alternated sequentially within the same experimental session. Although these and similar results were obtained under carefully controlled experimental conditions, such findings document the essential point that environmental conditions surrounding and/or supporting behavior play a very important role in determining the effects of drugs.

BEHAVIORAL CONTEXT

The environmental modulation of drug effects has been shown repeatedly, by using schedule-controlled responses and various types of events. These findings represent two areas of research demonstrating how drug effects are modified directly by existing environmental conditions:

1. More remote influences can also influence drug action. In behavioral history, for example, consequences that have occurred in the distant past can significantly alter the effects of abused drugs even though no traces of that experience are apparent in current behavior.
2. In other studies in which environmental influences helped determine the effects of an abused drug, behavioral consequences occurring under one experimental condition alter the action of drugs occurring under different conditions. In this case, the conditions that interact are relatively close in time. For example, in an experiment with monkeys, exposure to a procedure in which responses avoided the delivery of a mild electric shock completely reversed the effects of amphetamine on punished responses that had occurred in a different and adjacent context (i.e., under different stimulus conditions from the avoidance schedule and separated by only a few minutes).

Comparable results, although with different species, different schedule conditions, and different environmental events, have also been arrived at with Alcohol, cocaine, and Chlordiazepoxide (Barrett, 1987). The findings show the generality of this phenomenon—that the environment is an important variable contributing to the effects of drugs on behavior. The actions of a drug at its receptor site and the transduction mechanisms that ensue can be affected by events occurring in the environment.

SUMMARY

The studies described here indicate the powerful influences that exist in the environment that can alter the course of the effects of abused drugs. Such findings illustrate the need to examine those influences and the manner in which they occur, although it is often tempting to attribute all changes in behavior to the abused drug. Consequences that are immediate, as in the existing environment, or remote, such as in the individual's past experience, help determine the acute effects of drugs and may also contribute to long-term abuse and persistent drug use.

(See also: Adjunctive Drug Taking ; Causes of Substance Abuse ; Reward Pathways and Drugs ; Tolerance and Physical Dependence )

BIBLIOGRAPHY

Barrett, J. E. (1987). Nonpharmacological factors determining the behavioral effects of drugs. In H. Y. Meltzer (Ed.), Psychopharmacology: The third generation of progress. New York: Raven Press.

Dews, P. B. (1958). Studies on behavior: IV. Stimulant actions of methamphetamine. Journal of Pharmacology and Experimental Therapeutics, 122, 137-147.

Dews, P. B. (1955). Studies on behavior: I. Differential sensitivity to pentobarbital of pecking performance in pigeons depending on the schedule of reward. Journal of Pharmacology and Experimental Therapeutics, 113, 393-401.

James E. Barrett

Intracranial Self-Stimulation (ICSS)

The intracranial self-stimulation (ICSS) procedure is used to study the effects of drugs on reward processes, or regions involved in pleasurable feelings, in the brain. In humans undergoing brain surgery, researchers were able to induce limb movements or produce sensations by electrically stimulating various regions of the cortex. Similarly, electrical stimulation of certain brain regions in the rat was reinforcing, or pleasurable, thus creating a new area for brain research. An electrode capable of delivering varying intensities and durations of electrical impulses was implanted in the brain of a rat. These animals could be trained to press a lever that would activate the implanted electrode, sending a small impulse to a specific brain region. In addition, animals could also be trained to press a lever that would "shut off" brain impulses in other regions. These animals will give up food and water, and even sexual activities, in order to perform tasks that lead to brain stimulation in certain regions. Based on these results, this procedure was recognized as a method by which mechanisms underlying drug addiction could be studied.

Early work in brain stimulation involved mapping out which brain areas would support self-stimulation in animals, primarily rats. Animals were trained using operant procedures in which a press of the lever would deliver an electrical stimulus to the brain. Researchers found two systems of reward in the rat brain using ICSS: a dorsal (closer to the back of the animal) system projecting from the caudate/septal area through the dorsal thalamus to the tectum, and a ventral system (closer to the abdomen of the animal), the medial forebrain bundle. The "punishment" system seemed to be located in the diencephalon and the tegmentum. Rats will readily self-stimulate when electrodes are implanted into the ventral tegmental area (VTA) and substantia nigra, brain regions associated with reward. Researchers hypothesized that, by stimulating these brain regions, the rats were activating their own dopamine neurons electrically, thus producing the effects of reward. Dopamine is a neurotransmitter found in the brain of rodents and primates. This neurotransmitter is thought to be involved in the rewarding or pleasurable effects of drugs of abuse.

