General Features


Changes in Receptor Sensitivity

Enzyme Induction

Rebound Effects


Behavioral Tolerance

Pre-Post design.

Environment and ritual.

Opponent Process Theory



Self Administration

Reinforcement restructured.

Environmental bridges.

Breaking the Cycle






General Features

The introduction of a drug into a biological system is far more complicated than adding a compound into a test tube. The initial dosage of the drug, the route of administration, the rate of absorption, the rate of elimination and many other factors enter into this complex equation. Ultimately, the effectiveness of the drug is usually determined by the concentration of drug molecules in the plasma that are free to interact with receptor sites (cf., Chapter 3).

The distribution of drug molecules to the receptor sites may be complicated, but it is only the tip of the iceberg in terms of the organism's overall response to the drug. Early in the text, a somewhat loose distinction was made between a drug action (how the drug interacts with a specific receptor, e.g., mimicking acetylcholine at muscarinic receptors) and a drug effect (the physiological or behavioral results of this drug action; e.g., a decrease in heart rate or an increase in arousal level). There is yet another class of drug effects that may or may not involve the specific receptors that mediate the drug action: The presence of the drug may trigger any of several different responses that change the reaction to future encounters with the drug. This altered response to the drug is usually a decrease (tolerance), although increased response to the drug (sensitization) can also occur.

The body has two general methods of increasing its tolerance to a drug. One of these is to reduce the opportunity of the drug to reach the receptors (i.e., reduce the drug action), the other is to launch a biological counterattack against the drug effect (i.e., a compensatory reaction).

There are several different ways in which the receptors can be insulated from the drug. The entry of free drug molecules into the bloodstream can be reduced by lowering the rate of absorption from the stomach and intestines. This might be accomplished mechanically by a change in blood flow or peristaltic action, or biochemically by a reduction of the transport mechanisms that may be required to carry the drug molecules across the membranes and into the plasma compartment. Another possibility is to allow the drug to enter the bloodstream in exactly the same way, but to reduce the final level that is reached by increasing the rate at which the drug is eliminated (e.g., by the kidneys or liver). Finally, it may be possible to increase the binding of drug molecules into complexes with other, larger molecules to render them inert. Figure 9.1 summarizes these mechanisms of tolerance.

There are also several alternatives through which a compensatory response can be made to drug effects. One of the most straightforward ways is to increase the activity of an opposing system. For example, if a sympathetic agonist is increasing the heart rate, this could be countered by an increase in parasympathetic activity that reduces the heart rate. Although the contrapuntal relationship between the sympathetic and parasympathetic systems has been overrated, mutual feedback systems do tend to modulate and balance the activity of these systems, and some brain systems may have comparable patterns of organization. The details of this compensatory response are usually not clear for any individual case, but the general features probably involve neuromodulation, which has already been discussed in other contexts. If, for example, a drug reduces the amount of transmitter that is released, the postsynaptic cells may respond by increasing the number and or sensitivity of receptor sites to maximize the effect of the available transmitter molecules. (This process is basically the same as denervation supersensitivity, a phenomenon which can occur if the fibers coming into a cellular region have been cut and allowed to degenerate. Following the degeneration, the cells that have lost their inputs frequently show an increase in the number of receptor sites and become supersensitive to even small amounts of the missing transmitter substance.) Alternatively, agonists may cause the postsynaptic cells to reduce the number of receptors to prevent excessive levels of stimulation. At a still more complicated level, behavioral tolerance may enter into the picture, with the organism learning or otherwise adapting to the effects of the drug, such that the behavior is normalized in spite of any prevailing physiological changes that the drug may produce. These compensatory mechanisms are summarized in Figure 9.2 (cf., Ellison et al, 1978; Lee & Javitz, 1983; Schwartz & Keller, 1983).

