Animals self-administer drugs that produce addiction in humans, and they do not self-administer the drugs that do not result in addiction.1 These statements are based on studies that have important implications for the conceptualization of drug addiction.
Drug self-administration studies have evolved over the last 30 years. In 1962, James Weeks2 published a description of a techniciue that permitted rats m self-inject drugs intravenously. The technique Weeks described consists of placing a catheter in the rat's vein in such a way that it cannot be pulled out. The distal end of the catheter connects to a pump that infuses a small volume of a drug solution when the animal presses a lever. Weeks showed that rats would readily press a lever that resulted in an intravenous infusion of a morphine solution.
Weeks made one incorrect assumption. He thought animals would self-inject morphine only if they were physically dependent. He assumed that the major determinant of opioid self-administration was avoidance of withdrawal symptoms. Thirty years ago most researchers shared this assumption. Subsequent research, however, has shown that drugs can sene as excellent reinforcers when there is little or no physical dependence. (A positive reinforcer is a stimulus or event that increases the frequency of responses that produce it.) This fact has important clinical implications. It means detoxifying or withdrawing people from drugs docs little to alter the future probability of relapse.
Over the last 30 years, many investigators have replicated and extended Weeks' findings. Animals including baboons, rhesus monkeys, squirrel monkeys, and rats will selfadminister drugs that produce addiction in humans.
The research that stems from die work by Weeks and others invalidated one other assumption. Some individuals assumed that only humans could become addicted to drugs, because they believed that addiction required the capacity for abstract thinking. Since rats and other animals display drug-seeking behavior, it is unlikely that abstract thinking is necessary.
Animals will self-administer five categories of drugs:
* Psychomotor stimulants. This group includes drugs such as cocaine and amphetamine.
* Central nervous system depressants. This is a very large category and includes alcohol, barbiturates, non-barbiturate sedative hypnotics, benzodiazepines, volatile solvents, and gaseous anesthetics.
* Opioids. This category includes heroin, morphine, codeine, and certain mixed agonists/antagonists such as pentazocine.
* Arylcyclohexylamines. Two examples are phencyclidine and ketamine.
* Nicotine. Nicotine can serve as an effective reinforcer.3 This finding is important evidence for identifying cigarette smoking as a type of drug addiction.
Not all addicting drugs have been adequately analyzed in animal selfadministration studies. For example, delta-9-tetrahydrocannahinol, the active component of marijuana, is poorly soluble in water. This lack of solubility has impeded its use in self-administration experiments.
Drugs differ in the degree to which they strengthen or reinforce drug-seeking behavior. Some comparisons among drugs within the same pharmacological class have been made. For example, cocaine is generally an excellent reinforcer. In situations where another drug is available along with cocaine, cocaine is usually the preferred drug.4 However, comparisons across drug classes are methodologically difficult. At present it is not possible to conclude that a drug such as cocaine is a better reinforcer than a drug such as heroin.
DRUGS NOT SELF-ADMINISTERED
Animals will not self-administer some psychoactive drugs. For example, antipsychotics, tricyclic antidepressants, opioid antagonists, and LSD-type hallucinogens do not serve as reinforcers. These findings were extended in a very important way. Animals can learn to avoid or terminate injections of some drugs. That is, certain drugs have aversive effects and serve as negative reinforcers. (A negative reinforcer is a stimulus or event that increases the frequency of responses that terminate or postpone it.) If animals are physically dependent upon morphine, they quickly learn to press a lever that postpones or terminates a naloxone infusion. Rhesus monkeys will also avoid or terminate infusions of antipsychotic drugs.5
The noxious effects of antipsychotic drugs probably account for the poor compliance of patients in taking these medications. In contrast, the tricyclic antidepressants have neutral effects. These studies of drug avoidance illustrate how findings from one line of research, namely drug addiction research, can lead to unexpected and important results in other areas.
Two major generalizations that emerge from animal drug selfadministration studies are:
* animals do not self-administer drugs that do not cause addiction in humans, and
* animals do self-administer drugs that produce addiction in humans.
These generalizations have limits. One limit is that animals selfadminister certain drugs that infrequently result in addiction in humans, such as some local anesthetics. However, these drugs do not appear to be strong reinforcers.