Drugs can interact with the established pattern of self-stimulation in an animal. Interactions between drugs and ICSS suggest that these treatments act through the same mechanisms. The rate at which the animal presses the lever is correlated with the intensity of the current being delivered to the brain. However, the rate at which the animal presses the lever is not necessarily related to the amount of pleasure the animal is experiencing. The influences of various drugs on self-stimulation behavior can be due to a variety of effects, such as increases or decreases in general activity, changes in motivation or memory, etc. To state that a drug has an effect on self-stimulation, these possibilities must be ruled out. To do this, one can compare data describing the effects of the test drug in other behavioral paradigms (e.g., locomotor activity, self-administration) to the effects observed in ICSS.

Despite these limitations, researchers have collected interesting data, examining the effects of various drugs of abuse on rate of self-stimulation. Animals were trained to press a lever that would result in electrical stimulation of the brain. Then, the intensity of the stimulation was lowered so that the animals would not press the lever very often. When the animals were given the psychomotor stimulant amphetamine, the animals began to press the lever at a very high rate that gradually declined to the rate observed at low stimulation intensities. To rule out that animals might be pressing the lever more often due to the motor-activating effects of amphetamine, these researchers looked at the effects of amphetamine on lever-pressing in rats that were not receiving any brain stimulation. They saw no changes in lever-pressing before or after the rats were given amphetamine. Thus, they concluded that amphetamine enhances the reward produced by the subthreshold stimulation by activating reward pathways in the brain.

Another approach in using ICSS to measure the rewarding effects of drugs is to train animals to regulate the intensity of the stimulation that they receive in the brain. Animals are given access to two levers in the test chamber. When the animal pressed one of these levers for the first time, a relatively high level of brain stimulation was delivered. However, subsequent presses of the lever deliver decreasing levels of stimulation. The animal can "reset" the stimulation to the original high level by pressing the second lever. Under these conditions, the animals reliably reset their stimulation level once it drops below a certain point. From this measurement, researchers are able to determine each animal's reward threshold in a very reliable way. Regardless of the initial level of stimulation, these animals would press the reset lever at the same intensity of stimulation. Drugs such as amphetamine and morphine have "threshold-lowering" effects, such that the animals would press the reset lever at a lower intensity after receiving these drugs. This suggests that these drugs are themselves reinforcing, or pleasurable.

ICSS has been used to study the effects of the chronic administration of cocaine. Depending on the frequency of administration and amount of cocaine given, difference changes in ICSS responses have been observed. When low doses of cocaine were given once or several times a day, no changes in the ICSS threshold were observed. However, when higher doses of cocaine were administered for seven days, the reward threshold was increased in these animals, indicating that tolerance to the rewarding effect of cocaine had developed and/or that the effects of cocaine had become less pleasurable. In addition, animals that self-administered cocaine also exhibited this increase in the ICSS reward threshold. These experimental results are comparable with those observed in human drug users who take increasingly greater amounts of drug to achieve the same pleasurable effect over a long period of time.

BIBLIOGRAPHY

Greenshaw, A., & Wishart, T. (1987). Effects of drugs on reward processes. In A. Greenshaw & C. Dourish, (Eds.), Experimental psychopharmacology: Concepts and methods. Clifton, NJ: Humana Press.

Hammer, R., Egilmez, Y., & Emmett-Oglesby, M. (1997). Neural mechanisms of tolerance to the effects of cocaine. Behavioral Brain Research, 84, 225-239.

Silverman, P. (1978). Animal behaviour in the laboratory. New York: Pica Press.

Heather L. Kimmel

Learning, Conditioning, and Drug Effects—An Overview

The effects of abused drugs can be examined at many levels, ranging from the molecular to the cellular to the behavioral. Each of these research areas contributes significant information to understanding the mechanisms by which drugs of abuse and alcohol produce their diverse effects. The most tangible sign of both immediate and long-term actions of abused drugs is their effects on behavior. Often it is incorrectly assumed that behavior is a passive reflection of more significant events occurring at a different and (usually) more molecular level. Understanding those cellular events is occasionally viewed as the key to understanding drug abuse and to intervention strategies. In fact, however, behavior itself and the variables that control it play a prominent and often profound role in directly determining drug action and, most likely, those cellular and molecular events that participate in behavior and in the effects of drugs. The variables that guide and influence behavior also affect molecular substructures—therefore, behavioral and neurobiological processes are interdependent.

