Behavioral research has contributed to our understanding of many of the factors involved in drug abuse, including initiation, maintenance, cessation, and relapse. Prior to the 1960s, the general belief held by professionals and lay people was that drug abuse was caused by an underlying psychopathology that could be studied only in humans. Behavioral researchers, however, took advantage of the knowledge gained about the control of appetitive behaviors and developed an animal model of drug abuse. Although early work on drug abuse and drug-taking behaviors assumed that only those animals1 already physically dependent on opiates could be induced to take them (Thompson and Schuster, 1964), it soon became clear that when drugs were made available, drug-naive animals took them readily and to excess.
This chapter highlights some of the major accomplishments in behavioral research (including the development of behavioral models) and discusses opportunities for future research. Insights from behavioral research have made major contributions to our understanding of the addictive process, enabling researchers to study the behavior of drug taking separately from its pharmacological sequelae and making it possible to integrate the findings of other research disciplines (e.g., treatment and neurosciences).
The major contribution of behavioral research to the study of drug abuse has been the development of the self-administration model and the use of this model to test for abuse liability and to expand our understanding of addiction. This basic model has been augmented by other models based on the principles of learning and conditioning such as drug classification (drug discrimination); the relationship between drug use and variables controlling use (behavioral economics); the nature of transition states in drug abuse (initiation, abstinence, withdrawal); motivational states (e.g., incentive motivation); and the roles of tolerance and physical dependence in drug-seeking behavior.
Drug Self-Administration Model
The drug self-administration model is based on the learning principle that behavior is maintained by its consequences, called reinforcers. Laboratory animals (humans and nonhumans) will work to receive a range of different drugs administered orally, intramuscularly, intravenously, by smoking, or by insufflation. In this model, the laboratory animal performs some action, such as depressing a lever, to trigger the administration of a drug (e.g., through an indwelling catheter or a solution to drink). In general, those drugs (e.g., cocaine, heroin, nicotine, alcohol) that maintain drug taking in nonhumans are also commonly abused by humans, and those that are avoided by humans (e.g., antipsychotics) are also avoided by nonhumans. These results are replicable in virtually every species tested with the model and with different routes of administration. Such findings brought into question the traditional explanations of the etiology of drug abuse, such as psychopathology or various social deprivations.
This model also allows behavioral researchers to control past history and current environmental conditions, thus demonstrating that it is the interaction of the drug's pharmacological effects with past history and current environmental conditions (i.e., setting) that determines whether sampling an abusable drug will proceed to persistent use or abuse (e.g., Barrett and Witkin, 1986). This model points to the importance of a confluence of variables in drug-taking behavior and has broadened the clinician's understanding of the various causal factors that might be involved in drug abuse.
The drug discrimination paradigm is considered a model of the subjective effects of drugs in humans. In this paradigm, research subjects are
trained to respond differently to the test drugs (e.g., drug versus placebo or drug versus drug). For example, a research subject might be trained to press the left lever after a dose of amphetamine and the right lever after a dose of placebo. After training, research subjects (nonhuman or human) will respond differentially to drug and placebo, allowing for comparison among drugs and for conclusions about pharmacological and behavioral similarity, depending on the manner in which the trained research subject responds.
Animal Models of Drug Dependence
Drug dependence has also been modeled in laboratory animals. Drug dependence (or addiction), as noted in Chapter 1, is characterized in both the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; APA, 1994) and the International Classification of Diseases (ICD-10, WHO, 1992) as drug-seeking behavior involving compulsive use of high doses of one or more drugs, for no clear medical indication. Dependence is usually accompanied by tolerance and withdrawal; physicians often confuse the presence of a withdrawal syndrome (i.e., physical dependence) with the compulsive drug taking that is a part of the behavioral dependence syndrome. Models have been developed in which animals are maintained on specific drugs of abuse (e.g., opiates) for some period of time, either via self-administered or experimenter-administered drug, and then observed for the effects of abrupt cessation (e.g., Woods and Schuster, 1968). Manipulations using animal models have provided information about the relationship between repeated drug use and toxicity, as well as the likelihood that the drug will be taken in the future.
In addition to being a useful tool for investigating basic biobehavioral mechanisms underlying drug abuse, the drug self-administration model has provided the foundation for research in many other areas of drug abuse. For example, it has been shown that there is a significant positive correlation between the potencies of cocaine (and other stimulants) as dopamine reuptake blockers and their ability to maintain self-administration behavior, although the same is not true for norepinephrine and serotonin (Ritz et al., 1987; Bergman et al., 1989). This finding suggests that the action of cocaine at its binding site, which results in dopamine uptake blockade, mediates the effects that contribute to abuse (Fischman and Johanson, 1996).
Neuroscientists have taken advantage of this model to investigate the brain loci mediating the reinforcing and dependence-producing properties of morphine (Bozarth and Wise, 1984), the dopaminergic contributions to drug reinforcement, and the brain areas activated by specific drugs (Koob and Bloom, 1989; Cerruti et al., 1994; Nestler, 1994). Geneticists have used this technology to evaluate the heritability of drug abuse (e.g., Froelich et al., 1988); similarly, neurochemists have examined specific behavioral correlates in this model (Kalivas and Duffy, 1993; see Chapter 3).
Excitatory Amino Acids
Research on phencyclidine (PCP) provides a good example of the way in which behavioral studies provide a body of data for understanding the neural basis of learning and memory, as well as the development of novel medication strategies. In the early 1970s, the introduction of PCP as a drug of abuse was immediately recognized as different and potentially more devastating than abuse of other hallucinogens. Initial studies evaluated this drug and its analogues in self-administration and drug discrimination paradigms (reviewed in Balster and Willetts, 1996). It became obvious that PCP was a noncompetitive antagonist at the NMDA (N-methyl-D-aspartate) receptor. PCP became an important research tool for understanding the role of excitatory amino acid neurotransmission initiated by glutamate in the control of a variety of behaviors and in the pathophysiology of neuronal death.
