Back to book TOC
|(Search book contents)||(Search entire ASNet)|
An Overview of Assessing Drug Reinforcement
Michael A. Bozarth
Center for Studies in Behavioral Neurobiology
Department of Psychology
Montreal, Quebec, Canada H3G 1M8
|The various methods used to study drug reinforcement are briefly examined, and some of the advantages and limitations of each procedure are discussed. Although some measures may seem preferable to others, each experimental technique has certain applications where it is most appropriate. The concordance of different experimental paradigms is also examined by summarizing the results of studies attempting to identify the neural substrates of psychomotor stimulant and opiate rewards. In general, there is substantial agreement among different experimental paradigms purporting to identify the brain mechanisms involved in these drug rewards. Finally, a protocol is suggested for the routine screening of new compounds for addiction liability. This protocol uses several experimental procedures, including both indirect and direct indices of drug reinforcement.|
As evidenced by the chapters in this book, there are a variety of ways to study the reinforcing properties of abused drugs. This chapter will present a brief overview of the major methods of assessing drug reinforcement, examine the concordance among several experimental techniques, and outline a protocol for screening new compounds for their addiction potential. The overviews of the different experimental methods are necessarily cursory and only a few exemplary considerations for using these methods are outlined. These topics are fully explored in the various chapters, and the reader is referred to this material for a detailed discussion of the experimental methods summarized here.
Definitions and Terms
Before proceeding to an overview of the techniques used to assess drug reinforcement and their relationships to drug addiction and drug abuse liability, it is necessary to briefly define some of the terms used in this chapter. The term drug addiction has suffered from numerous re-interpretations and a general lack of standardization in its usage. The term addict and its derivatives were used in reference to alcohol by 1612 and tobacco by 1779 (Oxford English Dictionary) and described habitual morphine use by 1909 (Oxford English Dictionary [Supplement]). Both morphine addiction (Tatum, Seevers, & Collins, 1929) and cocaine addiction (Tatum & Seevers, 1929) were described in scientific papers in 1929 without explicit definition of the term addiction. Drug addiction was defined, however, by Tatum and Seevers in 1931 as "a condition developed through the effects of repeated actions of a drug such that its use becomes necessary and cessation of its action causes mental or physical disturbances (p. 107)." The nature of these disturbances was vividly described in the literature.
The early use of the term addiction appears to have described a condition where chronic and habitual drug use develops and the individual becomes intensely preoccupied with the drug and its effects. Later definitions focused on physiological dependence, and drug addiction has come to be applied commonly to describe a condition where the organism becomes physiologically dependent on a substance. The general confusion about the use of this term has led several agencies to advocate abandoning the use of this term altogether (e.g., American Psychiatric Association; World Health Organization) and to replace it by clearly defined terms that have not yet been adopted into common usage. The obvious advantage of coining new terms is that they do not carry the connotations that have been developed over a long period of poorly defined use. However, it is difficult to totally eliminate a word from language by edict, and it is perhaps preferable to re-define the term for scientific usage if the lay usage of the term carries a similar meaning.
The general lexical definition of addiction is notably different from the special application that it has been given by many pharmacologists and clinicians. The American Heritage Dictionary defines addiction under the verb addict as follows:
Jaffe (1975) has provided a more precise, scientific definition of addiction:
The term drug abuse has fewer problems with its definition. The lay and scientific communities have used this term in similar fashions, although several scientific agencies have again recommended that this term be dropped from professional usage. For the purposes of this chapter, and indeed much of this book, the term drug abuse has been used as defined by Jaffe (1975):
Role of Drug Reinforcement in Drug Addiction
Because the term addiction is limited to a description of behavior, other factors must be identified that control the development and expression of drug addiction. One approach to studying drug addiction involves describing the relationship of the behavior to the consequences of that behavior. That is, the behavior of drug taking is governed by the direct and immediate consequence of that behavior—drug administration and the ensuing pharmacological actions of the drug. Essentially, this model of drug addiction is a simple extension of basic operant psychology. Viewed with this perspective, drug addiction is just another behavior controlled by its consequences and does not represent any special class of motivation or behavior (see Bozarth, 1986). The term reinforcement is used to describe the relationship between the behavior (i.e., drug taking) and the consequences of that behavior (i.e., drug effect). Drug self-administration is the paramount case of drug reinforcement—the operant response directly produces administration of the addictive substance. In operant terms, reinforcement is said to occur when the presentation of the reinforcing stimulus (i.e., drug) increases the probability (or frequency) of the behavior that presentation of the stimulus is contingent upon. Positive reinforcement refers to the situation where the presentation of a reinforcing stimulus increases the frequency of a behavior, and negative reinforcement refers to the cases where removal of some stimulus (usually an aversive event such as electric shock) results in an increase in the frequency of some behavior. Although the motivational properties of stimuli serving as positive and negative reinforcers would appear to be much different, the consequence of these stimuli is the same—they both increase the frequency of the behaviors they are associated with. Punishment, on the other hand, represents a much different situation where the presentation of some stimulus (again, usually aversive) suppresses the behavior that it is associated with. It is important to keep in mind the distinction between punishment and negative reinforcement. These terms are frequently used incorrectly as synonyms because they both involve aversive stimuli, but they in fact represent much different processes involved in the control of behavior.
There are two general models of drug addiction that have evolved from reinforcement theory. One model focuses on the positive reinforcing properties of a drug and is probably related to drug-induced mood elevation and euphoria. The second model focuses on the potential negative reinforcing properties of drug administration; chronic administration of many drugs can produce physiological dependence, and drug abstinence can produce an aversive motivational state that is relieved by subsequent drug intake. Although both models are derived from reinforcement theory and emphasize observable behavior and the pharmacological actions of addictive drugs, their motivational effects can be much different. Negative reinforcement models have figured prominently in some traditional theories of drug addiction, but positive reinforcement models appear to have become dominant in recent years. Indeed, most of the chapters in this book reflect the new emphasis on positive reinforcement models of drug addiction.
The use of the phrase "the role of reinforcement in drug addiction" is somewhat misleading. Any event can be studied in terms of reinforcement theory, and the use of this phrase really implies a scientific approach rather than a cause of drug-taking behavior. There is, however, an important feature that distinguishes this approach from other theoretical orientations—the emphasis of this approach is usually on the contingencies of drug self-administration and the pharmacological properties of the drug.
It may be argued that reinforcement theory offers no more than a nominal explanation of drug addiction. In fact, there does appear to be an inherent circularity in describing a behavior as "produced by" the consequences of that behavior. Reinforcement models of drug addiction have, however, transcended their strict operational definitions and generate more implicit rules governing drug-taking behavior. These models generally focus on the intrinsic reinforcing properties of the drug and place far less emphasis on special conditions that are necessary for the drug to serve as a reinforcer. In particular, the observation that most drugs that are potent reinforcers of human behavior (and hence are potentially addictive) are also potent reinforcers of animal behavior has received considerable attention and further supports the notion that some inherent property of the drug is critical for this reinforcing action. Thus, reinforcement theory does not rely strictly on nominally explaining drug-taking behavior but provides an empirical construct that has evolved from this perspective of drug addiction.