The common feature of all of these mechanisms of tolerance is that the response to subsequent drug administration is changed. Depending upon the nature of the particular response, the tolerance might be evidenced by a change in the effective dose, the lethal dose, the time course of the drug effect, the range of effects, or some combination of these. Furthermore, the changes that occur within one system can even alter the future responses to drugs that are in a different pharmacological class (cf., Glowa & Barrett, 1983). We turn now to some specific examples of tolerance to demonstrate some of these reactions to repeated drug injections.


The term tachyphylaxis literally means rapid protection and is exemplified by the tolerance that develops to the effects of indirect acting drugs. Ephedrine, a drug which stimulates the sympathetic nervous system, is such a drug. As shown in Figure 9.3a, a standard dosage of ephedrine produces a rapid and short lived increase in blood pressure. If this same dosage is repeated at 10-minute intervals, the effect becomes smaller and smaller until, after several dosages, there is virtually no change in blood pressure. How could such a rapid tolerance develop?

The mechanism of this rapid tolerance can be inferred by the time course and by the effects of other drugs. The tolerance does not represent a permanent change, because the change in blood pressure will return to its original level if an interval of several hours is allowed between doses. This pattern of rapid tolerance that goes away quickly could be the result of fatigue of the smooth muscles that cause the vasoconstriction. However, the effects of epinephrine show that this is not the case. Repeated dosages of epinephrine continue to produce large elevations in blood pressure, and a single dosage of epinephrine given at a time when ephedrine has no effect, will produce a full scale change in blood pressure (see Figure 9.3b).

The interaction of these drugs with a third drug, reserpine, provides further information about the mechanism of tachyphylaxis. Reserpine causes the gradual depletion of norepinephrine from the sympathetic terminals. This results in a decline in blood pressure, which can be readily reversed by epinephrine. By contrast, ephedrine (even the first dosage) has no effect on blood pressure after the transmitter substance has been depleted (see Fig. 9.3c).

The conclusion is that the tachyphylaxis is the result of a rapid emptying of the transmitter substance from the synaptic vesicles (see Figure 9.3d). Ephedrine per se has no direct effect on the smooth muscle receptors that mediate the change in blood pressure. Rather, the elevated blood pressure is produced indirectly by stimulating the release of the neurotransmitter from nerve terminals. Repeated dosages of the drug in rapid succession release the transmitter faster than it can be replaced, and the effectiveness of the drug declines. These conclusions are further supported by the observations that norepinephrine administration not only produces an increase in blood pressure, but it also partially restores the effectiveness of ephedrine. The restoration occurs because the reuptake process (cf., Chapter 3) incorporates some of the norepinephrine into the vesicles where it can be released by the next dosage of ephedrine.

This form of tachyphylaxis is a special case of tolerance that does not involve any particular reaction of the systems involved. It is a simple case of the drug effect being limited by the capacity of the system to respond. The remaining types of tolerance that will be discussed involve a much more dynamic and longer lasting reaction to the effects of drugs.

Changes in Receptor Sensitivity

Tolerance can also be mediated by a change in the sensitivity of the relevant system to the drug or transmitter. An example of this sort of tolerance can be seen in the results of an experiment performed by Brodeur and DuBois in 1964. They administered daily dosages of an acetylcholinesterase inhibitor to rats. This blockade of the inactivation of acetylcholine allows the transmitter substance to accumulate. Initially, these drug injections produced a variety of parasympathetic symptoms, including tremor and convulsions. By the end of 60 days, however, tolerance had developed and none of these effects was observed.

There are several possible ways that such tolerance could be developed. For example, the drug could lose its ability to block acetylcholinesterase. However, assays demonstrated that the degree of cholinesterase inhibition remained unchanged over the 60-day treatment period. This leaves open the possibility that the acetylcholine levels were brought under control by some other mechanism, but measures of acetylcholine showed the same high levels were produced by the drug on Day 60 as on Day 1. How could tolerance develop if the actions of the drug remained constant?