Current work in this area has the potential to lead to novel treatment medication strategies for preventing neurotoxicity following brain trauma. As excitatory amino acid antagonists are developed for therapeutic uses, an important goal will be to avoid the abuse liability (the likelihood that a drug will be abused) and psychological disturbances produced by PCP; the animal models developed by drug abuse researchers are now being relied on in this area of medications development. An exciting research development suggests that excitatory amino acids may play an important role in the development of tolerance to and dependence on drugs of abuse such as the opiates, alcohol, and stimulants (Balster and Willetts, 1996). It is possible that this research will lead to completely novel strategies for the treatment of the addictions.
Development of Therapeutic Drugs Without Abuse Liability
A major concern in the development of new psychotropic medications is to maximize therapeutic efficacy while reducing the risks of abuse and dependence. In the 1920s, substitutes for morphine were sought that
lacked abuse liability. That research measured physical dependence and focused on the withdrawal syndrome by collecting behavioral and physiological data on both objective and self-reported measures. Work in that area led to the important discovery that physical dependence and abuse liability were not the same and that abuse liability could not be assessed solely on the basis of chemical structure.
The drug self-administration model continues to provide a bioassay for the evaluation of abuse liability. Animal self-administration studies have been used widely to predict the abuse liability of new drugs (Brady and Lukas, 1984). Self-administration data are frequently a part of the information submitted to the Food and Drug Administration (FDA) by pharmaceutical companies as part of their applications for approval of psychoactive drugs, including those targeted for psychiatric disorders (e.g., anxiety, depression). Those data are used by the FDA (and ultimately the Drug Enforcement Administration) in their scheduling recommendations.2 Behavioral assays using this model provide critical data for determining the appropriate regulatory status of drugs, since such determinations cannot be made simply on the basis of chemical structure or in vitro data. Because the particular schedule in which a drug is placed strongly influences the marketing success of the new compound, the pharmaceutical industry has been a major supporter of behavioral research. In fact, different preparations of the same medications, with the same active chemical constituent, are often regulated and scheduled differently based on their behavioral effects. For example, nicotine gum is unscheduled and sold over the counter; approval is pending for over-the-counter sale of nicotine patches; however, nicotine nasal spray may be placed in Schedule IV or V.
Medications for the Treatment of Drug Abuse
The drug self-administration model has been of major importance in the search for potentially useful pharmacological interventions to treat drug abusers. Early research, for example, demonstrated the efficacy of immunizing rhesus monkeys with an antigen that caused the formation of antibodies that bound morphine when it was injected intravenously (Bonese et al., 1974). Rates of heroin self-administration decreased almost to zero in immunized animals, although the toxicity of the procedure
limited its utility in humans. Other researchers, using monoclonal antibody techniques, recently reported the development of an artificial enzyme that inactivates cocaine by cleaving it into two inactive metabolites (Landry et al., 1993). This technique is effective in the test tube, but it must now be demonstrated in nonhumans before it proceeds to human trials; researchers are pursuing this work in conjunction with behavioral researchers experienced in drug self-administration research and medications development (J. Woods, University of Michigan, personal communication, 1995). Further evidence for the promising nature of immunopharmacotherapy is given in a recent report (Carrera et al., 1995) describing suppression of locomotor activity and stereotyped behavior in rats after active immunization with a cocaine immunogen. This response was specific to cocaine and was not seen after amphetamine administration.
Administration of antagonists or immunization against specific drugs, although clearly potentially important tools in our armamentarium against drug use (see discussion on behavioral economics, below), promises no more success than the available opiate antagonist naltrexone 3 for the treatment of heroin addiction. It is very clear that nonhumans, treated with naltrexone, will show extinction in their opiate responding (Koob et al., 1984), and humans, under residential laboratory conditions, also will stop using heroin after treatment with naltrexone (Mello et al., 1981). However, after leaving the structured setting of a residential laboratory individuals relapse to heroin use. Although laboratory studies on naltrexone, with nonhuman and human subjects, demonstrate the utility of the drug self-administration model in the initial assessment of the utility of a new medication, the model does not allow for an evaluation of the contextual (social and environmental) factors that could ultimately affect drugtaking behavior. Thus, there is a need for behavioral models that pattern complex behaviors (e.g., studies that give heroin users the choice of taking naltrexone and explore the range of conditions under which it is taken). Studies focused on compliance are becoming increasingly important because the most efficacious medications are useless if the patient does not take them.
The National Institute on Drug Abuse (NIDA) Medications Development Program relies on drug self-administration and drug discrimination models for its preclinical evaluation of new medications (IOM, 1995; Mello
and Negus, in press). Animal models of self-administration (versus human models) have several advantages for medications development. For example, drugs that are not approved for use in humans can be evaluated; the effects of new treatment medications on patterns of drug self-administration can be evaluated quantitatively under controlled experimental conditions; social factors such as peer pressure or expectancy do not complicate interpretation of data; and accurate baseline measures of the daily dose and patterns of drug self-administration can be determined before, during, and after administration of the treatment medication. Additionally, the safety of the medication can be evaluated continually. Thus, the use of animal models for those aspects of medications development is parallel in importance to the earlier reliance on animal models of drug self-administration for evaluation of the abuse liability of new drugs. To the extent possible, however, these laboratory models should be employed across species to include humans.
Learning and Conditioning
A major contribution of behavioral research has been an understanding of the ways in which basic principles of learning and conditioning can be used to modify drug-taking behavior. These principles have been precisely defined so that they can be studied and replicated across conditions and species.