Some investigators have tried to identify common mechanisms or neuroanatomical pathways that underlie all drug reinforcing effects. However, drugs that serve as reinforcers come from very different pharmacological classes and have diverse mechanisms of action. There is no striking commonality except for their reinforcing effects. The identification of a shared neural substrate or other commonality is an objective of some current research. For example, dopamine serves an important function in mediating the reinforcing effects of cocaine and other psychomotor stimulant drugs,'1 and dopamine may also be important in mediating the reinforcing effects of opioids.*' However, there is far less evidence at present for a role of dopamine in the reinforcing effects of other drugs.'1 The feasibility of identifying neuropharmacological actions common to all addicting drugs requires more research.
ROUTES OF ADMINISTRATION
A number of routes of administration have been studied. In addition to the intravenous route, animals self-administer drugs via the oral, intramuscular, intragastric, pulmonary, and intraperitoneal routes. Recently developed techniques permit animals to self-administer drugs in nanolitcr volumes directly into brain tissue.7 One of the many similarities between drug-taking by laboratoiy animals and by humans is that drugs serve as reinforcers via multiple routes of administration.
What types of animals show drugseeking behavior? A drug such as cocaine serves as a reinforcer for nearly every animal tested: baboons, rhesus monkeys, squirrel monkeys, dogs, rats, and mice.8 These findings are troubling for some people, for the results suggest that it is biologically normal for certain drugs to have reinforcing effects. It is important to note that reinforcing effects of certain drugs are the rule and not the exception. Although most animals in a population will self-administer drugs such as cocaine, there are individual differences in drug intake among animals within a species. Reinforcing effects, like cither drug effects, are probably influenced by genotype. Thus, genetic differences may be one factor accounting for individual differences.9
One implication of these research findings concerns theories about the etiology of drug addiction. The fact that practically all animals tested will self-administer certain drugs means that it is unlikely that drug addiction is due to a biochemical abnormality or a genetic defect. An explanation based on self-medication is also unlikely, because animals used in laboratory research do not display signs of a psychiatric disorder that is antecedent to drug self-administration. Explanations of drug addiction exclusively in terms of factors such as low self-esteem and hopelessness are not plausible given the results of laboratory studies. These conclusions do not rule out the possibility that entities such as psychiatric illness or low self-esteem may be risk factors for drug abuse in humans.
SIMLAR DRUG-TAKMG PATTERNS IN HUMANS AND ANIMALS
Humans and laboratory animals not only find the same drugs reinforcing, they also show similar patterns of intake over time.10 For example, with psychomotor stimulant drugs such as cocaine and amphetamine, animals show a characteristic pattern of intravenous drug use. Periods of high drug intake that can last for several days alternate with periods of low or negligible intake. Different species ranging from baboons to rats show this cyclic pattern of intake of psychomotor stimulant drugs. During periods of high drug intake, the animals eat very little and do not sleep. Between drug infusions, .stereotypic patterns of motor behavior occur.
This cyclic pattern is very similar to that observed in humans who lake these drugs intravenously. Drug "runs" or binges can last several days and terminate when the user becomes exhausted or disorganized and falls asleep. Upon awakening the user is depressed and anergic; these states are terminated by initiating a new bout of drugtaking.
A very different pattern of intake occurs if drug access is limited to a few hours each day. Under these limited access conditions, animals will not show cycles of high and low intake. Instead, they take the drug in a regular pattern each day. During the interval when the drug is not available, the animals will sleep and eat and thereby maintain much belter health than when the drug is continuously available.
HUMAN STUDIES OF SELFADMINISTRATION
Humans have served as subjects in laboratory studies of drug selfadministration. Drugs that produce addiction in humans outside the laboratory serve as reinforcers under controlled test conditions. Those that have higher addiction potential, such as the short-acting barbiturates, are better reinforcers than drugs with lower abuse potential, such as the benzodiazepines." Drugs that do not produce addiction in humans do not serve as reinforcers.1'-' Unfortunately, most studies with humans and laboratory animals differ in many ways and thereby make comparisons difficult.
CONSEQUENCES OF DRUG INTAKE
Like humans, animals will show tolerance and physical dependence to certain drugs. Other adverse effects occur. Long-term effects on health depend on multiple factors. Two important factors are the type of drug taken and whether the drug is available continuously or for limited periods each day.