Suppose the observed tolerance occurred because the neural systems had become refractory to the high levels of acetylcholine. This notion was tested by administering carbachol, a synthetic drug that acts on receptors for acetylcholine, but is immune to the inactivating effects of acetylcholinesterase. In animals that had not received prior drug treatment, the LD-50 was 2 mg/kg. The rats that had developed tolerance to the cholinesterase inhibitor were twice as resistant to the effects of carbachol, showing an average lethal dose of 4 mg/kg. In other words, the drug continued to inhibit the action of acetylcholinesterase, the resulting increase in acetylcholine levels were maintained, but the tremors and convulsions disappeared: The drug actions remained constant while the drug effects declined.

The most likely mechanism for this form of tolerance is a change in the sensitivity of the postsynaptic membrane. This can occur through the process of neuromodulation (see figure 9.4). Synaptic activity is a dynamic process, which can be controlled by either a change in the amount of transmitter substance that is released or a change in the response to the transmitter. In the example described above, it would appear that the neural systems responded to the high acetylcholine levels by reducing the number of receptors (cf., Schwartz and Keller, 1983).

Although it is not necessary to go through the details of a specific example, it should be pointed out that the same mechanisms can result in tolerance to drugs that produce a decrease in transmitter substance. The initial effects of transmitter reduction are typically greater than the chronic effects. This type of tolerance can be attributed to an increase in the number of postsynaptic receptors. This effect has been described in earlier discussions as it applies to the phenomenon of denervation supersensitivity. The increased sensitivity that follows nerve damage can be viewed as tolerance to the physical damage.

These changes in receptor populations involve some rather major commitments of cellular metabolism. As such, they have the properties of both inertia and momentum; it takes some time (perhaps days or weeks) for the tolerance to develop and perhaps even more time for the system to return to initial levels when the drug is no longer present. These are very important considerations which will be seen in more detail in the later discussion of rebound phenomena.

Enzyme Induction

Enzymes are protein molecules that increase the speed of chemical reactions. They typically have a rather high affinity for a particular chemical structure (the substrate) and the enzyme-substrate complex proceeds through the chemical reaction faster than the substrate alone. We already have seen several examples of enzymes that are involved in neurotransmitter systems (e.g., AChE, MAO, COMT and tyrosine hydroxylase). The liver has a rather extensive library of enzymes that facilitate physiological processes (especially digestion) and help to break down toxic substances from both internal and external sources.

The chemical specificity of enzymes allow for the precise control of chemical reactions, but it also poses a problem. It would be very inefficient (not to mention impossible) for the body to produce and store all the enzymes that might be needed. It would be much simpler to have a way to limit production to those that actually are needed. This is what happens in the process known as enzyme induction (see Figure 9.5). When a new foodstuff or drug is encountered it may induce the metabolic machinery to produce an enzyme that has the specific ability to break it down into simpler components that can be used by the body (in the case of foodstuffs) or inactivated and eliminated (in the case of drugs).

The induction of enzymes involves protein synthesis, a process that may require several hours or more to take place. What this means in terms of the metabolic fate of drugs is that the drug molecules from the first injection may induce the formation of the appropriate enzyme (usually by liver cells), but undergo metabolism through the existing, sluggish pathways. Thus, the drug may stay in the system and produce its effects for a long period of time. However, once the liver cells have begun production of the enzyme, it is more readily available for encounters with the drug molecules, and the breakdown reactions for subsequent dosages will proceed more rapidly. It should be noted that this is not an all or none process, but rather one which can be regulated by the number of times the inducing substance is encountered and the amount that is presented. In any event, the induction of enzymes can result in dramatically different rates of drug metabolism that are seen as examples of tolerance.

The barbiturate drugs provide a good example of tolerance that is at least partially the result of enzyme induction. Remmer (1962) administered high anesthetic doses of pentobarbital to rats on three successive days. Rats in a control group received daily injections of saline. Then all the rats received a lower test dosage to determine if tolerance had developed. The rats that had been pre-treated with pentobarbital slept only half as long as the rats in the control group (30 min vs 67 min). This change in sleeping time was paralleled by a change in the rate of eliminating the drug from the system. The half-life of the drug (the time required to inactivate half of the injected drug) in the control group was twice as long as that of rats that had been pre-treated with pentobarbital.