For example, research on drug effect expectancies suggests that learned beliefs and attitudes may serve as risk factors for the initiation and use of drugs (Brown, 1993). Further, epidemiological research has pointed to the importance of social modeling and attitudes as having strong impacts on drug use and abuse. Research on learning and conditioning has led to successful treatment models for drug abusers, including relapse prevention, community reinforcement, and focused techniques such as extinction training, relaxation training, contingency management, and job skills training. Two well-studied behavioral interventions are discussed below: contingency management and relapse prevention.
Contingency management research is based on the fact that, although drugs are potent reinforcers, there are non-drug reinforcers that can compete with drug use (see discussion of behavioral economics, below). Manipulation of the environment can shift the focus toward or away from drug reinforcers (e.g., Azrin et al., 1966; Barrett and Witkin, 1986). In the laboratory, monkeys will choose saccharine over phencyclidine if they are required to work substantially harder for the drug (Carroll and Rodefer, 1993). Research with humans has shown that experienced cocaine users will choose money or tokens over cocaine when the appropriate quantity and quality of alternative reinforcers are available (Foltin and Fischman,
1994; Higgins et al., 1994). In addition, direct reinforcement of drug abstinence can be effective in methadone maintenance programs (Iguchi et al., 1988), and drug consumption can be reduced significantly when valued alternatives (commodities and recreational activities) are provided in exchange for clean urine (Higgins et al., 1993).
Relapse prevention research also combines cognitive and behavioral approaches (e.g., Marlatt and Gordon, 1985; Carroll KM et al., 1991). Behavioral analysis of drug abusers has demonstrated that learning and conditioning (both classical and operant) play an important role in the initiation, maintenance, cessation, and relapse to drug use. Early work in rats (Wikler and Pescor, 1967) and in humans (O'Brien et al., 1977) showed that signs of abstinence can become classically conditioned to the specific environmental conditions under which withdrawal has occurred in the past. Thus, even though a previously opiate-dependent person has remained drug free for a prolonged period of time, specific environmental conditions could trigger opiate withdrawal symptoms, which in turn might motivate relapse. This effect has been modeled in the laboratory, where rhesus monkeys, dependent in the past on morphine, showed clear signs of physical dependence and relapse in the presence of stimuli that in the past signaled opiate withdrawal (Goldberg and Schuster, 1967, 1969; Goldberg et al., 1970, 1971). Conditioned opiate withdrawal and craving have also been demonstrated experimentally in humans (O'Brien et al., 1977; Childress et al., 1988). Although Wikler (1973), for example, observed that conditioned withdrawal plays a substantial role in relapse to opiate use, even years after the drug-dependent person has ceased using opiates, the role of conditioned responses in relapse in the nonlaboratory setting is not yet clear.
Drug Administration and Withdrawal
A variety of behavioral studies have been used to characterize and quantitate the potential deleterious effects of drug administration and withdrawal for both illicit and licit drug use and have been useful in guiding policy development. For example, while cigarette smoking has long been associated with increased alertness, sustained performance in situations of fatigue, and increased cognitive performance (Rusted and Warburton, 1992), dependent individuals experience decreases in performance stemming from nicotine withdrawal. These decrements are reversed rapidly by the re-administration of tobacco or medically approved forms of nicotine such as nicotine gum or patch (Henningfield, 1994). Characterizing the course and timing of this behavioral degradation has been critical in determining how to manage nicotine-dependent airline pilots. Since performance decrements do not emerge until approximately
four hours after the last cigarette, the prohibition of smoking by pilots would not be expected to compromise their performance on flights of two hours or less, and non-disrupted pilot behavior might be sustained by nicotine-delivering medications in the absence of tobacco (Fiore et al., 1994). Those findings point the way to research with other drugs of abuse where similar effects may cause a reluctance to stop use.
Behavioral research has revealed the complexity of drug use and has shown that the conditions under which drugs are used may be as powerful in motivating drug use as the drug itself. It has been observed that even noxious stimuli in the environment will serve as reinforcers and result in the self-administration of drugs. Thus, the concept of motivation—why people use drugs—is far more complicated than initially believed and may include adjunctive behaviors, non-drug reinforcers, appetitive behaviors, and single priming doses.
Although drug abuse is frequently described as a direct consequence of exposure to a drug with abuse liability, the great majority of people experimenting with such drugs do not become abusers (Anthony et al., 1994). The intrinsic effects produced by certain drugs (i.e., their physiological and subjective effects) can serve as motivating factors for drug abuse, but there is another, less direct, yet powerful way in which drugs can gain control over a person's life. Drug taking can develop as an adjunct to another strongly motivated behavior when that behavior becomes intermittently blocked or cannot be completed. When so blocked, that person may turn to an easy, satisfying alternative—an adjunctive behavior (Falk, 1984, 1993).
In the laboratory, excessive adjunctive behavior (e.g., excessive water drinking, aggression, eating, smoking) is related to the intermittent availability of an important commodity or activity. The conditions or ''generator schedules" under which such behaviors become excessive are similar to conditions in natural and social environments that provide what we need, but only in small amounts, and with delay intervals. The adjunctive behavior generated may be noninjurious (e.g., drinking water) or creative (e.g., an intense hobby), but it also can result in aggression or drug taking, depending upon personal history, skills, and currently available alternatives. Drug abuse can arise from conditions already generating behav-
ioral excesses, and this may be one of the reasons that drug abusers often have other behavioral problems.
Adjunctive behavior studies indicate that drug abuse may stem more from environmental-generating conditions, together with a lack, or poor utilization, of other opportunities, than from any intrinsic attractiveness of drugs with abuse liability. Those studies clarify how drugs can become so attractive to some individuals, and the abuse behavior so persistent, in light of the trouble it causes for them. Understanding the conditions that comprise economically or socially restricted schedules of reinforcement and, therefore, can generate and sustain drug abuse behavior has important implications for the design of therapeutic and prevention strategies. Adjunctive behavior research procedures may also serve as models of limited opportunities in the natural ecology.