Rationale for the Study of Drug Reinforcement
Compulsive drug-taking behavior is the defining characteristic of an addiction, and drug reinforcement is a primary determinant in drug-taking behavior. Thus, the study of drug reinforcement is, in essence, the study of drug addiction. The general rationale for the study of drug reinforcement requires no explanation, but a brief outline of some of the more obvious reasons is nonetheless appropriate. The methods used to study drug reinforcement provide an instrument for:
2. the determination of brain mechanisms mediating drug addiction (see chapters by Broekkamp; Phillips & Fibiger; Roberts & Zito);
3. the screening of new compounds for potential addiction liability (see chapters by Brady, Griffiths, Hienz, Ator, Lukas, & Lamb; Reid; Weeks & Collins; Yanagita; see also chapters by de Wit & Johanson; Haertzen & Hickey; Henningfield, Johnson, & Jasinski);
4. the development of treatments for drug addiction (see chapter by Amit, Smith, & Sutherland);
5. the study of structure-activity relationships in medicinal chemistry (see chapter by Glennon & Young); and
6. an approach to studying basic motivational processes (see chapter by White, Messier, & Carr; see also Bozarth, 1982, 1986).
Principal Methods For Assessing
The following section presents a brief overview of the major methods used to assess drug reinforcement in laboratory animals and in humans. Some of the main advantages and disadvantages of each technique are outlined, although no attempt is made to provide a detailed discussion of the various techniques. This section is presented only to summarize these methods, and the reader should consult the individual chapters specifically addressing the various experimental methods for details.
Intravenous self-administration involves preparing animals with intravenous catheters and allowing them to self-administer a drug (see chapters by Brady, Griffiths, Hienz, Ator, Lukas, & Lamb; Weeks & Collins; Yanagita; Yokel). Some behavioral response, usually lever pressing, is followed by intravenous drug administration, and the ability of the drug injection to directly reinforce behavior is determined. Two basic procedures are used to establish the self-administration of a drug. The cross substitution procedure involves initially training animals to self-administer a standard reinforcing drug such as cocaine and then substituting the test drug and determining if the new drug maintains self-administration. This method has the advantage of circumventing the drug’s influence on learning when screening potentially reinforcing drugs, but has the disadvantage associated with possible pharmacological interactions between the training drug and the test drug. The direct acquisition method involves testing drug-naive animals for the acquisition of intravenous self-administration.
The largest advantage of the intravenous self-administration technique is that the principles of operant conditioning can be directly applied to the study of drug reinforcement. Intravenous drug self-administration can control behavior in much the same manner as traditional reinforcers such as food and water in hungry or thirsty animals (for a review, see Spealman & Goldberg, 1978; see also Katz & Goldberg, this volume; cf. Wise, this volume).
There are several specific advantages of using intravenous self-administration to study drug reinforcement. First, the patterns of drug intake can be used to distinguish reward from motoric effects of pharmacological challenges (see chapters by Yokel; Wise). Second, principles of operant conditioning can be directly applied to assess drug reinforcement. For example, progressive-ratio testing can determine the relative reinforcing strength of several compounds (see chapters by Brady, Griffiths, Hienz, Ator, Lukas, & Lamb; Weeks & Collins; Yanagita; Yokel), and persistence of drug-seeking behavior and relapse can be directly studied by examining extinction patterns of responding (see Stewart & de Wit, this volume). Third, the pattern of drug intake can be evaluated and cases of regulated drug-intake and binge intake can be easily distinguished (see Yokel, this volume). Fourth, choice procedures have been developed where the animals choose one of two reinforcing drug injections; either different doses of the same compound or two different drugs can be tested with this method.
There are also several disadvantages to using intravenous self-administration, and these disadvantages are severe enough to restrict the number of laboratories currently using this technique. First, the intravenous preparation is relatively difficult to maintain; subject loss due to blocked or leaky catheters and illness can be quite high. Second, although animals learn quite quickly to self-administer some drugs (e.g., cocaine, heroin), weeks and sometimes months of testing is often required for response patterns and drug-intake levels to stabilize. Third, the dose range tested for self-administration is very important; if excessively high drug doses are tested, aversive side-effects may inhibit drug self-administration; if very low doses are tested, the drug may not be sufficiently rewarding to establish or maintain operant behavior. Fourth, possible drug interactions and learning variables can be important when the cross substitution procedure is used; abrupt decreases in the reward magnitude can produce a negative contrast effect, and animals might fail to maintain self-administration of a drug that would otherwise be reinforcing. Also, important interactions may occur between the effects of the training drug and the test drug. For example, animals initially trained to self-administer heroin would be unlikely to self-administer an agonist/antagonist like pentazocine because the latter compound would precipitate withdrawal which would presumably be aversive. In direct acquisition studies, drugs that disrupt learning may retard the acquisition of drug self-administration, even though they can serve as reinforcers.
Intravenous self-administration has a high degree of face validity. It is the most direct method of studying drug reinforcement because it directly measures the ability of a drug to reinforce behavior; it is viewed by most drug addiction specialists to be unequivocal demonstration of a drug’s reinforcing action, provided certain behavioral control procedures are tested. Furthermore, there is excellent concordance between the ability of a drug to support intravenous self-administration in animals and its addiction potential in humans (see Griffiths & Balster, 1979; Weeks & Collins, this volume).
Other Self-Administration Procedures
There are several other self-administration procedures that have been used in laboratory animals. Each of these methods has been applied in special situations, although their use is generally not as widespread and they lack the consentaneous validity given the intravenous technique.
Oral self-administration has been shown for several drugs (see chapters by Amit, Smith, & Sutherland; Meisch & Carroll). This method has the advantage of being relatively easy but suffers from controversy regarding the interpretation of some data derived from this technique. Nonetheless, oral drug self-administration has three advantages that establish it as a useful technique for studying drug reinforcement. First, it is easy and does not require surgical preparation of the subjects or special apparatus. Any laboratory with animal housing facilities and drinking bottles can use this technique. Second, the experimental preparation remains viable for long periods of time. Studies that require extensive testing of the subjects can be completed where the attrition rate in intravenous self-administration would be prohibitive. Third, large numbers of subjects can be easily tested. When assessing the effects of brain lesions or pharmacological challenges, individual subject reactivity to the treatment may require that large sample sizes be used. The oral self-administration procedure, because of its low equipment and labor requirements, is especially suitable for testing such large numbers of subjects.
Intragastric self-administration has been shown for several compounds (e.g., see Yanagita, this volume), but it does not appear to have any significant advantages over intravenous self-administration for most drugs. The argument has been made that drugs abused by the oral route of administration are best studied by this technique, but that argument has not been substantiated.
Intracranial self-administration has been shown for several compounds (see Bozarth, 1983, this volume). Self-administration procedures have been modified to produce drug microinjections directly into brain tissue. This method has the distinct advantage of localizing the reinforcing action of a drug (provided certain control procedures are observed), but it is methodologically difficult and only a few laboratories have reported success with this technique. Intracranial self-administration is a valuable technique for asking certain experimental questions that cannot be addressed with other experimental methods.
The concordance of the results from these methods with the documented addiction liability of various drugs in humans has not been evaluated. Some addictive drugs are correctly categorized with each of these measures, but too few compounds have been tested to fully evaluate the overall accuracy of these methods in predicting addiction liability.
Conditioned Place Preference
Conditioned place preference is a relatively new method of assessing the rewarding effects of drugs. With this method the animal develops an association between the rewarding action of a drug and specific environmental cues. When tested in the drug-free state, the animal approaches and maintains contact with the environmental cues that have been associated with the rewarding drug. This procedure involves several conditioning trials where the animal is injected with the test compound and placed in a specific compartment of the test apparatus containing various cues (e.g., tactile, visual, and/or olfactory). The animal is later tested to determine if it increases the amount of time spent in the compartment associated with the drug injection. If so, a conditioned place preference is said to have developed, and the drug is presumed to have some rewarding action. There are numerous variations of this basic method (see chapters by Bozarth; Phillips & Fibiger; Reid; van der Kooy; White, Messier, & Carr), but all of these procedures appear to produce strikingly similar results.