Humans and laboratory animals show similar patterns of psychomotor stimulant intake and also show similar toxic effects.'" When animals selfinject psychomotor stimulants and the drugs arc available continuously, their behavior is characterized by hyperactivity, stereotypic motor movements, and excessive grooming that can evolve into self-mutilation. These effects of high-dose amphetamine intake are similar to those reported in humans who intravenously use psychomotor stimulants. Humans will display repetitive purposeless behaviors. Delusions of parasitism and repetitious grooming behavior are common. Pricking and probing of the skin occur.
When rats and rhesus monkeys have unlimited access to psychomotor stimulants, they usually die within four weeks.13 During this time they show marked weight loss, and the duration of the drug binges becomes longer. This toxicity is not seen if periods of drug intake are limited to a few hours each day.
Toxic or adverse effects of selfadministered drugs manifest themselves in many ways, such as impairment in occupational or social functioning. However, if toxicity is assessed in terms of shortened life span, then relative to other reinforcing drugs, the psychomotor stimulant drugs are the most toxic. One other drug is notable for its ledial effects, and this is alcohol. The toxicity of alcohol is partly due to the suppression of" food intake. When given continuous access to alcohol, rhesus monkeys self-administer the drug in bouts that last several days." Remarkably, the animal terminates these bouts, and a major abstinence syndrome follows the termination. After several days of no alcohol intake, another bout is initiated. Death may occur during the abstinence syndrome following a bout. Alcoholics studied in a research ward also showed bouts of drinking followed by periods of sell-initiated abstinence and withdrawal.1' In animals, these toxic effects do not occur when intake is limited to several hours a day.
Other general ClNS depressants such as pentobarbital do not suppress food intake and do not have the short-term lethality ol" alcohol."' When pentobarbital, other CNS general depressants, opioids, or arylcyclohcxvlamines are available 24 hours a day, rhesus monkeys will persistently sell-administer the drugs and not initiate periods of abstinence. The animals remain chronically intoxicated and become tolerant and dependent, but their life span is not shortened as it is with alcohol and the psychomotor stimulants.
FACTORS CONTROLLING DRUGREINFORCED BEHAVIOR
Many variables affect drugseeking.1' The most commonly studied factors have been drug dose and schedule of reinforcement. A schedule of reinforcement specifies the relationship between responses and the delivery of a reinforcer. For example, in the simplest case one can deliver a reinforcer after each response. This is a fixed-ratio 1 schedule. Under a fixed-ratio 10 schedule, ten responses must occur before delivery of a reinforcer. Different schedules yield different patterns of responding. Schedules of reinforcement are an important determinant of all operant behavior.
Schedules of reinforcement have been of interest in drug-taking studies for several reasons. By varying the size of the schedule one can vary the effort or cost of obtaining the drug. The greater the number of" lever presses required per dose of drug, the greater is the effort or cost. A consistent finding across many studies is that drug consumption decreases as cost increases. Laboratory studies with humans reveal findings similar to those with laboratory animals. Drug consumption decreases as cost or work requirement increases. In society at large, alcohol consumption goes down as price goes up. Some individuals have argued that increasing the price of alcohol is one of the most effective measures I hat the government has for decreasing alcohol consumption and alcoholism. When prices increase, death rates from cirrhosis and automobile accidents decrease.
FUNDAMENTAL NATURE OF ADDICTION
Some individuals maintain that drug addiction is immoral behavior or willful misconduct. Others hold that it is illegal, antisocial behavior. However, the same individuals would probably be much less willing to characterize cigarette smoking as immoral or antisocial. Vet, high levels of cigarette smoking are as much instances of drug addiction as are high levels of use of other addictive drugs, and nicotine, like certain other drugs, can be an effective reinforcer for laboratory animals.
Moral or legalistic interpretations are not helpful in understanding the behavior of laboratory animals. The fact that animals self-administer the drugs that produce addiction in humans means there is a biological component to drug addiction. The fundamental core of drug addiction is that certain drugs can, under some conditions, serve as powerful reinforcers. Researchers have identified a number of conditions that control drug reinforced behavior, but there are no laboratory experiments that explain why drugs can serve as reinforcers. Nevertheless, it is possible to offer a plausible speculation as to why certain drugs act as reinforcers. Normal biological reinforcement systems exist such as the ones for food and water. Such reinforcement systems are a product of evolution. Dews1'1 has suggested that drug addiction is an accident of nature whereby a normal biological process, namely reinforcement, produces a pathological outcome. In other words, drug addiction, like autoimmune diseases, is a pathological outcome of what is usually a normal process (also see Goldstein and Kalant-0 and Miller et al-1).