Drug use has an obvious effect on motivation. Individuals who, in the absence of drug use, will go to work, support a family, seek an education, and engage in other aspects of a productive life style, can become totally involved in drug seeking and drug taking, neglecting all other activities they previously found rewarding. Behavioral research with human subjects is now focusing on understanding the determinants of the reinforcing effects of drugs in an environment in which alternative reinforcers are available.
Research on the factors controlling the choice to use drugs is best carried out in the laboratory, where multiple behaviors can be measured and manipulated (see Fischman et al., 1991). Such research has established some of the determinants of choice both between drugs and between drugs and non-drug reinforcers. For example, the frequency of drug choice depends on dose (Johanson, 1975; Nader and Woolverton, 1991), as well as on environmental factors such as the availability of a non-drug option (Carroll ME et al., 1991), simultaneous delivery of an adverse consequence (e.g., electric shock) (Johanson, 1975), or increased response requirement for the drug (Nader and Woolverton, 1991).
Research with humans has shown that the choice between cocaine and alternative reinforcers can be a sensitive assay for the efficacy of new medications (e.g., Foltin and Fischman, 1994). The question then becomes whether a specific medication increases the likelihood that drug abusers will choose non-drug rather than drug options, a question much closer to the natural ecology of a treatment setting. This type of research is highly complex, and methodologies for conducting research, analyzing data, and developing theoretical frameworks must be developed in order to elucidate interactions among the organism, the drug, and the environment.
The behavioral economics approach, described below, is one such possibility, but others must also be considered.
Research into other appetitive behaviors may be directly relevant to drug abuse. For example, studies have shown that food deprivation can increase drug intake (Carroll, 1995), but it is not known whether this is true for deprivation of money, social factors, or other commodities. It may well be that research into disorders such as bulimia and anorexia is directly relevant to the study of drug abuse: they have similar topographies, although there are currently no data indicating that their underlying processes are also similar.
Research with nonhumans has repeatedly shown that a single (priming) dose of a drug can reinstate drug-reinforced responses even after the animal has ceased responding to that drug due to extinction (e.g., de Wit and Stewart, 1981; Slikker et al., 1984). This phenomenon is believed to occur in humans as well, which is why most drug abuse treatment programs stress the need for total abstinence. Recent studies have shown that nonalcoholic human research subjects chose an ethanol-containing beverage over money on days when they were pretreated with an ethanolcontaining beverage but not on days when they were pretreated with placebo (de Wit and Chutuape, 1993; Chutuape et al., 1994). These observations support the hypothesis that priming doses of a drug can reinstate drug taking in those who are currently not seeking or taking a drug.
It is possible that priming effects are not specific to drugs but may be a more general phenomenon. For example, food-satiated humans presented with a specific food "prime" will choose to eat more of that food but not of other foods (Cornell et al., 1992). This phenomenon, in the context of incentive motivational theory, suggests that drugs of abuse (specifically heroin and cocaine) produce their motivational effects by acting directly on the central nervous system, and that administration of such drugs has the ability to induce motivation (incentive) for them (Stewart et al., 1984). It has been suggested that after repeated use, the conditioned incentive effect of a drug can mimic its neural activity, which can initiate drug-taking behavior. Research in this area has the potential to provide information about the neural substrates underlying appetitive motivation and offers opportunities for integrated behavioral and neurobiological research into the mechanisms underlying relapse.
Patients seeking treatment for their drug use report irresistible craving that leads to continued use, despite firm resolutions to remain abstinent. From the perspective of the drug abuser, this motivation to continue use is a major impediment to abstinence. Drug treatment specialists refer to drug craving by their patients as an important determinant of success or failure in treatment. Despite the clear importance of this concept from a descriptive perspective, it is an extremely difficult concept to measure in the laboratory (or the clinic) in a meaningful fashion. Clinicians try to gauge the efficacy of treatment interventions by assessing changes in reports of drug craving by their patients. Studies attempting to correlate reported craving with actual use of cocaine in a laboratory setting have not been successful. Under some conditions, use remained unchanged as reported craving decreased (Fischman et al., 1990); under other conditions, use decreased while craving remained unchanged (Foltin and Fischman, 1994). A similar dissociation has been found in tobacco smokers: nicotine chewing gum decreased tobacco intake, but measures of the desire to smoke were unaffected (Nemeth-Coslett and Henningfield, 1986). Studies of heroin addicts found that craving increased under conditions of precipitated withdrawal, but the choice of self-administering an opiate was unaffected (Schuster et al., 1995).
Conditioned craving has been reported in former opiate addicts (O'Brien, 1975) and in former cocaine addicts (O'Brien et al., 1990) when presented with stimuli associated with prior drug use. This phenomenon has been used to screen medications for potential anticraving activity (Robbins et al., 1992; Berger et al., 1996). Reports of craving in response to drug-related stimuli have been accompanied by significant changes in skin temperature, skin resistance, pulse, and other autonomic measures. Recently, specific limbic system activation has been noted using oxygen15 PET (positron-emission tomography) measures of regional cerebral blood flood (Childress et al., 1995). Although the relationship of the conditioned craving phenomenon to actual relapse is unclear, it has been possible to demonstrate an effect on drug-taking behavior in nicotine-dependent smokers. Droungas and colleagues (1995) demonstrated that smoking cues provoked craving for cigarettes and a reduced latency to smoke in smokers who were not aware that they were being observed.
In the context of a cocaine treatment intervention, reports of craving during the first week of treatment were only weakly predictive of treatment outcome, and changes in craving over the course of the treatment intervention were uncorrelated with success in abstaining from cocaine use (S.T. Higgins, University of Vermont, personal communication, 1995). Data collected thus far suggest that craving and the increased probability
of drug self-administration may both be related to other variables (e.g., withdrawal states, drug-associated environmental cues) but that craving is not causally related to increased drug taking. In fact, one author has suggested that drug use may be mediated by processes different from those mediating craving and that craving may represent the cognitive "battle" going on in the drug abuser related to whether or not to seek and take the drug (Tiffany, 1990). Based on these insights, a new multi-item questionnaire has been developed that might more accurately reflect the multidimensional aspect of what drug users are reporting, thus better predicting treatment-related behavior (Tiffany et al., 1993). At this time, although attempts to understand craving may be important clinically, it seems unwarranted to employ drug craving as a surrogate measure of drug self-administration.