The conditioned place preference method appears to be quick and easy. Only simple equipment is required, and no special surgical preparation of the animals is needed. Procedural differences do not appear to be important, although some methods of testing may be preferred over others (see van der Kooy, this volume). The motor facilitating or impairing effects of some drug treatments that adversely influence most measures of drug reward do not have a significant influence with this method, because the response requirement is very simple and because the animals are usually tested in the drug-free state when the motoric effects of these drugs are absent. Lastly, this technique offers a direct method of studying conditioning effects which are important in the control of drug-taking behavior.
The main disadvantage of this method is that little is really known about the basis of this effect. Manipulations that should have a strong influence on classical conditioning have a relatively small effect on conditioned place preference. Also, investigators using much different conditioning and testing procedures report surprisingly similar strengths of conditioning; the subjects usually show around a 200 second increase in the time spent in the conditioning environment during a 15-minute test. Finally, although not documented in the literature, the conditioned place preference technique may be prone to spurious results; that is, individual tests where rewarding drugs do not produce a preference and other tests where nonrewarding drugs produce a slight, but statistically significant, conditioned place preference may occur with an unacceptably high frequency; this makes replication of important findings derived from this technique very important.
The concordance of conditioned place preference studies with human addiction liability has not yet been established. Conditioned place preferences have been shown for a wide variety of drugs that are addictive in humans, and nonaddictive drugs generally do not produce a place preference (see Bozarth, this volume). Furthermore, the place preference produced by heroin is very reliable over numerous replications (see Bozarth, this volume). Because this technique is relatively new and because the variables controlling place preference conditioning are not well understood, caution should be exercised when interpreting the results of these studies. Specifically, the results of conditioned place preference studies should be considered in concert with the findings from more established techniques, and this method is probably most valuable as a preliminary screen for addiction liability and for answering experimental questions that cannot be suitably addressed with other methods.
Other Conditioning Procedures
There are several other conditioning procedures that have been used to study the rewarding effects of drugs. The two techniques that have received the most study are conditioned reinforcement (see Davis & Smith, this volume) and reinstatement of responding (see Stewart & de Wit, this volume).
Conditioned reinforcement studies involve developing an association between a rewarding drug injection and some stimulus (e.g., visual, auditory) that is concurrently presented with the drug infusion. The animal is later tested to determine if it will lever press to produce presentation of the drug-associated stimulus. If the drug injections were reinforcing, then lever-pressing behavior should be maintained (or established) by the presentation of the stimulus event associated with the drug reward. One of the most important applications of this technique is testing manipulations that may interfere with operant responding. Because the animal passively receives the rewarding drug injections during the conditioning trials, pharmacological treatments that interfere with response performance should not affect the establishment of the conditioned reinforcer. Testing for the putative rewarding drug effect occurs when the animal is drug-free (as in the place preference conditioning method) and thus should not be influenced by motor-impairing drug effects.
Reinstatement of operant responding involves a much different procedure with a similar conceptual basis. The animal is first trained to intravenously self-administer a drug. It is then subjected to extinction trials where the drug vehicle (usually physiological saline) is substituted for the reinforcing drug. After several extinction trials, the response rates drop to control levels. Next, a priming drug injection is given noncontingently to the animal. If the drug effect is similar to the rewarding effect of the drug that the animal was trained to self-administer, the animal will reinstate lever pressing even though it only receives the drug vehicle with each self-administered infusion. This phenomenon is probably related to the priming effect that is seen in many behavioral studies, and it may be related to the sensations of drug craving in humans. There are two important potential applications of this technique. First, this procedure may provide a model of relapse to drug self-administration (and addiction), and conditioning factors involved in drug-taking behavior can be directly studied. Second, this method appears to have a great potential as a preclinical screening tool. Animals can be trained on standard drugs such as cocaine or heroin and the abilities of new compounds to reinstatement responding determined. If the test drug produces an increase in operant responding, then it probably has properties very similar to the training drug. A major advantage of this method over other techniques that are designed to test similarities across compounds (e.g., drug discrimination) appears to be the ease of training the subjects. Another potential advantage is that pharmacologically dissimilar drugs that share similar addiction liabilities may show cross generalization. For example, animals trained on cocaine show a reinstatement of responding when given a noncontingent injection of morphine (de Wit & Stewart, 1981). This cross generalization combined with the ease of training and the fact that (unlike conditioned place preference) these animals can be repeatedly tested on various drugs makes this method of assessing drug reinforcement one of the most promising new methods for screening compounds for addiction liability.
Brain Stimulation Reward
This method of studying drug reward involves training animals to work for electrical brain stimulation and determining the effects of drugs on brain stimulation reward (see chapters by Esposito, Porrino, & Seeger; Lewis & Phelps; Reid). Addictive drugs appear to enhance or facilitate brain stimulation reward; this facilitation effect can be demonstrated as an increased rate of lever pressing for fixed intensity brain stimulation (see Reid & Bozarth, 1978) or as a lowering of current thresholds for brain stimulation (see Esposito & Kornetsky, 1978). Although many manipulations can attenuate brain stimulation reward including both reward and response-impairing manipulations, facilitation of brain stimulation reward has not been reported for pharmacological treatments that lack independently confirmed, directly reinforcing actions.
Both rate and threshold measures of brain stimulation should reveal any facilitatory effect that the test drug might produce. Rate measures are easy to obtain and require only simple equipment. The primary disadvantage seems to be that rates can vary considerably over the course of a test session. Brief pauses in responding can greatly affect the lever-pressing rate for small time samples (e.g., 1 to 5 minutes). If one attempts to determine the drug’s time-course by continuously testing the subject for an hour or more and to resolve response-rate measures into short time samples (e.g., 10-minute time periods), the effects of fatigue and response pauses can adversely influence the rate measures. It is probably better to test the subject for only a few minutes of each hour (or perhaps at 30-minute time periods) to increase the likelihood that the animal will not take breaks or experience significant fatigue. Such testing has yielded stable baseline rates of responding over the course of 24 hours (e.g., Atalay & Wise, 1983). Alternatively, most threshold measures are not influenced by response pauses or fatigue, and the animal’s threshold can be continuously "tracked" over the course of several hours of testing. Threshold procedures require, however, more sophisticated equipment or more laborious testing procedures although microcomputers can easily automate the procedure.
This technique is very easy and considerable experience has been gained with brain stimulation reward over the past three decades. The surgical procedure is simple and the animals are easy to maintain. The drug effect is assessed against an already established behavior so learning effects are not important. The initial drug dose tested is not critical, because the animal is working for the rewarding effects of the electrical stimulation and behavior will not be extinguished with subrewarding drug doses as in the intravenous self-administration method. Repeated drug testing does not disrupt the animal’s behavior, and full time-course and dose-response data can be easily generated. The pharmacokinetics of the drug’s action can provide direction to other, more direct methods of assessing drug reinforcement, and the use of this technique for determining dose and time-after-injection parameters is probably the most important application of brain stimulation reward in screening drug addiction liability.
There has been controversy regarding the use of some measures. For example, the use of response rates has been challenged (e.g., Kornetsky & Esposito, 1979), and the two-lever autotitration method appears to have some problems (see Fouriezos & Nawiesniak, this volume). The most serious limitation of this technique involves consideration of what is actually being measured—the interaction of the addictive drug with the neural substrate presumed to mediate its rewarding action. Brain stimulation reward studies provided some of the earliest evidence that addictive drugs may interact with brain reward systems, and early applications of this technique (e.g., Broekkamp, 1976; see also Broekkamp, this volume) have been very influential in subsequent work studying the neural basis of drug reward (e.g., Bozarth & Wise, 1981). Nonetheless, the effect of drugs on brain stimulation reward remains an indirect method of assessing addiction liability.