EXPERIMENTAL ANALYSIS OF BEHAVIOR AND DRUGS OF ABUSE

The progression of behavioral approaches in the study of the effects of abused drugs is characteristic of the cumulative and evolutionary nature of scientific progress. A number of techniques are now available that permit the development and maintenance of a variety of behaviors that are remarkably stable over time, sensitive to a number of interventions, and reproducible within and across species. These procedures have evolved over the past several years and reflect the combined efforts of individuals in different disciplines ranging from psychology, pharmacology, physiology, and ethology. For the most part, research studying the effects of abused drugs on behavior has been conducted by two basic procedures. One procedure uses unconditioned behavior, such as locomotor activity that is more spontaneous in its occurrence (but still influenced by environmental conditions) and requires no specific training before it can be studied. Many Psychomotor Stimulants such as Cocaine and Amphetamine, for example, produce large and consistent increases in locomotor activity in laboratory animals. Frequently, however, unconditioned behavior is produced or elicited by the presentation of specific stimuli, and it is then brought under experimental control by arranging for the production of a response to a stimulus other than that originally responsible for its occurrence. The Russian physiologist Ivan Pavlov, for example, performed extensive studies in 1927, in which he used the unconditioned salivary response to food and to conditioned stimuli paired with food to study processes of classical or respondent conditioning. Although this approach has been used somewhat less often than other techniques, respondent conditioning procedures still serves as a very useful method for studying drug action (Barrett & Vanover, 1993).

The second procedure, which is designated as operant conditioning, uses the methods and techniques developed by the pioneering American psychologist B. F. Skinner (1938) to investigate behavior controlled by its consequences. The body of experimental research using operant conditioning techniques to study the effects of abused drugs is extensive (see Iversen & Lattal, 1991, for general reviews of the techniques and applications).

Unconditioned and Conditioned Respondent Behavior.

Respondent behavior is elicited by specific stimuli and usually involves no specific training or conditioning, since the responses studied are typically part of the behavioral repertoire of the species and are expressed under suitable environmental conditions. Although the factors responsible for the occurrence of these behaviors presumably lie in the organism's distant evolutionary past, certain unconditioned responses can be brought under more direct and immediate experimental control through the use of procedures first discovered and systematically explored by Pavlov. These procedures consist of expanding the range of stimuli capable of producing an elicited response. In respondent conditioning, previously noneffective stimuli acquire the ability to produce or elicit a response by virtue of their temporal association with an unconditional stimulus, such as food, which is capable of eliciting a response without prior conditioning. Thus, when a distinctive noise, such as a tone, is repeatedly presented at the same time that or shortly before food is given, the tone acquires the ability to elicit many of the same responses originally limited to food.

Respondent behaviors depend primarily on antecedent events that elicit specific responses. Typically, these responses do not undergo progressive differentiation, that is, the responses to either a conditioned or an unconditioned stimulus are generally quite similar. These procedures also do not establish new responses but simply expand the range of stimuli to which that response occurs.

Operant Behavior.

In contrast to respondent behavior, operant behavior is controlled by consequent events, that is, it is established, maintained, and further modified by its consequences. Operant behavior occurs for reasons that are not always known. The responses may have some initially low probability of occurrence or they may never have occurred previously. Novel or new responses are typically established by the technique of "shaping," in which a behavior resembling or approximating some final desired form or characteristic is selected, increased in frequency and then further differentiated by the provision of a suitable consequence, such as food presentation to a food-deprived organism. This technique embodies the principle of reinforcement and has been widely used to develop operant responses such as lever pressing by rodents, humans, and nonhuman primates. Behavior that has evolved under such contingencies may bear little or no resemblance to its original form and can perhaps only be understood by careful examination of the organism's history. Although some behaviors often appear unique or novel, it is likely that the final product emerged as a continuous process directly and sequentially related to earlier conditions. The manner in which operant responses have been developed and maintained, as well as further modified, has been the subject of extensive study in the behavioral pharmacology of abused drugs and has had a tremendous impact on this field. Many of the potent variables that influence behavior, such as reinforcement, punishment, and precise schedules under which these events occur, also are of critical import in determining how a drug will affect behavior.

Respondent Versus Operant Behavior.

Although it is possible to tell operant behavior from respondent behavior in a number of ways, these processes occur concurrently and blend almost indistinguishably. For example, the administration of a drug may elicit certain behavioral and physiological responses such as increased heart rate and changes in perception that are respondent in nature; stimuli associated with the administration of that drug may also acquire some of the same ability to elicit those responses. If the administration of the drug followed a response and if the subsequent frequency of that response increased, then the drug also could be designated a reinforcer of the operant response. Thus, these important behavioral processes frequently occur simultaneously and must be considered carefully in experimental research, and also in attempting to understand the control of behavior by abused drugs. The primary distinctions between operant and respondent behavior now appear to be the way these behaviors are produced and the possible differential susceptibility to modification by consequent events. Respondent behavior is produced by the presentation of eliciting stimuli; characteristic features of these behaviors are rather easily changed by altering the features of the eliciting stimulus such as its intensity, duration, or frequency of presentation. Under all of these conditions, however, the response remains essentially the same.