Patients frequently report craving that is associated with increased thoughts about drugs and drug use. Modification or alleviation of those thoughts, while not resulting in abstinence, may well shift their focus away from drug seeking and drug taking toward more acceptable behaviors that are in keeping with the goals of a treatment program. Because of the belief, shared by patients and clinicians, that craving has a major impact on relapse, craving should be studied further. An understanding of the nature of craving, what it is and how it impacts behavior, is an important opportunity for drug abuse research.
Violence and Aggression
Although illicit drugs (particularly cocaine) have been associated with a dramatic upsurge in violence in the United States, it has been difficult to attribute causality to the pharmacological effects of the drugs being used (see Chapter 7). Animal models have shown that acute cocaine administration enhances aggressive responding (Miczek et al., 1994), but the majority of data on humans has been based on epidemiological rather than experimental findings. The ethical issues involved in actually engendering violent or aggressive behavior are formidable, and substantial creativity is required to design ethically acceptable and valid models.
One such laboratory model has demonstrated differential effects related to the drug being tested (Cherek, 1981; Cherek et al., 1991). In this model, each research participant is told that responding on one lever will earn points exchangeable for money, whereas responding on the second lever will subtract points from an unseen research participant in another location. The person making this latter (aggressive) response gains no points by making the response and, in fact, there is no second person. Although the model appears to have face validity, it has not undergone rigorous testing to verify that it models aggressive behavior. Recent ef-
forts to verify the model by studying prisoners found guilty of either violent or nonviolent crimes suggest that the model shows behavioral differences in these two subpopulations (D. Cherek, University of Texas, personal communication, 1995). Continued research is needed to develop models of aggression in animals and humans.
Vulnerability to Drug Use
Animal models of drug self-administration have identified factors that facilitate the addiction process. For example, genetic strain (e.g., George and Goldberg, 1989) and individual differences in activity level (e.g., Piazza et al., 1993) can predict vulnerability to repetitive drug use. Environmental conditions such as lack of alternative reinforcers, restricted access to food (e.g., Carroll, 1995), and drug history (e.g., Horger et al., 1991) can accelerate the onset of drug self-administration. Behavioral studies to delineate those facilitating variables in animals may lead to data-based programs for targeting high-risk human populations, making education and prevention efforts more focused and presumably more effective.
Research in the area of etiology has focused on risk factors, with the underlying assumption that some drug use is pathological. However, one of the messages from animal research using the self-administration model is that drugs easily serve as reinforcers and that conditions do not need to be pathological for drugs to be repeatedly self-administered by all animals. Research in the area of neurobiology is beginning to demonstrate that drug-taking behavior is controlled by brain mechanisms developed through evolution to ensure the reinforcing effects of biologically essential activities of eating, drinking, and copulating. The implication of these research findings is that, were it not for countervailing influences, drug use would be the norm, not an aberration. That inference may be somewhat strong, since there are individual differences in those brain systems that contribute to vulnerability, but it points to a research effort in prevention that takes into account the biological foundations against which these efforts are made.
For example, a variety of environmental risk factors can affect responsivity to drugs, including personality, family, and peer influences. Etiological research has identified issues of interest in these areas, including questions related to risk taking, impulsivity, and deviance (see Chapter 5). Those areas have received attention from behavioral researchers in other contexts, and it should be possible to adapt existing models or to develop new ones with direct relevance to drug abuse and dependence. One etiological hypothesis that might be tested with these models is drug effect expectancy. It has been hypothesized that individuals learn about
the effects of drugs directly or from other sources (e.g., media, others), creating a memory network (i.e., an expectancy) that can be activated by drug-related cues. These cognitively developed networks are associated with a drug or drugs and potentially can change as new learning occurs. Use of this construct has important implications for treatment, since it implies that changes in cognition can result in changes in drug abuse behavior. Little research has been carried out in this area, with most of it currently concentrated on alcohol use, and the paradigms for implementing and measuring these changes are not yet well developed. As the role of expectancies in the development and maintenance of drug abuse is delineated more clearly, procedures for preventing or changing drug effect expectancies may well be a useful aspect of a more general cognitivebehavioral approach to drug abuse treatment.
Drug self-administration models have been developed to provide finer-grained analyses of the dynamic interplay among variables. One example is the application of behavioral economics, which focuses on concepts from consumer demand theory. This approach provides a way of understanding the relationship between consumption (i.e., drug use or self-administration) and variables such as price, income, and the characteristics of the goods to be consumed. The utility of this approach stems from its ability to integrate the effects of multiple independent variables (e.g., unit price and the conditions under which the subject chooses reinforcers) into a single term—elasticity of demand4—that may be used to better understand the consequences of various treatments.
Drug taking occurs in the context of multiple interacting and competing reinforcers. Some are directly related, in that the consumption of one leads to the consumption of others (e.g., cigarettes and alcohol), whereas some show an inverse relationship (e.g., involvement in some religious activities and drug use). It has been generally assumed that providing individuals, particularly children and adolescents, with alternative sources of reinforcement will decrease the use of drugs. Behavioral economic analysis can use the concept of cross-price elasticity to quantify precisely whether each reinforce acts as a substitute (a viable alternative) for, serves as a complement (a promoter) to, or is independent (ineffective) of the one against which it is compared. Making sense of the relationships among qualitatively different reinforcers will give us the ability
Defined here as the degree of responsiveness of drug consumption to changes in price (see Chapter 10).
to suggest concrete ways in which drug use can be affected in both animals and humans (Bickel et al., 1993; Carroll, 1993).