The overall concordance of the results of brain stimulation reward studies with human addiction liability data is very good (see Bozarth, 1978). In fact, brain stimulation reward studies predicted the long-term abuse liability of several compounds that were initially "missed" by intravenous self-administration studies (see Bozarth, 1978). The potential usefulness of this technique for determining the dose and time course of potentially rewarding drug-effects is probably unequaled by any other method, and this technique can potentially make a unique contribution to the preclinical screening of drugs for addiction liability.
Drug discrimination involves training an animal to make one of two (or more) alternate responses when under the influence of a specific drug (see chapters by Colpaert; Overton). For example, the animal is injected with a training drug, and one of two possible responses (e.g., depressing the left-hand or right-hand lever; turning left or right in a T-maze) is reinforced (e.g., food reward presented to a food-deprived animal). Alternative trials are conducted under either placebo or another drug; the subject must learn to make a different response to obtain reinforcement when it has not been injected with the training drug. Once the animal has learned to discriminate the training drug effect from that of placebo and to make the correct response, new compounds can be tested for their ability to produce stimulus properties similar to the training drug. If the test drug produces similar cue properties, the subject should make the same response as when it is injected with the training drug.
Drug discrimination offers a very sensitive measure of the stimulus properties of various drugs. Animals can successfully discriminate various classes of drugs (e.g., ethanol, benzodiazepines, psychomotor stimulants, opiates), and they can even distinguish between two drugs in the same general pharmacological class (e.g., the stimulants amphetamine and apomorphine; Hernandez, Holohean, & Appel, 1978). Dose-response effects are easily obtained, and new compounds can be fairly quickly tested once the animal has been successfully trained in the discrimination task. This general technique has been used to study structure-activity relationships for several classes of drugs (e.g., see Glennon & Young, this volume), and relatively small changes in molecular structure can produce large changes in the compound’s stimulus properties. With the procedures used by most investigators, drug discrimination studies are relatively insensitive to the response-impairing effects of some drugs since a disruption of motor capacity would only decrease response rates and not influence which choice the animal makes.
Most methods of training subjects to reliably discriminate the stimulus properties of a drug involve long training periods and a substantial investment in time and effort. Retraining periods must be interspersed with test sessions to insure that the animal still makes the correct response when given the training drug. The effect of various treatments on the motivational properties of the primary reinforcer have not been evaluated, but changes in motivational level may influence response selection and this variable needs to be systematically evaluated. Most importantly, drug discrimination is based on the stimulus properties of a drug and there is no reason a priori to suppose that the most salient drug cues are necessarily related to the attributes that the investigator wishes to assess. For example, opiate drugs produce a number of effects including reward, analgesia, physiological dependence, changes in gastric motility, and changes in other autonomic nervous system functions. When a drug is reported to have stimulus properties similar to morphine, any one (or more) of the myriad of stimulus effects could be responsible for this effect. It is indeed erroneous to conclude that the test drug is an analgesic or that it will have an addiction liability similar to morphine.
The concordance of drug discrimination studies with human addiction liability appears reasonably good, but there has been no systematic evaluation conducted across a wide range of drugs. The stimulus properties of a drug in humans (as measured by instruments such as the Addiction Research Center Inventory) may be an excellent predictor of the drug’s potential for abuse, but there is no reason to presuppose that similar stimulus properties in animals are responsible for the generalization in the drug discrimination method. In reporting the subjective effects of addictive drugs in humans, special attention has been directed toward identifying response items (e.g., euphoria, drug liking) that correctly classify highly addictive drugs. Thus, not all of the stimulus properties of a drug necessarily constitute its subjective-effects profile that predicts addiction liability.
Subjective-Effects Measures in Humans
Some of the earliest studies attempting to identify variables that would predict a drug’s addiction liability involved assessing the subjective effects of the drug after administration in humans. Much of this work was pioneered at the Addiction Research Center in Lexington, Kentucky, and it usually involved determining the effects of various addictive drugs in ex-addict, prisoner volunteers from the local federal penitentiary.
Questionnaires have been designed to evaluate the subjective effects of drugs in humans. One of the tests, the Addiction Research Center Inventory (ARCI), has several specific scales that identify drug effects similar to prototypical addictive agents (see Haertzen & Hickey, this volume). The ARCI has been very successful in accurately classifying addictive drugs. It is an empirically keyed questionnaire based on a very large sample size, and it can be used with either single dose or chronic drug evaluation. Several other questionnaires have also been developed, some of which are based on items derived from the ARCI (see Henningfield, Johnson, & Jasinski, this volume).
The major advantage of subjective-effects questionnaires is that they have been shown to accurately classify the addiction liability of known addictive drugs. When used following a single drug injection, they also minimize the subject’s exposure to the potentially addictive drug. Two important limitations of these tests are that new addictive compounds may possess properties that are not identified as "like" the prototypical addictive drugs the scoring system is based on and that repeated exposure or subject-controlled exposure may influence the addiction liability of the compound. Nonetheless, subjective-effects measures are frequently employed even where other measures of drug reinforcement are the primary interest (e.g., see de Wit & Johanson, this volume). The concordance of these measures with the observed addiction liability of most drugs is very high.
Human Drug Self-Administration
A more direct method of assessing the reinforcing properties of addictive drugs in humans involves tests for drug self-administration (see chapters by Henningfield, Johnson, & Jasinski; Mello & Mendelson). Laboratory procedures very similar to those used in animal intravenous self-administration can be adapted for use in humans. Not surprising, some of the same patterns of responding seen in animals are also seen during human drug self-administration (e.g., see Mello & Mendelson, this volume).
The major advantage of this technique is that it directly assesses the reinforcing properties of the drug. The major disadvantages are that this method is costly and that it usually involves considerable exposure to the potentially addictive substance, thus risking the development of an addiction in the experimental subjects. This technique has not been used extensively to screen new compounds (Its primary use seems to have been to study known addictive drugs.), so it is premature to judge its usefulness for predicting the addiction liability of new drugs. The face validity of this method, however, is extremely high.
Choice Testing in Humans
A relatively new procedure for assessing drug reinforcement in humans involves choice testing (see de Wit & Johanson, this volume). Subjects are given alternative trials with two or more color-coded capsules and are instructed to associate the drug effects with the different capsules. After several training trials have been completed, the subjects are tested in a choice condition where they can select one of the two (or more) color-coded capsules for ingestion. The relative preference for the capsules is used as a measure of drug reinforcement, and subjects generally show a significant preference for drugs with a known addiction liability.
The are several advantages to using this technique for assessing drug addiction liability in humans. First, relatively few drug exposures are necessary to demonstrate a preference for highly addictive drugs; this minimizes the exposure of the experimental subjects to potentially dangerous or addictive drugs. Second, choice testing can be conducted between drug and placebo or among two or more drugs. Relative measures could potentially demonstrate preferences for various addictive drugs or different dosages of the same drug. Third, subjective-effects measures can be easily taken during the training trials without disrupting the subject’s performance. Thus, choice testing can be conducted concurrently with subjective-effects assessment, and the drug evaluation can be based on the outcome of both measures. Fourth, this experimental method is relatively inexpensive and does not require any special apparatus or facilities. Fifth, this technique does not require that the subjects be tested as inpatients. This greatly increases the range of subject populations that can be easily recruited for this procedure, and it can allow the subjects to experience the drug effects in their natural environment.
Most choice testing involves relatively few exposures to the test drug. Some addictive drugs may require repeated administration, higher doses, or special conditions to be reinforcing, and this technique may not accurately predict abuse liability in these situations.
Choice testing accurately classifies the addiction liability of amphetamine, but an insufficient number of other compounds has been tested with this method to determine if other classes of addictive drugs show similar effects. The technique is very promising, but considerably more work needs to be completed before its usefulness for screening new compounds can be accurately evaluated.