1. Johanson CE, Balster Ri.. A summary of the results of a drug self-administration study using substitution procedures in rhesus monkeys. Bull Nan. Nac 1978; 30:4354.
2. Weeks JR. Experimental morphine addiction: method for automatic intravenous injections in unrestrained rats. Science. 1902: 138:143-144.
3. Henuingfield JE. Goldberg SR. Nicotine as a reinforcer in human subjects and laboratory animals. Pharmacol Biochem Behim 1983; 19:9894)92.
4. Johanson CE, Aigner T. Comparison of the reinforcing properties of cocaine and procaine in rhesus monkeys. Pharmacol Biochem Behav. 1981; 15:49-53.
5. Hoffmeister E Reinforcing properties of perphenazine, haloperidol and amitryptiline in rhesus monkevs. J Pharmacol Exp lhei. 1977; 200:516-522.
6. Wise RA, Rompre P-R. Brain dopamine and reward, Annu Rev Psychol. 1989; 40:191-225.
7. Goeders NE. Smith JE. Intracranial selfadministration methodologies. Neurosa BioMifw Riv. I !187: 11:319-329.
K. Kishnian MW. Cocaine and the amphetamines. In: Meltzer I IV ed. Psychopharma cology: The Third Generation of Progress. New York, NY: Raven Press; 1987:13431553.
9. George ER, Goldberg SR. Genetic approaches io the analysis of addiction processes, irends Pharmacol Sci. 1989: 10:78-83.
10. Griffiths RR. Bigelow GE. Henningfteld [E. Similarities in animal and human drug-taking behavior. In: Mello NK. cd. Advances in Substance. I. Greenwich. Conn: JAI Press Ine: 1980:1-90.
11. Griffiths RR. Koache JD. Abuse liability of benzodiazepines: a review of human studies evaluating subjective anchor reinforcing effects. In: Smith I)K. Wesson I)R. eds. ihe Benzodiazepines: Current Standards for Medical Practice. Lane aster. England: MTP Press Limited; 1985: 209-22*).
12. Henninglielcl JE. Lukas SE.. Bigelow C.V.. Human studies of drugs as ? ein forcers. In: Goldberg SR. Stolcrman IP. eds. Behavioral Analysis of Drug Dependence. Orlando. FIa: Academic Press; 1986:691 22.
13. Johanson CE. Balsler RL. Bune.se K. ScIiadministiation of psychomotor stimulanl drugs: the effects ol unlimited access. Pharmacol linn hi m Belimi. 1970; 4:1551.
14. Winger JD. Woods JII. Ihe reinforcing properly of ethanol in the rhesus monkey: I. initiation, maintenance and termination of intravenous ethaliol-reinlbrced responding. .Inn .V )' Acad Sci. 1973: 215:102-175.
15. Mc-IIo NK, Meiidelson JH. Drinking palterns during work-contingent and nonconlingcni alcohol acc|uisition. Psychosum Med. 1972: 31:139-164.
16. YanagilaT. Takahashi S. Development of tolerance Io and physical dependence on barbiturates in rhesus monkeys. J Pharmacol Exp Ther. 1970: 172:163-169.
17. Meise h RA. Kaclors controlling eli 1114 reinforced behavior. Pharmacol liuti Item Behau 1987; 27:307-371.
18. See ley JR. Death bv liver cirrhosis and the price of beverage alcohol. CWf? Med Assoc J. I960; 83:1361-1 300.
19. Dews PB. Maintenance of behavior bv "schedules": an unfamiliar contributor to the maintenance of the abuse of substances and the tike. Uv. Levisene PK. ed. Su bsla nee Abuse. Habitual Behavior, and Self Control. Boulder. Colo: West view Press: H)Ht: 19-80.
20. Goldstein A. Kulant H. Drug policy: striking the right balance. Science. 1990: 249:1513-1521.
21. Miller NS, Dackis CA. Gold MS. The relationship of addiction, tolerance, and dependence Io alcohol and chugs: a neurochemical approach. / Snbst Abuse lient. 1987: 4:197-207.