In contrast, operant behavior depends to a large extent on its consequences, and with this process, complex behavior can develop from quite simple relationships. One has only to view current behavior as an instance of the organism's previous history acting together with more immediate environmental consequences to gain some appreciation for the continuity and modification of behavior in time. Current behavior is often exceedingly difficult to understand because of the many prior influences or consequences that no longer operate but which may leave residual effects. The effects of a particular consequence or intervention can be quite different depending on the behavior that exists at the time the event occurs. An individual's prior history, then, is important not only because it has shaped present behavior but also because it will undoubtedly determine the specific ways in which the individual responds to the current environment. Accordingly, prior behavioral experience can have a marked effect in determining how a drug will change behavior.

BEHAVIORAL METHODOLOGY AND THE EVALUATION OF ABUSED DRUGS

Experiments with drugs and behavior were initiated in Pavlov's laboratory in Russia during the time that Pavlov was studying the development of conditioned respondent procedures (see Laties, 1979, for a review of this early work). Early experiments with the effects of drugs on operant behavior were initiated shortly after Skinner began his pioneering work (Skinner & Heron, 1937). More intensive studies using drugs and operant-conditioning techniques were not conducted, however, until effective drugs for the treatment of various psychiatric disorders such as Schizophrenia were introduced in the 1950s. These discoveries prompted the development and extension of behavioral techniques to study these drugs, and many of the procedures were subsequently used in the study of abused drugs. From these combined efforts, several key principles evolved that have served as the foundation for understanding and evaluating the effects of abused drugs.

Environmental Events.

As already discussed, behavior can be controlled by a wide range of environmental events. One question that arose early in the study of the behavioral effects of drugs was whether the type of environmental event that controlled behavior contributed to the effects of a drug—that is, whether a behavior controlled by a positive event, such as food presentation, would be affected in the same manner as a behavior controlled by a more negative event, such as escape from an unpleasant noise or bright light. Although seemingly straightforward, the issue is not easily addressed because other known factors contribute to the actions of drugs, such as the rate at which a behavior controlled by the event occurs. If rates are not similar, any comparison between drug effects on behavior controlled by those different events might be spurious. Indeed, when such comparisons have been conducted in nonhuman primates under carefully controlled conditions, it has been shown that the type of environmental event controlling behavior can play an important role in determining the qualitative effects of a drug on behavior (Barrett & Witkin, 1986; Nader, Tatham, & Barrett, 1992). For example, when the effects of certain drugs such as Alcohol or Morphine were studied by using behavioral responses of monkeys who were similarly maintained by a food stimulus or a mild electric-shock stimulus, the drugs produced different effects depending on the maintaining event (Barrett & Katz, 1981). These findings suggest that the manner in which behavior is controlled by its environmental consequences—that is, the characteristics of the environment—can be of considerable importance in determining how an individual will be affected by a particular drug. This was one of the experiments that supported the view that a drug is not a static molecule with uniform effects, but rather that the way the substance interacts with its receptor and initiates the cascade of biochemical processes depends very much on the dynamic interaction of behavior within its environment. When the issue is viewed in this light, it is clear that environmental events and the way they impinge on behavior contribute substantially to the specific effects of a drug and its impact on the individual organism.

Examples of similar types of environmental control over pharmacological effects of drugs also come from studies that employed respondent conditioning procedures to demonstrate that stimuli paired with morphine or heroin injections can influence the development and manifestation of fundamental pharmacological processes such as tolerance and lethality (Siegel, 1983). These studies add to the rather convincing body of evidence that environmental conditions accompanying the administration and effects of the drug can be of considerable importance in determining the effects of that drug when it is administered, as well as when it is subsequently administered.

Behavioral and Pharmacological History.

In addition to pointing to the contribution of the immediate environment in determining the effects of abused drugs, a number of studies demonstrated that the consequences of past behavior could also contribute significantly to the effects of drugs, often by resulting in an action that is completely opposite to that shown in organisms without that history. These findings convey the complexity involved in understanding the effects of drugs of abuse, and the difficulties in attempting to understand their actions in humans with more complex life histories than those of experimental animals. In addition, related studies showed that prior experience with one drug could also directly affect the manner in which behavior is influenced by other drugs.