Cross Validation of Experimental Procedures
During the past decade, considerable progress has been made in identifying and characterizing the neural basis of psychomotor stimulant and opiate rewards. The ventral tegmental dopamine system appears to be involved in both classes of drug rewards (see Bozarth, 1985, 1986; Bozarth & Wise, 1983; Phillips & Fibiger, this volume; Wise & Bozarth, 1982, 1984). This system has its cell bodies in the ventral tegmental area and sends its axonal projections forward to terminate in several brain regions including the nucleus accumbens and frontal cortex. Although other brain mechanisms may also participate in reward from these compounds, the data suggesting the involvement of this brain system will be briefly examined to illustrate how various methods of assessing drug reinforcement have provided similar answers to the same general question. The concordance across markedly different methods of study provides a type of empirical test for the validity of each measure. The conclusions drawn from this research far exceed the limitations of any single method of studying drug reinforcement.
Psychomotor Stimulant Reward
No fewer than nine independent studies using four different
paradigms have suggested that the nucleus accumbens (NAS) terminal
of the ventral tegmental dopamine system is critically involved in the
rewarding effects of psychomotor stimulants (see Table 1).
of amphetamine into the NAS facilitate brain stimulation reward (BSR).
Dopamine-depleting lesions of the NAS or of the cell body region of
system (ventral tegmental area: VTA) disrupt intravenous
(IVSA) of cocaine. Dopamine-depleting lesions of the NAS disrupt the
and the maintenance of amphetamine self-administration. Kainic acid
that destroy the target cells of the NAS dopamine terminals also
cocaine self-administration. Amphetamine is intracranially
(ICSA) directly into the NAS. Finally, NAS microinjections of
produce a conditioned place preference (CPP), and the conditioned place
preference produced by systemic amphetamine injections is attenuated by
dopamine-depleting lesions of the ventral tegmental system.
|BSR/NAS amphetamine microinjections||Broekkamp et al., 1975|
|IVSA/NAS dopamine-depleting lesions||Lyness et al., 1979
Roberts et al., 1977, 1980
|IVSA/VTA dopamine-depleting lesions||Roberts & Koob, 1982|
|IVSA/NAS kainic acid lesions||Zito et al., 1985|
|ICSA/NAS amphetamine||Hoebel et al., 1983|
|CPP/NAS amphetamine microinjections||White et al., this volume|
|CPP/NAS dopamine-depleting lesions||Spyraki et al., 1982|
Some evidence suggests that the frontal cortex projection of the ventral tegmental dopamine system may also be involved in psychomotor stimulate reward. The relative importance of the nucleus accumbens and frontal cortex terminal fields has not been established, but nine studies have suggested a role for the nucleus accumbens (see Table 1) while only three studies have suggested a role for the frontal cortex (i.e., Goeders & Smith, 1983; Glick & Marsanico, 1975; Phillips, Mora, & Rolls, 1981).
At least 14 independent studies using four different experimental paradigms have suggested that the ventral tegmental area is involved in opiate reward (see Table 2). Morphine microinjections into the ventral tegmental area (VTA) facilitate brain stimulation reward (BSR). Dopamine-depleting lesions in the VTA attenuate the facilitatory action of systemic morphine injections on BSR. The acquisition of intravenous heroin self-administration (IVSA) is disrupted by dopamine-depleting lesions of the VTA, and narcotic antagonist microinjections into the VTA disrupt heroin self-administration. Animals trained and later extinguished on intravenous morphine self-administration show a reinstatement of responding following VTA morphine microinjections. Both fentanyl and morphine are intracranially self-administered (ICSA) directly into the VTA. Morphine, an opioid peptide, and an enkephalinase inhibitor all produce a conditioned place preference when microinjected into the VTA. Finally, the conditioned place preference produced by systemic morphine injections is attenuated by dopamine-depleting lesions of the VTA system.
There is evidence that opiate reward may involve additional systems.
Other brain regions that have been suggested to mediate opiate reward
the caudate nucleus (Glick, Cox, & Crane, 1975), the nucleus
(Goeders, Lane, & Smith, 1984; Olds, 1982; Vaccarino, Bloom, &
Koob, 1985), and the lateral hypothalamus (Olds, 1979). The
gray region may also contribute to the net reinforcing impact of
delivered opiates (see Bozarth, 1986; Bozarth & Wise, 1983).
|BSR/VTA morphine microinjections||Broekkamp et al., 1976
Broekkamp et al., 1979
|BSR/VTA dopamine-depleting lesions||Hand & Franklin, 1985|
|IVSA/VTA dopamine-depleting lesions||Bozarth & Wise, 1986|
|IVSA/VTA narcotic antagonist microinjections||Britt & Wise, 1983|
|IVSA/VTA morphine reinstatement of responding||Stewart, 1984
Stewart & de Wit, this volume
|ICSA/VTA fentynal||van Ree & De Wied, 1980|
|ICSA/VTA morphine||Bozarth & Wise, 1981, 1982|
|CPP/VTA morphine||Bozarth & Wise, 1982
Phillips & LePiane, 1980
|CPP/VTA opioid peptide||Phillips & LePiane, 1982|
|CPP/VTA enkephalinase inhibitor||Glimcher et al., 1984|
|CPP/VTA dopamine-depleting lesions||Spyraki et al., 1983|
The overall concordance of these various experimental methods is excellent. Markedly different experimental paradigms and several variations of these methods of study have yielded very similar conclusions. Both psychomotor stimulants and opiates appear to derive a major part of their reinforcing effects by activation of the ventral tegmental dopamine system, although at different synaptic elements (i.e., the nucleus accumbens terminal field and the ventral tegmental area cell body region, respectively; for a review, see Bozarth, 1986). A recent study has provided a direct test of this hypothesis by demonstrating a partial cross substitution of ventral tegmental morphine microinjections for intravenous cocaine reinforcement (Bozarth & Wise, 1986). Most importantly, even though there is evidence to suggest that additional mechanisms may be involved in the reinforcing actions of both classes of drugs, the concordance among markedly different experimental techniques and among various independent laboratories provides important cross validation of these methods of assessing the reinforcing properties of abused drugs.
It is interesting to note that the neural substrates for both amphetamine and morphine reward were first identified by brain stimulation reward studies combined with microinjection methodology (see chapter by Broekkamp for a description of the experimental technique; see also Broekkamp  for a description of the studies). The results of these studies seem to have been initially ignored by most drug addiction specialists because the relevance of these findings to drug reward was not generally appreciated. Ironically, later studies confirming the rewarding sites of action for these drugs (such as the work with intracranial self-administration) have helped to provide validation of the brain stimulation reward method as a technique for assessing the reinforcing properties of these abused drugs.
Screening Drugs For Addiction Liability
One of the most important applications of assessing drug reinforcement is screening new compounds for addiction liability. Several pharmacological classes of drugs used for specific therapeutic applications are traditionally associated with a moderate to high addiction potential. Most notable are the strong analgesics, appetite suppressants, stimulants, and hypnotic sedatives. Minor tranquilizers have also been suggested to be somewhat addictive, although their addiction liability appears far less than their abuse potential.
Medicinal chemists can generate new compounds at a rapid rate by making slight alterations in the basic drug molecule. The objective is to develop compounds that retain a high therapeutic efficacy while minimizing the undesirable side-effects such as addiction liability. Fast and effective methods are required to assess the addiction liability of these new compounds before clinical trials are begun. Probably a very high proportion of compounds never progress beyond the initial phase of drug screening because alterations in the drug molecule fail to significantly diminish the undesirable side-effects of the drug or do so by also diminishing the therapeutic efficacy.