The concepts of elasticity and cross-elasticity and the behavioral methods that have been developed to study drug taking from this perspective can also be utilized in the development of medications. The use of behavioral economics could provide information about whether a specific medication, in addition to directly decreasing drug use, also increases sensitivity to other factors of the treatment regimen (e.g., counseling). The opportunity to screen potential new medications by using this conceptual approach, as well as the objective measure of drug self-administration, expands the utility of laboratory analyses and makes them invaluable screening procedures prior to the use of uncontrolled and expensive clinical trials.
CONCLUSION AND RECOMMENDATION
In summary, there are opportunities for continued progress in behavioral approaches to drug abuse research. Behavioral models are useful for developing drug abuse pharmacotherapies; improving treatment engagement and compliance; developing novel procedures for both strengthening weak positive behaviors and attenuating strong drug-related behaviors; addressing questions related to mechanisms of craving and relapse; and promoting better understanding of drug use over the life span of drug users. Increased understanding of various drugs' mechanisms of action can also lead to better understanding of behavior and of vulnerability to drug abuse, which may not be elucidated with familial and drug use histories. The continued development of behavioral models is necessary to improve integration of data and variables being studied. To this end, combining neurobiological and behavioral models should be a primary research goal of the future.
Identification of the mechanisms by which drugs produce behavioral effects is important in the development of new treatment approaches, especially medications development. Behavioral assays, including drug self-administration by animals and humans, as well as subjective effects assessment in humans, are the cornerstone of medications development research because they enable efficient means of screening new chemicals that are highly predictive in their effects on human drug taking. The opportunities for future behavioral research are in the continued development and utilization of those behavioral assays of drug effects.
Although current research into the development of new medications for the treatment of drug abuse is important, drug abuse will likely continue to be treated with a combination of treatment modalities (i.e., behavioral and pharmacological interventions). Research on the modifica-
tion of drug-taking behavior must continue, with special emphasis on the use of alternative reinforcers. Outcomes should be analyzed with sophisticated approaches that integrate the effects of multiple variables (e.g., behavioral economic analyses). This area of research has only begun to be applied within the clinic, and more sophisticated interventions must be developed based on carefully collected data.
Although a number of animal models have been developed, the use of behavioral models with human participants is a necessary step in expanding the field of drug abuse research. The models being developed should combine a range of behavioral approaches including conditioning, social learning, and cognitive models, integrating them to emulate most effectively the complex behaviors represented by the various aspects of drug seeking and taking. Controlled environment research, behavioral economic analyses, and vulnerability studies are a first step toward addressing some of the complex behaviors associated with drug abuse.
The committee recommends the use of behavioral models (involving both humans and nonhumans) to further our understanding of the various aspects of drug use, abuse, and dependence (such as initiation, relapse, prolonged abstinence, craving, and transitions from drug use to abuse); to develop improved behavioral and pharmacological interventions for the treatment of drug abuse and dependence; and to inform prevention efforts.
Although research in this area is difficult, it is an important investment, and researchers should be encouraged to explore new paradigms to model complex behaviors related to drug abuse. Importantly, the models must be validated if they are to have utility for the field and should combine a range of behavioral approaches including conditioning, social learning, and cognitive models, integrating them to most effectively emulate the complex behaviors represented by the various aspects of drug seeking and taking.
Anthony JC, Warner LA, Kessler RC. 1994. Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: Basic findings from the National Comorbidity Survey. Experimental and Clinical Psychopharmacology 2:244-268.
APA (American Psychiatric Association). 1994. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: APA.
Azrin NH, Hutchinson RR, Hake DF. 1966. Extinction-induced aggression. Journal of the Experimental Analysis of Behavior 9:191-204.
Balster RL, Willetts J. 1996. Phencyclidine: A drug of abuse and a tool for neuroscience research. In: Schuster CR, Gust SW, Kuhar MJ, eds. Pharmacological Aspects of Drug Dependence: Towards an Integrated Neurobehavioral Approach. Handbook of Experimental Pharmacology 118:233-262.
Barrett JE, Witkin JM. 1986. The role of behavioral and pharmacological history in determining the effects of abused drugs. In: Goldberg SR, Stolerman IP, eds. Behavioral Analysis of Drug Dependence. Orlando: Academic Press. Pp. 195-224.
Berger SP, Hall S, Mickalian JD, Reid MS, Crawford CA, Delucchi K, Carr K, Hall S. 1996. Haloperidol antagonism of cue-elicited cocaine craving. Lancet 347:504-508.
Bergman J, Madras BK, Johnson SE, Spealman RD. 1989. Effects of cocaine and related drugs in nonhuman primates. Part III. Self-administration by squirrel monkeys. Journal of Pharmacology and Experimental Therapeutics 251:150-155.
Bickel WK, Degrandpre RJ, Higgins ST. 1993. Behavioral economics: A novel experimental approach to the study of drug dependence. Drug and Alcohol Dependence 33:173-192.
Bonese KR, Wainer BH, Fitch FW, Rothberg RM, Schuster CR. 1974. Changes in heroin self-administration by rhesus monkey after morphine immunization. Nature 252:708-710.
Bozarth MA, Wise RA. 1984. Anatomically distinct opiate receptor fields mediate reward and physical dependence. Science 224:516-517.
Brady JV, Lukas SE. 1984. Testing Drugs for Physical Dependence Potential and Abuse Liability. NIDA Research Monograph 52. Rockville, MD: NIDA.
Brown SA. 1993. Drug effect expectancies and addictive behavior change. Experimental and Clinical Psychopharmacology 1:55-67.
Carrera MRA, Ashley JA, Parsons LH, Wirsching P, Koob GF, Janda KD. 1995. Suppression of psychoactive effects of cocaine by active immunization. Nature 378:727-730.