Early studies using different training conditions to develop a visual discrimination in pigeons demonstrated that an antipsychotic drug, Thorazine (chlorpromazine), and an antidepressant drug, imipramine, had different effects on that discriminative behavior, depending on how the training occurred (Terrace, 1963). Similarly, studies that used exploratory behavior of rats in mazes demonstrated that the effects of a mixture of amphetamine (Stimulant) and a Barbiturate drug (Depressant) depended on whether the rats had been previously exposed to the maze (Steinberg, Rushton, & Tinson, 1961). More recently, studies with squirrel monkeys showed that prior behavioral experience can influence the effects of a wide range of drugs, including morphine, cocaine, and amphetamine, as well as alcohol, under a variety of experimental conditions (summarized by Barrett, Glowa, & Nader, 1989; Nader et al., 1992). In one study, for example, the effects of amphetamine were studied on behavior reinforced by food that was also suppressed by punishment. Under these conditions, amphetamine produced a further decrease in punished responding. If those same monkeys, however, were then exposed to a procedure in which responding postponed or avoided punishing shock and were then returned to the punishment condition, amphetamine no longer decreased responding; instead, it produced large increases in suppressed responding. Thus, the effects of amphetamine in this study depended on the prior behavioral experience of the animal.

These findings raise a number of issues surrounding the etiology of drug abuse as well as issues pertaining to an individual's risk for or vulnerability to abusing particular drugs. If, as seems likely, certain drugs are abused because of their effects on behavior, and those behavioral effects are related to past history, then the historical variables become exceptionally important in eventually understanding and treating, as well as preventing, drug abuse. Perhaps previous behavioral experience generates conditions under which a drug may have quite powerful actions on behavior and on the subjective effects that drug produces; by virtue of their previous history, the susceptible individuals may be predisposed to drug abuse. If these arguments are valid, it should be possible, after achieving a better understanding of the factors, to develop behavioral strategies for "inoculating" or "immunizing" individuals against particular drug effects. Although such possibilities may seem remote at this time, it is very clear that behavioral variables can direct the effects of abused drugs in striking and significant ways.

SUMMARY

Although drugs of abuse have a reliable and predictable spectrum of effects under a broad range of conditions, the implications from studies are that many of the more characteristic effects of abused drugs can be altered by the organism's history and by the environmental conditions under which the drug is and has been administered. As Folk (1983) said so eloquently, "Pharmacological structure does not imply motivational destiny"; the reasons for the effects of an abused drug depend on more than the static molecular properties of that drug. Both past and present environmental factors can play an overwhelming role in determining the behavioral effects of abused drugs, and they may indeed be a major source of the momentum behind the continued use and abuse of those substances.

(See also: Abuse Liability of Drugs ; Addiction: Concepts and Definitions ; Adjunctive Drug Taking ; Causes of Substance Abuse ; Reinforcement ; Vulnerability as Cause of Substance Abuse )

BIBLIOGRAPHY

Barrett, J. E., Glowa, J. R., & Nader, M. A. (1989). Behavioral and pharmacological history as determinants of tolerance- and sensitization-like phenomena in drug action. In M. S. Emmett-Oglesby & A. J. Goudie (Eds.), Tolerance and sensitization to psychoactive agents: An interdisciplinary approach. Clifton, NJ: Humana Press.

Barrett, J. E., & Katz, J. L. (1981). Drug effects on behaviors maintained by different events. In T. Thompson, P. B. Dews, & W. A. McKim (Eds.), Advances in behavioral pharmacology (Vol. 3). New York: Academic Press.

Barrett, J. E., & Vanover, K. E. (1993). 5-HT receptors as targets for the development of novel anxiolytic drugs: Models, mechanisms and future directions. Psychopharmacology, 112, 1-12.

Barrett, J. E., & Witkin, J. M. (1986). The role of behavioral and pharmacological history in determining the effects of abused drugs. In S. R. Goldberg & I. P. Stolerman (Eds.), Behavioral analysis of drug dependence. New York: Academic Press.

Falk, J. L. (1983). Drug dependence: Myth or motive? Pharmacology Biochemistry and Behavior, 19, 385-391.

Iversen, I. H., & Lattal, K. A. (1991). Experimental analysis of behavior (Parts 1 and 2). New York: Elsevier.

Laties, V. G. (1979). I. V. Zavodskii and the beginnings of behavioral pharmacology: An historical note and translation. Journal of the Experimental Analysis of Behavior, 32, 463-472.

Nader, M. A., Tatham, T. A., & Barrett, J. E. (1992). Behavioral and pharmacological determinants of drug abuse. Annals of the New York Academy of Sciences, 654, 368-385.

Pavlov, I. (1927). Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex. London: Oxford University Press.

Siegel, S. (1983). Classical conditioning, drug tolerance, and drug dependence. In Y. Israel et al. (Eds.), Research advances in alcohol and drug problems (Vol. 7). New York: Plenum.

Skinner, B. F. (1938). Behavior of organisms. New York: Appleton-Century-Crofts.

Skinner, B. F., & Heron, W. T. (1937). Effects of caffeine and benzedrine upon conditioning and extinction. Psychological Record, 1, 340-346.