Screening new compounds for addiction liability is divided into two phases—preclinical and clinical. Although both phases are concerned with assessing addiction liability, they differ markedly in their immediate objectives. Preclinical studies involve screening compounds using animal models and must progress at a quick pace and need not concern themselves with the long-term effects on the experimental subjects. Methods used for preclinical screening should maximize the potential for drug addiction by using a wide range of dose levels and exposures to the compound. Clinical studies must investigate the addiction liability of a compound without producing long-term adverse effects in the subjects. This poses a particularly difficult task, because clinical screening must expose the subjects to a potentially addictive compound under conditions that will maximize the probability of detecting the drug’s addiction potential while minimizing the likelihood of producing an actual drug addiction. Unfortunately, the conditions that maximize the potential for detecting the addictive properties of a drug also maximize the potential for producing an addiction in the experimental subjects. This situation has led to the development of methods that detect the underlying properties of a compound that are intimately related to its addictive properties under conditions that minimize the risk of actual addiction to the compound.
Two important results ensued, in part, from this dilemma. First, methods have been developed that can directly assess addiction liability in lower animals, and these techniques provide the first line for eliminating compounds with a high addiction liability. The animal models are so good as to almost eliminate the need for screening in humans, but clinical assessment remains the final determination of whether a new compound will be advanced to clinical trials. The techniques used to assess addiction liability in humans have also greatly progressed. Instead of directly measuring the ability of the compound to produce compulsive drug use, measures have been developed that determine the subjective effects and "liking" for the compound. These variables appear to be intimately related to a drug’s addiction potential and are usually, in themselves, accurate predictors of a drug’s addiction potential. The limiting factors in the accuracy of clinical assessment involve conditions of drug exposure (e.g., drug dosage or days of repeated exposure) and subject variables (e.g., subjective impressions of the psychological state produced by the compound) that are not important in animal studies. Thus, preclinical screening remains an important step in the process of drug development not only for its ability to rapidly screen large numbers of compounds, but also for its ability to use higher drug doses and longer conditions of exposure that would be impractical (or unethical) in routine clinical screening. The final assessment in humans, however, may detect properties of a compound that ultimately produce an addiction that might be missed in lower animals.
The preclinical screening of new compounds for addiction liability can be divided into three stages. The first stage determines the drug dosage and time course of the drug effects that will be studied in the subsequent tests. The second stage involves an initial demonstration that the new compound has (or is likely to have) a substantial reinforcing or rewarding effect. The third stage is the most difficult but also provides the most direct demonstration that the administration of the compound is reinforcing and, hence, is very likely to have significant addictive properties.
Stage I: Initial Determination of Drug Dosage & Time Course
The initial determination of appropriate drug dose and time course for rewarding effects should use a measure that is quick, easy, and allows repeated testing of the subject across various conditions. Brain stimulation reward is uniquely suited for this purpose. Animals with lateral hypothalamic electrodes can be maintained for long periods of time without noticeable deterioration in performance or substantial change in the rewarding effects of the electrical stimulation. A pool of trained subjects can be used to determine the relevant dose and time-course of a new compound’s effect. Testing should continue for several hours after drug administration, and relatively large doses of the compound may be tested. Any delayed facilitation of brain stimulation reward should be apparent if the sessions continue 4 to 6 hours after drug injections.
Threshold-tracking measures are the best approach because they generate stable baseline values when testing the subjects for long session durations. Rate measures may also be suitable, but they would probably require that the subjects be tested periodically for short periods of time (e.g., every 30 minutes for 5-minute test durations). Otherwise, a marked deterioration in the subject’s performance may result from fatigue, and the drug’s effect would have to be evaluated against a changing baseline response rate. Automation of the testing procedure and data analysis is a definite asset because considerable data must be collected to fully evaluate the compound.
From this stage of testing, the dose range to be tested with the other measures can be selected. The dose most likely to be self-administered is probably the one which produces a facilitation of brain stimulation reward with the shortest latency after injections. Larger doses that produce a delayed facilitation, however, may also be self-administered.
The reinstatement of responding method may also be suitable for the initial screening of new compounds, although the feasibility of this technique for this application, particularly the effects of a delayed rewarding action, has not been evaluated. Two important disadvantages to using this technique for Stage I testing are the difficulty in maintaining intravenous subjects for repeated testing (as in the use of a subject pool for screening multiple drugs) and the possible influence that repeated extinction trials may have on this measure. Thus, until the use of the reinstatement of responding method for determining initial drug dosages and time courses of potentially addictive drugs has been evaluated, brain stimulation reward remains the preferred technique for Stage I screening of new compounds.
Stage II: Preliminary Demonstration of Directly Rewarding Drug Action
The next stage of testing a compound for addiction liability involves a preliminary assessment of the compound’s directly rewarding effect. The conditioned place preference method is the preferred procedure for this stage of testing because large numbers of subjects can be quickly and easily screened. The lowest drug dose effective in facilitating brain stimulation reward should be used as the low end of the dose range, and doses one and two orders of magnitude greater than this dosage should also be tested. (There is probably no need to test intermediate doses because conditioned place preferences appear to be produced across a wide range of doses of a compound.) Unless there is evidence from the brain stimulation reward studies that the drug effect is markedly delayed following injection, the subjects can be conditioned immediately after the drug injections. Conditioning duration may be an important variable if the drug has a particularly short or long duration of action, but 30-minute conditioning sessions will probably be satisfactory for many compounds. Again, the time course determined from brain stimulation reward testing can be used to direct the duration of the conditioning trials.
An alternative procedure particularly useful for screening large numbers of compounds and obtaining dose-response data is the drug discrimination method. This method yields additional data, such as whether the test compound is similar or dissimilar to the training drug, and also shows clear dose- dependent effects. The major limitation of this technique for assessing addiction liability is that this method does not necessarily measure drug reward. It is likely that the rewarding action of a compound significantly contributes to the stimulus properties of that drug, but the relative importance of this and other interoceptive cues has not been established. Furthermore, drug discrimination studies are very laborious, and selection of an inappropriate training drug may cause the test drug’s addiction liability to be missed.
The reinstatement of responding method may also be very effective for this stage of testing. A potential advantage of this method over drug discrimination testing is that the training drug and test drug can be markedly different (as in the case of opiates and psychomotor stimulants, see de Wit & Stewart, 1981) and generalization may still occur. The utility of this method for screening potentially addictive drugs merits further exploration.
Stage III: Confirmation of Directly Rewarding Drug Action
The last stage of preclinical screening for addiction liability involves testing the compound for its ability to directly reinforce behavior. Intravenous self-administration is the preferred method for this, but intragastric, intraventricular, and oral self-administration may be useful in some situations. Drug substitution procedures are commonly used to determine if a new compound will maintain intravenous self-administration. This approach has the advantage that the subjects have already been trained to self-administer a drug with known reinforcing properties, so learning factors do not affect the outcome of the test. However, important pharmacological and behavioral interactions may occur that invalidate the conclusions drawn from this procedure. For example, if a compound is a partial narcotic antagonist with low intrinsic reinforcing properties, then testing this compound in an animal that has been self-administering morphine is likely to precipitate withdrawal and the animal may actively avoid lever pressing. Similarly, if a subject has been self-administering a potent reinforcer like cocaine or heroin and a compound with much lower reinforcing efficacy is substituted, then a behavioral contrast effect may be produced and the subject may fail to maintain self-administration. That is, the shift to a reinforcer of much lower incentive value is likely to produce a contrast effect where that reinforcer will be unable to maintain the behavioral response, even though the same reinforcer would normally maintain responding if the abrupt shift in reinforcement magnitude were not present. In both of these cases, the test compound is reinforcing, but its reinforcing efficacy is relatively small.