Carroll KM, Rounsaville BJ, Keller DS. 1991. Relapse prevention strategies for the treatment of cocaine abuse. American Journal of Drug and Alcohol Abuse 17:249-265.
Carroll ME. 1993. Drugs and Environment. Netherlands: Open University Press.
Carroll ME. 1995. Interactions between food and addiction. In: Niesink RJM, Kornet ML, eds. Behavioral Toxicology and Addiction: Food, Drugs and Environment. Netherlands: Open University Press.
Carroll ME, Rodefer JS. 1993. Income alters choice between drug and an alternative nondrug reinforcer in monkeys. Experimental and Clinical Psychopharmacology 1:110-120.
Carroll ME, Carmona GG, May SA. 1991. V. Cocaine self-administration in rats: Effects of non-drug alternative reinforcers on acquisition. Psychopharmacology 110:5-12.
Cerruti C, Pilotte NS, Uhl G, Kuhar MJ. 1994. Reduction in dopamine transporter MRNA after cessation of repeated cocaine administration. Molecular Brain Research 22:132-138.
Cherek DR. 1981. Effects of smoking different doses of nicotine on human aggressive behavior. Psychopharmacology 75:339-345.
Cherek DR, Bennett RH, Grabowski J. 1991. Human aggressive responding during acute tobacco abstinence: Effects of nicotine and placebo gum. Psychopharmacology 104:317-322.
Childress AR, McLellan AT, Ehrman R, O'Brien CP. 1988. Classically conditioned responses in opioid and cocaine dependence: A role in relapse? NIDA Research Monograph 84:2543.
Childress AR, Mozley D, Fitzgerald J, Reivich M, Jaggi J, O'Brien CP. 1995. Limbic activation during cue-induced cocaine craving. Society for Neuroscience Abstracts 21(3):1956.
Chutuape MA, Mitchell S, de Wit H. 1994. Ethanol preloads increase ethanol preference under concurrent random-ratio schedules in social drinkers. Experimental and Clinical Psychopharmacology 2:310-318.
Cornell CE, Rodin J, Weingarten H. 1992. Stimulus-induced eating when satiated. Physiology and Behavior 45:695-704.
de Wit H, Chutuape MA. 1993. Increased ethanol choice in social drinkers following ethanol preload. Behavioural Pharmacology 4:29-36.
de Wit H, Stewart J. 1981. Drug reinstatement of heroin-reinforced responding in the rat. Psychopharmacology 79:29-31.
Droungas A, Ehrman RN, Childress AR, O'Brien CP. 1995. Effect of smoking cues and cigarette availability on craving and smoking behaviors. Addictive Behaviors 20:657-673.
Falk JL. 1984. The environmental generation of excessive behavior. In: Mule SJ, ed. Behavior in Excess: An Examination of the Volitional Disorders. New York: Free Press. Pp. 313-337.
Falk JL. 1993. Schedule-induced drug self-administration. In: Van Haaren F, ed. Methods in Behavioral Pharmacology. Amsterdam: Elsevier. Pp. 301-328.
Fiore M, Shi FY, Heishman SJ, Henningfield JE. 1994. The Effect of Smoking and Smoking Withdrawal on Flight Performance: A 1994 Update. Submitted by the Centers for Disease Control to the Federal Aviation Administration.
Fischman MW, Johanson CE. 1996. Cocaine. In: Schuster CR, Gust SW, Kuhar MJ, eds. Pharmacological Aspects of Drug Dependence: Towards an Integrated Neurobehavioral Approach. Handbook of Experimental Pharmacology 118:159-195.
Fischman MW, Foltin RW, Nestadt G, Pearlson GD. 1990. Effects of desipramine maintenance on cocaine self-administration by humans. Journal of Pharmacology and Experimental Therapeutics 253:760-770.
Fischman MW, Kelly TH, Foltin RW. 1991. Residential laboratory research: A multidimensional evaluation of the effects of drugs on behavior. NIDA Research Monograph 100:113-128.
Foltin RW, Fischman MW. 1994. Effects of buprenorphine on the self administration of cocaine by humans. Behavioural Pharmacology 5:79-89.
Froelich JC, Harts J, Lumen L, Li T-K. 1988. Differences in response to the aversive properties of ethanol in rats selectively bred for oral ethanol preference. Pharmacology, Biochemistry and Behavior 31:215-222.
George FR, Goldberg SR. 1989. Genetic approaches to the analysis of addiction processes. Trends in Pharmacological Sciences 10:78-83.
Goldberg SR, Schuster CR. 1967. Conditioned suppression by a stimulus associated with nalorphine in morphine-dependent monkeys. Journal of the Experimental Analysis of Behavior 10:235-242.
Goldberg SR, Schuster CR. 1969. Nalorphine: Increased sensitivity of monkeys formerly dependent on morphine. Science 166:1548-1549.
Goldberg SR, Woods JH, Schuster CR. 1970. Conditioned nalorphine-induced abstinence changes: Persistence in post morphine-dependent monkeys. Journal of the Experimental Analysis of Behavior 14:33-46.
Goldberg SR, Woods JH, Schuster CR. 1971. Nalorphine-induced changes in morphine selfadministration in rhesus monkeys. Journal of Pharmacology and Experimental Therapeutics 176:464-471.
Henningfield JE. 1994. Comments on West's editorial, ''Beneficial effects of nicotine: fact or fiction?" Addiction 89:135-146.
Higgins ST, Budney AJ, Bickel WK, Hughes JR, Foerg F, Badger G. 1993. Achieving cocaine abstinence with a behavioral approach. American Journal of Psychiatry 150:763-769.
Higgins ST, Bickel WK, Hughes JR. 1994. Influence of an alternative reinforcer on human cocaine self-administration. Life Science 55:179-187.