It could be postulated that the relevant brain cells became less responsive to the effects of pentobarbital in the same way that the cells became less responsive to acetylcholine in the previous discussion. The evidence does not support this. The concentration of pentobarbital in the blood at the time of awakening can be used as an index of the sensitivity of the cells to the drug. As the concentration gradually falls, it eventually reaches a level that is low enough to allow the animal to awaken. The rats that had been pre-treated were at least as sensitive to the drug as control rats, with waking levels of the drug that were even slightly lower than those of the control group.

The phenomenon of enzyme induction can produce a dramatic tolerance in terms of the effective dosage of a drug, but it does not necessarily confer the same degree of protection against lethal dosage. It fact, the LD-50 can remain virtually the same, while tolerance increases the requirements for an effective dose (the ED-50) until it may be almost identical to the lethal dose. Let us examine this curious phenomenon more carefully with a hypothetical extension of the pentobarbital tolerance shown above.

The upper panel of Figure 9.6 shows the normal course of a barbiturate drug. After injection, the drug is rather quickly absorbed into the plasma compartment. When a certain concentration is reached, sleep ensues while the drug levels continue to rise and produce a deeper level of anesthesia. Eventually, the drug will reach its peak concentration, which is determined by the amount of drug, route of administration and other factors discussed in Chapter 3. Meanwhile, the drug is being metabolized and the plasma concentration begins to decline. When it reaches a certain level, the animal awakens and the drug metabolism continues until the drug has been eliminated from the system.

After several exposures to the drug, the enzyme that degrades the drug has been induced, and the drug is removed from the system more rapidly (Figure 9.6b). This shortens the sleeping time by allowing the animal to awaken more quickly. However, the onset of sleep and the peak concentration of the drug in the plasma may show little or no change if the absorption of the drug is fast relative to the drug metabolism.

Now, suppose an attempt is made to duplicate the original drug effect (60 min of anesthesia) by increasing the drug dosage. The faster rate of drug metabolism requires a very high dosage to forestall awakening for the full hour. In this hypothetical case (see Figure 9.6c), the peak plasma levels are very near the lethal dosage.

With the margin of safety (the ratio of the LD-50 to ED-50) reduced by tolerance, it may be advisable to administer multiple doses over time (e.g., a supplemental dosage every 20 min) to attain the same duration of action (see Figure 9.6d). These effects demonstrate that tolerance can render a drug considerably more dangerous, a finding that has important implications in the clinic, the laboratory, and on the street.

Rebound Effects

The various types of tolerance are, in some sense, an extension of the concept of homeostasis. The physiology of the organism reacts to a challenge by attempting to return the system back to normal. In the case of drugs, this can have important consequences not only for the changes in the effectiveness of the drug, but also for the rebound changes that occur when the drug is no longer present.

The rebound phenomena can be observed readily in the case of nicotine and caffeine. On the surface, it would seem that individuals who use these relatively mild CNS stimulants should be easily identifiable. They should, perhaps, have faster reflexes, be more vigilant, require less sleep, or be more aware of the environment. Or perhaps they should be more irritable, anxious, or jumpy. None of these effects, positive or negative, is observed. Virtually every attempt to extract an identifiable difference in physiology or personality between smokers, coffee drinkers, and nonusers has failed. The differences are revealed when these groups are compared without drug. The details of the effects are complicated, but nearly everyone has seen or been either a coffee drinker before the first cup in the morning or a smoker trying to quit. The major effect of these stimulant drugs is not to produce an average state of arousal that otherwise could not be attained, rather they come to prevent the rebound effects that would occur without the drug (sleepiness, lack of energy, dysphoria, etc.). The mechanisms of tolerance actively counterbalance the effects of the drug, and this balance can be unmasked by removing the drug from the system (see Figure 9.7).