Direct acquisition of intravenous self-administration is the best approach to assessing a compound’s directly reinforcing effect. This method, however, suffers from several serious limitations. First, the drug dose and pharmacokinetics must be satisfactory for the reinforcing action to be associated with the lever-pressing response. Second, the response must be learned and compounds that disrupt learning may fail to support (or severely retard the acquisition of) self-administration. Third, a drug may be reinforcing only after considerable exposure to that drug. This requires that the animals be involuntarily given drug injections prior to assessing drug self-administration. And fourth, the compound may be reinforcing only under certain circumstances or conditions. Anxiolytic compounds may not be reinforcing in normal experimental animals but be reinforcing in stressed subjects.
It is obviously not practical to test a new compound under all of the conditions that may affect its reinforcing properties. Organismic variables or long-term drug exposure may modify a drug’s reinforcing efficacy. The general pharmacological properties (e.g., analgesic, anxiolytic, or sedative effects) may provide information regarding likely conditions that will reveal a compound’s reinforcing effect not seen with a simple test of direct response acquisition of intravenous self-administration.
A positive finding in any of the three stages of preclinical screening is strongly suggestive that the compound will possess a significant addiction liability. The lack of concordance across these measures does not constitute evidence that the compound is devoid of addiction potential but, rather, suggests that more extensive tests need to be conducted with the procedures that failed to reveal the drug’s addiction potential. The fact that a compound has some addiction liability, however, does not preclude its further assessment as a potentially useful therapeutic agent. A surprisingly large number of medicinal compounds probably possess some addiction liability.
Stage IV: Clinical Assessment
Initial clinical screening for addiction liability should probably use psychometric assessment of the compound’s subjective effects. Instruments, such as the Addiction Research Center Inventory, are very reliable and have considerable accuracy in predicting addiction liability. Single-dose administration (or limited dose-range testing) minimizes the exposure of the potentially addictive compound to human subjects.
Choice testing or drug self-administration is the final step in clinical screening before limited clinical trials are begun. Here, the subjects must be exposed to repeated administration of the compound so the risk of developing an addiction in the human subjects is significant. This approach, however, may reveal addictive properties that cannot be detected using the single dose methods of assessment. The selection of the test compound during choice testing or its voluntary self-administration remains the most direct method of demonstrating an addiction potential. As with the animal self-administration studies, the compound may be reinforcing under special conditions related to repeated exposure or specific circumstances. Even voluntary self-administration testing in humans may miss this reinforcing action.
Stage V: Clinical Trials
After the drug has progressed through the first four stages of testing and has been shown not to be reinforcing in these tests, it is ready for limited clinical trials. During this phase, it is important that dispensing and use pattern information be accurately recorded to determine if there are any indications that the compound possesses a high addiction liability. As mentioned above, conditions that cannot be duplicated in either a laboratory or outpatient settings may influence whether a drug is potentially addictive, and the fact that the earlier tests failed to reveal any addiction liability could be related to any number of variables present in "the real world" but not duplicated in the experimental conditions.
The study of drug reinforcement is in essence the study of the motivational properties of drugs and hence the study of drug addiction. The development of better experimental methods and a full understanding of existing techniques is an important step in elucidating the mechanisms of drug addiction, the development of new, less addictive compounds, and even the identification of basic neural processes involved in motivation and reward.
Some experimental techniques have a firmer empirical basis than others, but the most compelling conclusions are drawn from studies where several different experimental methods produce similar findings. In general, methods of assessing drug reinforcement that have been adopted across several laboratories and/or have been used for some time have obvious merit documented by their widespread application. Although ease of implementing the procedure is not a substitute for a validated measure of drug reinforcement, most currently used procedures are useful in some situations and possess a certain degree of inherent validity.
This book has attempted to provide a compendium of the most important methods used to assess drug reinforcement. The individual chapters were written by authors who have extensive experience with these techniques. There is a mixture of senior investigators involved in the early development of these methods and of new scientists who are further developing and exploiting these experimental techniques. Many chapters in the book summarize considerable work with a given method, while others present original data illustrating how the techniques are actually applied. The inclusion of diverse experimental methods and of both laboratory animal and human assessment techniques appears to be unique to this volume. The topics and format were selected to provide a thorough understanding of the principles and applications of the methods of assessing the reinforcing properties of abused drugs.
Preparation of this chapter was facilitated by a grant from the National Institute on Drug Abuse (U.S.A.) and by a University Research Fellowship from the Natural Sciences and Engineering Research Council of Canada.
American heritage dictionary. (1969). New York: American Heritage Publishing Company.
Atalay, J., & Wise, R.A. (1983). Time course of pimozide effects on brain stimulation reward. Pharmacology Biochemistry & Behavior18: 655-658.
Bozarth, M.A. (1978). Intracranial self-stimulation as an index of opioid addiction liability: An evaluation. Unpublished master’s thesis, Rensselaer Polytechnic Institute, Troy, NY.
Bozarth, M.A. (1982). The neural substrate of opiate reward in the rat. Unpublished doctoral dissertation, Concordia University, Montréal.
Bozarth, M.A. (1983). Opiate reward mechanisms mapped by intracranial self-administration. In J.E. Smith and J.D. Lane (Eds.), The neurobiology of opiate reward processes (pp. 331-359). Amsterdam: Elsevier/North Holland Biomedical Press.
Bozarth, M.A. (1985). Biological basis of cocaine addiction. In C.J. Brink (Ed.), Cocaine: A symposium (pp. 32-36). Madison, WI: Wisconsin Institute on Drug Abuse.
Bozarth, M.A. (1986). Neural basis of psychomotor stimulant and opiate reward: Evidence suggesting the involvement of a common dopaminergic system. Behavioural Brain Research 22: 107-116.
Bozarth, M.A., & Wise, R.A. (1981). Intracranial self-administration of morphine into the ventral tegmental area. Life Sciences 28: 551-555.
Bozarth, M.A., & Wise, R.A. (1982). Localization of the reward-relevant opiate receptors. In L.S. Harris (Ed.), Problems of drug dependence, 1981 (National Institute on Drug Abuse Research Monograph 41, pp. 158-164). Washington, DC: U.S. Government Printing Office.
Bozarth, M.A., & Wise, R.A. (1983). Neural substrates of opiate reinforcement. Progress in Neuro-Psychopharmacology & Biological Psychiatry 7: 569-575.
Bozarth, M.A., & Wise, R.A. (1986). Involvement of the ventral tegmental dopamine system in opioid and psychomotor stimulant reinforcement. In L.S. Harris (Ed.), Problems of drug dependence, 1985 (National Institute on Drug Abuse Research Monograph 67, pp. 190-196). Washington, DC: U.S. Government Printing Office.
Britt, M.D., & Wise, R.A. (1983). Ventral tegmental site of opiate reward: Antagonism by a hydrophilic opiate receptor blocker. Brain Research258: 105-108.
Broekkamp, C.L.E. (1976). The modulation of rewarding systems in the animal brain by amphetamine, morphine, and apomorphine. Druk, The Netherlands: Stichting Studentenpers Nijmgen.
Broekkamp, C.., Phillips, A.G., & Cools, A.R. (1979). Facilitation of self-stimulation behavior following intracerebral microinjection of opioids into the ventral tegmental area. Pharmacology Biochemistry & Behavior 11: 289-295.
Broekkamp, C.L.E., Pijnenburg, A.J.J., Cools, A.R., & van Rossum, J.M. (1975). The effect of microinjections of amphetamine into the neostriatum and the nucleus accumbens on self-stimulation behavior. Psychopharmacologia42: 179-183.