Steinberg, H., Rushton, R., & Tinson, C. (1961). Modification of the effects of an amphetamine-barbiturate mixture by the past experience of rats. Nature, 192, 533-535.

Terrace, H. S. (1963). Errorless discrimination learning in the pigeon: Effects of chlorpromazine and imipramine. Science, 140, 318-319.

James E. Barrett

Learning Modifies Drug Effects

A general framework within which to understand the basic processes and principles of respondent conditioning (as first discovered in the 1920s by Russian physiologist Ivan Pavlov [1849-1936] and subsequently elaborated in many laboratories over the next six decades) is described elsewhere. Here, specific examples of the role of conditioned drug effects are provided in an effort to more fully develop the point that conditioned or learned responses come about as a reaction to stimuli that have been associated with drug injections. These stimuli can play a powerful role in governing subsequent behavior in the absence of the drug.

CONDITIONED EFFECTS OF DRUGS

In addition to studies described previously showing that tolerance to the effects of a drug, as well as lethality, can depend on respondent-conditioning phenomena, a number of additional studies have demonstrated the conditioning of Withdrawal and other responses that are typically associated only with the presentation or removal of the drug. For example, by pairing a tone stimulus with the administration of nalorphine, an Opioid antagonist that precipitates withdrawal signs or the abstinence syndrome (agitation, excessive salivation, and emesis) in morphine-dependent subjects, it was possible to show in rhesus monkeys that the tone acquired the ability to elicit withdrawal responses when presented in the absence of natorphine (Goldberg & Schuster, 1967; 1970). Striking illustrations of similar conditioned withdrawal responses in Heroin addicts, as well as Craving, in which environmental stimuli trigger the disposition to self-administer the drug, also have been described. These behavioral responses to stimuli that have been previously associated with drug withdrawal or administration often occur after a prolonged period of time spent without drugs (O'Brien, 1976).

In some cases, drugs also acquire stimulus control over behavior in a procedure known as state-dependent learning. This procedure is different in some ways from that used to study drugs as discriminative stimuli. State-dependent learning refers to the finding that subjects exposed to a particular procedure when injected with a drug often are impaired upon reexposure to that condition if the drug is not present. Thus, the drug can be viewed as part of the original context in which a response was learned. One concern that stems from the finding that behavior learned during a drug-related condition is impaired in the absence of the drug is that of the potentially enduring problems related to frequent abuse of drugs during adolescence—a period often associated with major developmental and cognitive growth.

REINFORCING EFFECTS OF DRUG-PAIRED STIMULI

Thus far, the focus has been on the effects of environmental stimuli paired with the administration of a drug rather than on stimuli paired with a drug as a reinforcer. As has been frequently demonstrated, and as is true of many stimuli, drugs can have multiple functions. These include discriminative effects, which set the occasion for certain responses to occur, and they also include reinforcing effects, whereby a response is increased in probability when a reinforcing drug follows the occurrence of that response. Drug self-administration techniques have been very informative and useful in the study of the effects of abused drugs.

One additional experimental procedure that has been used in this field of research is that of repeatedly pairing a rather brief visual or auditory stimulus (e.g., a light or a tone, respectively) with the reinforcing administration of the drug and then using that stimulus also as a reinforcer to maintain behavior without drug administration. Perhaps the most compelling work in this area stems from a procedure in which a stimulus was presented according to a schedule to follow a particular response. On certain occasions, that stimulus also was associated with the administration of a drug—that is, the stimulus occurred at various times without the drug and then also just preceding the drug. Known technically as a "second-order schedule," this technique exerts powerful control over the occurrence and patterning of behavior, and it results in sustained responding for extended time periods in the absence of anything but the stimuli that have been paired with the administration of the drug itself (Katz & Goldberg, 1991). In other words, conditioned stimuli that have been paired with a drug can exert considerable control over behavior.

SUMMARY

To summarize, conditioned drug effects play an important role in the behavior stemming from drug abuse. Stimuli correlated with the administration of a drug, as well as behavior in the presence of that drug, frequently result in those stimuli gaining considerable control over the discriminative effects or reinforcing effects of that drug (or both). Perhaps this is one of the main reasons that drug effects are so compelling and problematic: Not only does the drug itself have powerful effects, but stimuli correlated with the drug also acquire the ability to produce similar effects.

(See also: Addiction: Concepts and Definitions ; Causes of Substance Abuse ; Memory and Drugs: State Dependent Learning ; Research )

BIBLIOGRAPHY

Goldberg, S. R., & Schuster, C. R. (1970). Conditioned nalorphine-induced abstinence changes: Persistence in post morphine-dependent monkeys. Journal of the Experimental Analysis of Behavior, 14, 33-46.