Horger BA, Wellman PJ, Morien A, Davies BT, Schenk S. 1991. Caffeine exposure sensitizes rats to the reinforcing effects of cocaine. Neuro Report 2:53-56.
Iguchi MY, Stitzer ML, Bigelow GE, Liebson IA. 1988. Contingency management in methadone maintenance: Effects of reinforcing and aversive consequences on illicit polydrug use. Drug and Alcohol Dependence 22:1-7.
IOM (Institute of Medicine). 1995. The Development of Medications for the Treatment of Opiate and Cocaine Addictions. Washington, DC: National Academy Press.
Johanson CE. 1975. Pharmacological and environmental variables affecting drug preference in monkeys. Pharmacology Reviews 27:343-355.
Kalivas PW, Duffy P. 1993. Time course of extracellular dopamine and behavioral sensitization to cocaine. I. Dopamine axon terminals. Journal of Neuroscience 13:266-275.
Koob GF, Bloom FE. 1989. Nucleus accumbens as a substrate for the aversive stimulus effects of opiate withdrawal. Psychopharmacology 98:530-534.
Koob GF, Pettit HO, Ettenberg A, Bloom FE. 1984. Effects of opiate antagonists and their quaternary derivatives on heroin self-administration in the rat. Journal of Pharmacology and Experimental Therapeutics 229:481-486.
Landry DW, Zhao K, Yang GX-Q, Glickman M, Georgiadis TM. 1993. Antibody-catalyzed degradation of cocaine. Science 259:1899-1901.
Marlatt GA, Gordon JR. 1985. Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors. New York: Guilford Press.
Mello NK, Negus SS. In press. Preclinical evaluation of pharmacotherapies for treatment of cocaine and opiate abuse using drug self-administration procedures. Neuropsychopharmacology.
Mello NK, Mendelson JH, Kuehnle JC, Sellers MS. 1981. Operant analysis of human heroin self-administration and the effects of naltrexone. Journal of Pharmacology and Experimental Therapeutics 216:45-53.
Miczek KA, DeBold JF, Haney M, Tidey J, Vivian J, Weerts E. 1994. Alcohol, drugs of abuse, aggression and violence. In: Reis AJ, Miczek KA, Roth JA, eds. Understanding and Preventing Violence: Biobehavioral Perspectives on Violence, Vol. 2. Washington, DC: National Academy Press. Pp. 245-514.
Nader MA, Woolverton WL. 1991. Effects of increasing the magnitude of an alternative reinforcer on drug choice in a discrete-trials choice procedure. Psychopharmacology 105:169-174.
Nemeth-Coslett R, Henningfield HE. 1986. Effects of nicotine chewing gum on cigarette smoking and subjective and physiologic effects. Clinical Pharmacology and Therapeutics 39:625-630.
Nestler EJ. 1994. Molecular neurobiology of drug addiction. Neuropsychopharmacology 11:77-87.
O'Brien CP. 1975. Experimental analysis of conditioning factors in human narcotic addiction. Pharmacological Reviews 27:535-543.
O'Brien CP, Testa T, O'Brien TJ, Brady JP, Wells B. 1977. Conditioned narcotic withdrawal in humans. Science 195:1000-1002.
O'Brien CP, Childress AR, McLellan AT, Ehrman R. 1990. Integrating systemic cue exposure with standard treatment in recovering drug dependent patients. Addictive Behaviors 15:355-365.
Piazza PV, Deroche V, Deminiere JM, Maccari S, Le Moal M, Simon H. 1993. Corticosterone in the range of stress-induced levels possesses reinforcing properties: Implications for sensation-seeking behaviors. Proceedings of the National Academy of Sciences (USA) 90:11738-11743.
Ritz MC, Lamb RJ, Goldberg SR, Kuhar MJ. 1987. Cocaine receptors on dopamine transporters are related to self-administration of cocaine. Science 237:1219-1223.
Robbins S, Ehrman R, Childress AR, O'Brien CP. 1992. Using cue reactivity to screen medications for cocaine abuse: Amantadine hydrochloride. Addictive Behaviors 17:491-499.
Rusted JM, Warburton DM. 1992. Facilitation of memory by post-trial administration of nicotine: Evidence for an attentional explanation. Psychopharmacology (Berl) 108(4):452-455.
Schuster CR, Greenwald MK, Johanson CE, Heishman S. 1995. Measurement of drug craving during naltrexone-precipitated withdrawal in methadone-maintained volunteers. Experimental and Clinical Psychopharmacology 3:424-431.
Slikker W, Brocco MJ, Killam KF. 1984. Reinstatement of responding maintained by cocaine or thiamylal. Journal of Pharmacology and Experimental Therapeutics 228:43-52.
Stewart J, de Wit H, Eikelboom R. 1984. Role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants. Psychological Review 91:251-268.
Thompson T, Schuster CR. 1964. Morphine self-administration, food-reinforced and avoidance behaviors in rhesus monkeys. Psychopharmacology 5:87-94.
Tiffany ST. 1990. A cognitive model of drug urges and drug-use behavior: Role of automatic and nonautomatic processes. Psychological Review 97:147-168.
Tiffany ST, Singleton E, Haertzen CA, Henningfield JE. 1993. The development of a cocaine craving questionnaire. Drug and Alcohol Dependence 34:19-28.
WHO (World Health Organization). 1992. International Statistical Classification of Diseases and Related Health Problems. Tenth Revision. Geneva: WHO.
Wikler A. 1973. Dynamics of drug dependence: Implications of a conditioning theory for research and treatment. Archives of General Psychiatry 28:611-616.
Wikler A, Pescor F. 1967. Classical conditioning of the morphine abstinence phenomenon. Reinforcement of opioid drinking behavior and relapse in morphine addicted rats. Psychopharmacologia 10:255-284.
Woods JH, Schuster CR. 1968. Reinforcement properties of morphine, cocaine, and SPA as a function of unit dose. International Journal of the Addictions 3:231-236.