Broekkamp, C.L., van den Bogaard, J.H., Heynen, H.J., Rops, R.H., Cools, A.R., & van Rossum, J.M. (1976). Separation of inhibiting and stimulating effects of morphine in self-stimulation behavior by intracerebral microinjections. European Journal of Pharmacology 36: 443-446.
de Wit, H., & Stewart, J. (1981). Reinstatement of cocaine-reinforced responding in the rat. Psychopharmacology 75: 134-143.
Esposito, R., & Kornetsky, C. (1978). Opioids and rewarding brain stimulation. Neuroscience and Biobehavioral Reviews 2: 115-122.
Goeders, N.E., Lane, J.D., & Smith, J.E. (1984). Self-administration of methionine enkephalin into the nucleus accumbens. Pharmacology Biochemistry & Behavior 20: 451-455.
Goeders, N.E., & Smith, J.E. (1983). Cortical dopaminergic involvement in cocaine reinforcement. Science 221: 773-775.
Glick, S.D., Cox, R.D., & Crane, A.M. (1975). Changes in morphine self-administration and morphine dependence after lesions of the caudate nucleus in rats. Psychopharmacology 41: 219-224.
Glick, S.D., & Marsanico, R.G. (1975). Time-dependent changes in amphetamine self-administration following frontal cortex ablations in rats. Journal of Comparative and Physiological Psychology 88: 355-359.
Glimcher, P.W., Giovino, A.A., Margolin, D.H., & Hoebel, B.G. (1984). Endogenous opiate reward induced by an enkephalinase inhibitor, thiorphan, injected into the ventral midbrain. Behavioral Neuroscience 98: 262-268.
Griffiths, R.R., & Balster, R.L. (1979). Opioids: Similarity between evaluations of subjective effects and animal self-administration results. Clinical Pharmacology and Therapeutics 25: 611-617.
Hand, T.H., & Franklin, K.B.J. (1985). 6-OHDA lesions of the ventral tegmental area block morphine-induced but not amphetamine-induced facilitation of self-stimulation. Brain Research 328: 233-241.
Hernandez, L.L., Holohean, A.M., & Appel, J.B. (1978). Effects of opiates on the discriminative stimulus properties of dopamine agonists. Pharmacology Biochemistry & Behavior 9: 459-463.
Hoebel, B.G., Monaco, A.P., Hernandez, L., Aulisi, E.F., Stanley, B.G., & Lenard, L. (1983). Self-injection of amphetamine directly into the brain, Psychopharmacology 81: 158-163.
Jaffe, J.H. (1975).Drug addiction and drug abuse. In L.S. Goodman & A. Gilman (Eds.), The pharmacological basis of therapeutics (pp. 284-324). New York: MacMillan.
Kornetsky, C., & Esposito, R.U. (1979). Euphorigenic drugs: Effects on the reward pathways of the brain. Federation Proceedings 38: 2473-2476.
Lyness, W.H., Friedle, N.M., & Moore, K.E. (1979). Destruction of dopaminergic nerve terminals in nucleus accumbens: Effect on d-amphetamine self-administration. Pharmacology Biochemistry & Behavior 11: 553-556.
Olds, M.E. (1979). Hypothalamic substrate for the positive reinforcing properties of morphine in the rat. Brain Research 168: 351-360.
Olds, M.E. (1982). Reinforcing effects of morphine in the nucleus accumbens. Brain Research 237: 429-440.
Oxford English dictionary. (1933). Oxford: Oxford University Press.
Oxford English dictionary (supplement). (1933). Oxford: Oxford University Press.
Phillips, A.G., & LePiane, F.G. (1980). Reinforcing effects of morphine microinjection into the ventral tegmental area. Pharmacology Biochemistry & Behavior 12: 965-968.
Phillips, A.G., & LePiane, F.G. (1982). Reward produced by microinjection of (D-ala 2), Met-enkephalinamide into the ventral tegmental area. Behavioural Brain Research 5: 225-229.
Phillips, A.G., Mora, F., & Rolls, E.T. (1981). Intracerebral self-administration of amphetamine by rhesus monkeys. Neuroscience Letters 24: 81-86.
Reid, L.D., & Bozarth, M.A. (1978). Addictive agents and pressing for intracranial stimulation (ICS): The effects of various opioids on pressing for ICS. Problems of Drug Dependence, 729-741.
Roberts, D.C.S., Corcoran, M.E., & Fibiger, H.C. (1977). On the role of the ascending catecholaminergic systems in intravenous self-administration of cocaine. Pharmacology Biochemistry & Behavior 6: 615-620.
Roberts, D.C.S., and Koob, G.F. (1982). Disruption of cocaine self-administration following 6-hydroxydopamine lesions of the ventral tegmental area in rats, Pharmacology Biochemistry & Behavior 17: 901-904.
Roberts, D.C.S., Koob, G.F., Klonoff, P., and Fibiger, H.C. (1980). Extinction and recovery of cocaine self-administration following 6-hydroxydopamine lesions of the nucleus accumbens. Pharmacology Biochemistry & Behavior12: 781-787.
Spealman, R.D., & Goldberg, S.R. (1978). Drug self-administration by laboratory animals: Control by schedules of reinforcement. Annual Reviews of Pharmacology & Toxicology 18: 313-339.
Spyraki, C., Fibiger, H.C., & Phillips, A.G. (1982). Dopaminergic substrates of amphetamine-induced place preference conditioning. Brain Research 253: 185-192.
Spyraki, C., Fibiger, H.C., & Phillips, A.G. (1983). Attenuation of heroin reward in rats by disruption of the mesolimbic dopamine system. Psychopharmacology79: 278-283.
Stewart, J. (1984). Reinstatement of heroin and cocaine self-administration behavior in the rat by intracerebral application of morphine in the ventral tegmental area. Pharmacology Biochemistry & Behavior 20: 917-923.
Tatum, A.L., & Seevers, M.H. (1929). Experimental cocaine addiction. Journal of Pharmacology and Experimental Therapeutics 36: 401-410.
Tatum, A.L., & Seevers, M.H. 1931). Theories of drug addiction. Physiological Reviews 11: 107-121.
Tatum, A.L., Seevers, M.H., & Collins, K.H. (1929). Morphine addiction and its physiological interpretation based on experimental evidences. Journal of Pharmacology and Experimental Therapeutics 36: 447-475.
Vaccarino, F.J., Bloom, F.E., & Koob, G.F. (1985). Blockade of nucleus accumbens opiate receptors attenuates intravenous heroin reward in the rat. Psychopharmacology 86: 37-42.
van Ree, J.M., & de Wied, D. (1980). Involvement of neurohypophyseal peptides in drug-mediated adaptive responses. Pharmacology Biochemistry & Behavior 13 (Suppl. 1): 257-263.
Webster’s third new international dictionary. (1981). Springfield, MA: Merrian-Webster.
Wise, R.A., & Bozarth, M.A. (1982). Action of drugs of abuse on brain reward systems: An update with specific attention to opiates. Pharmacology Biochemistry & Behavior 17: 239-243.
Wise, R.A., & Bozarth, M.A. (1984). Brain reward circuitry: Four elements "wired" in apparent series. Brain Research Bulletin12: 203-208.
Zito, K.A., Vickers, G., & Roberts, D.C.S. (1985). Disruption of cocaine and heroin self-administration following kainic acid lesions of the nucleus accumbens. Pharmacology Biochemistry & Behavior23: 1029-1036.
Click here to enter the Addiction Science Network Discussion Forum
|© 1999-2009 Addiction Science Network This page was last revised 28 March 2009 12:35 EDT. Send comments to: feedback@AddictionScience.net Report technical problems to: webmaster@AddictionScience.net|