Goldberg, S. R., & Schuster, C. R. (1967). Conditioned suppression by a stimulus associated with nalorphine in morphine-dependent monkeys. Journal of the Experimental Analysis of Behavior, 10, 235-242.

Katz, J. L., & Goldberg, S. R. (1991). Second-order schedules of drug injection: Implications for under standing reinforcing effects of abused drugs. In N. K. Mello (Ed.), Advances in substance abuse, behavior and biological research (Vol. 4). London: Jessica Kingsley.

O'Brien, C. P. (1976). Experimental analysis of conditioning factors in human narcotic addiction. Pharmacological Review, 27, 533-543.

James E. Barrett

Operant Learning Is Affected by Drugs

According to psychologist B. F. Skinner, behavior that is rewarded or reinforced is more likely to occur again. The family dog soon learns that hanging around the kitchen table brings food. In this example, the food is a reinforcer because it increases the likelihood that the dog will spend time near the kitchen table. Thus, the dog's behavior "operates" on the environment to produce an effect. This process is called operant conditioning. The techniques of operant conditioning are used widely to establish new behaviors both in humans as well as in animals. Because behavior that is operantly conditioned is very sensitive and reliable, it is often used to examine drug effects.

A TYPICAL OPERANT CONDITIONING EXPERIMENT

In most operant conditioning experiments, an animal is placed in a special chamber which is called a Skinner box after the man that developed operant conditioning. A typical operant chamber, which is shown in Figure 1, has a response key or lever and a place for delivering food. The animal's responses are counted by a computer and also recorded on a roll of paper that shows the distribution of responses over time. Although the experimental chamber in Figure 1 is designed for animals, operant conditioning procedures are also used to examine drug effects in humans. In these studies, the person may sit in a chair and respond by moving a joystick or perhaps sit at a keyboard and respond to stimuli on a computer screen.

SCHEDULES OF FOOD DELIVERY

In most operant conditioning experiments in animals, responses on a lever or key produce food according to some schedule. Behavior maintained by a schedule of reinforcement is called schedule-controlled behavior. For example, the pigeon or rat may have to make a specific number of responses in order to receive food. When this occurs, the organism is responding under a fixed ratio schedule. A similar schedule is the variable ratio schedule in which reinforcement occurs after an unpredictable number of responses. With both the fixed ratio and the variable ratio schedules, animals respond very quickly, in fact, under a fixed ratio schedule that requires thirty responses for food delivery, pigeons may respond as fast as five times a second. Another operant schedule is the fixed interval schedule in which the first response that occurs after a specified period of time produces food. With this schedule, rates of responding increase as the time for food delivery approaches. For example, in a fixed interval five-minute schedule. responding is very low during the first two minutes of the interval; responding picks up speed during the third and fourth minutes of the interval and becomes very rapid during the last minute, just before the food is delivered.

By comparing drug effects on different schedules of Reinforcement, scientists have shown that the way in which a drug alters responding depends on the rate of responding produced by a given schedule of reinforcement as well as the amount (or dose) of drug given. Thus, a drug's effects are rate-dependent as well as dose-dependent. The rate-dependency theory of drug action was first proposed by Peter Dews in the early 1960s and is best exemplified by the effects of amphetamine. Dews noted that amphetamine alters responding differently under a schedule of reinforcement that produces low rates of responding than under a schedule of reinforcement that produces high rates of responding. Specifically, a small amount of Amphetamine increases very low rates of responding, whereas the same amount of amphetamine either decreases or does not change high rates of responding. Other drugs in the amphetamine class such as Cocaine and Methylphenidate (Ritalin) also alter responding in a rate-dependent manner.

One of the most interesting aspects of the rate-dependency theory of drug action is that it emphasizes the importance of behavioral as well as pharmacological factors in determining a drug's effect on behavior. Thus, the rate at which an animal responds is an important determinant of the way in which a drug alters behavior. It also helps to explain why drugs such as amphetamine and methylphenidate, which generally increase motor activity, might be useful in treating hyperactivity. Since hyperactive children respond at a very high rate, amphetamine would be expected to decrease these high rates of responding.

In contrast to the rate-dependent effects observed for amphetamine-like drugs. the most notable effect of drugs such as Morphine is that they decrease rates of responding under several different schedules of reinforcement. The extent to which morphine decreases rate of responding depends on how much morphine is given. Thus, its effects are dose-dependent. Moreover, like all schedule-controlled behavior, these decreases in rate of responding are very consistent and reliable. If a rat is trained to respond under a fixed ratio schedule of food presentation and then given morphine, morphine will decrease rates of responding by about the same amount each time it is given; however, if morphine is given daily for a week or more, its rate-decreasing effects diminish. In other words, Tolerance develops. Interestingly, the development of tolerance depends on the behavior examined as well as how much drug is given.

Morphine's effects on responding under schedules of reinforcement are also used as a baseline to investigate the biochemical and physiological events that occur when morphine is given. Opioid antagonists, which block the binding of morphine to opioid receptors, are able to reverse morphine's effects on schedule-controlled behavior. Since morphine's effects on responding are blocked when opioid receptors are blocked, these data suggest that morphine produces its behavioral effects by interacting with opioid receptors. Responding under schedules of reinforcement is also used to examine the biochemical and physiological effects of other drugs. For example, amphetamine's effects on schedules of reinforcement are altered by drugs that interfere with the neurotransmitter dopamine, suggesting that the dopamine system is involved in amphetamine's behavioral effects.

SCHEDULES OF PUNISHMENT

Although schedule-controlled behavior generally is maintained by the delivery of food, in some situations, responding is punished by the presentation of an unpleasant event. In a typical punishment procedure, responding is first maintained by a schedule of food delivery. Brief periods are then added during which responding is both reinforced by food and also punished by an unpleasant event. As a result, responding occurs at a lower rate during periods in which responding is punished than during unpunished periods. Figure 2 shows the design of a typical punishment procedure. First, note in the first panel that responding maintained by food alone occurs at a high rate. In the second panel, responding is punished by the addition of an unpleasant event and, as a result, rate of responding is decreased during the punishment period. The third panel shows that a drug such as alcohol selectively affects responding during the punishment period by restoring rates of responding to their baseline levels. Because these increases in punished responding occur following alcohol as well as a number of other antianxiety agents, but not following drugs such as morphine or amphetamine, increases in punished responding may reflect the antianxiety properties of these drugs. Indeed, the punishment procedure is used by a number of pharmaceutical companies to predict whether a drug might be useful in treating anxiety.

SCHEDULES OF REINFORCEMENT AS A WAY TO MEASURE LEARNING

Schedules of reinforcement are also used to examine the rate at which new behaviors are learned. Clearly, it takes some time to train an animal to respond under a schedule of reinforcement. This period of training is called the acquisition period and provides a measure of learning. One way to design a learning experiment is to measure how long it takes a group of rats to learn to respond under a schedule of reinforcement when a drug is given and compare that to how long it takes another group of rats to learn the same task without a drug. In experiments such as these, animals are usually trained to respond under very complicated schedules of reinforcement. Sometimes the animal has to complete the requirements of several different schedules in order to obtain food; in other procedures, the animal responds differently in the presence of different kinds of stimuli. In another procedure, the time it takes an animal to learn a pattern of responses is determined when a drug is given and compared to the time it takes the same animal to learn a different pattern of responses without a drug. Ethanol, the Barbiturates, and several antianxiety drugs all increase the number of errors animals make in learning new response sequences. Studies using a similar procedure in humans show that ethanol and certain antianxiety drugs also increase the number of errors people make when they learn new response sequences.

SUMMARY

Schedules of reinforcement offer several advantages for studying the behavioral effects of drugs. First, schedule-controlled responding is very consistent and remains unchanged for long periods of time. This consistency makes it easy to examine changes in behavior after a drug is given. Second, schedule-controlled behavior can be used with human subjects as well as with several different animal species, including mice, rats, pigeons, and monkeys. Finally, schedule-controlled behavior is recorded with automatic devices so that the experimenter is completely removed from the experiment and the nature of the behavior is easy to measure. From these studies, several important concepts have emerged. Scientists have shown that the behavioral effects of drugs depend not only on the amount of drug given, but they also depend on the nature of the behavior being examined. Both the rate of occurrence of a behavior as well as the presence of punishing stimuli are very important determinants of how drugs alter behavior.

The Psychomotor Stimulants increase responding under schedules of reinforcement when responding occurs at a low rate; when responding occurs at higher rates, the psychomotor stimulants decrease rates of responding. The most notable effect of morphine is that it decreases overall rates of responding. Alcohol and the antianxiety agents are unique in that they increase responding that is suppressed by the presentation of a punishing stimulus. Finally, several drugs interfere with the learning of complex patterns of responding.

(See also: Adjunctive Drug Taking ; Behavioral Tolerance ; Memory and Drugs: State Dependent Learning ; Memory, Effects of Drugs on ; Reinforcement ; Tolerance and Physical Dependence )

BIBLIOGRAPHY

Carlton, P. L. (1983). A primer of behavioral pharmacology. New York: W. H. Freeman.

Mc Kim, W. A. (1986). Drugs and behavior. Englewood Cliffs. NJ: Prentice-Hall.

Seiden, L. S., & Dykstra, L. A. (1977). Psychopharmacology: A biochemical and behavioral approach. New York: Van Nostrand Reinhold.

Linda A. Dykstra