The rebound effects can be considerably more serious than a little early morning grumpiness. The chronic, heavy use of CNS depressant drugs such as barbiturates or alcohol can set up dangerous counter-effects. When these effects are released by the abrupt withdrawal from the drug, hyperexcitability occurs, which in severe cases can lead to convulsions and death. With chronic, heavy use of alcohol, this rebound hyperexcitability may be seen as largely irreversible motor tremors, especially of the hands (the DT's or delirium tremens). These rebound effects are most pronounced when the drugs are withdrawn abruptly after a period of sustained high dosages, but this is not a prerequisite. So swift is the body's ability to counter these drugs that a single dosage can set up rebound effects. An overdose of alcohol or barbiturate first presents the danger of death through its depressant effects, followed by the susceptibility to seizure activity that may be equally dangerous. It is for this reason that the most superficially obvious treatment of barbiturate overdose-- the administration of stimulant drugs such as strychnine or picrotoxin-- is contraindicated.

Rebound effects are also major factors in the use of amphetamine and related drugs. The actions of amphetamine are complicated and include both direct effects on the postsynaptic receptors and the indirect release of the neurotransmitter substance from the presynaptic vesicles. The behavioral effects include increased arousal, greater physical energy, a heightened sense of well being, and even euphoria. Tolerance to these effects occur readily and probably include all of the mechanisms (transmitter depletion, receptor changes, and enzyme induction) that were discussed above. A common sequel to the stimulant properties of amphetamine is a profound depression, the depth of which is related to the amount and duration of the drug administration. In practice, it is almost impossible to avoid the rebound phenomena. In part because of the indirect actions of the drug, the effects tend to be self limiting, with the depletion of transmitter rendering the drug ineffective until a period of time has been allowed to restore the system toward its previous state.


In studying drug actions and drug effects, it is sometimes easy to forget a basic fact about the physiology of the nervous system: It was not designed for the purpose of responding to drugs invented or harvested by man. The actions of drugs can magnify, interrupt, speed up and delay, but the processes are those that are inherent to normal functions and the maintenance of the brain and behavior. Accordingly, we must not limit the phenomena of tolerance and withdrawal to the realm of drug use, but must search out the relevance to normal functions. Neurotransmitters can be viewed as endogenous drugs and they surely trigger their own neuromodulatory and rebound effects. We turn now to some situations in which behavioral processes impose dramatic limits on the effects of both drugs and environmental situations.

Behavioral Tolerance

The phenomenon of tolerance can be no simpler than the actions of the drug to which tolerance is developing. One of the fundamental principles of pharmacology is that no drug has a single action (cf., side effects in Chapter 3). The extended implication of this is that no drug can induce a single type of tolerance. If a drug has a major effect and two distinct side effects, then it is very likely that tolerance can develop to each of these independently. In some cases, this can be the determining factor in the development of a drug for clinical use. For example, if tolerance develops to a troublesome side effect within a few days or weeks, then the patient may be able to benefit from the long term use of the drug. On the other hand, if tolerance develops to the main effect, but not the side effect, then the drug becomes less and less useful over time.

The development of tolerance to specific aspects of drug action may present some difficulties, but the picture becomes still more complex when behavior is considered. In the previous section, we developed the notion that behavior per se could be likened to a drug, triggering rebound effects comparable to those occurring in response to the administration of drugs. We turn now to a consideration of behavioral contributions to drug tolerance and present some of the most intriguing findings in the pharmacological literature.

The development of tolerance to the effects of a drug poses some of the same problems of interpretation that are encountered in the recovery from brain damage. In both instances, the initial effects frequently are more pronounced than the long term effects. In the case of brain damage, the problems of interpretation are particularly intractable. Does the recovery of some of the lost function reflect the "take-over" by a related area of the brain? Is it due to some neuromodulatory effect such as denervation supersensitivity? Or, is it the result of learning to accomplish the same behavioral goal in a different fashion? There is evidence to support each of these possibilities, but the conclusions remain tentative because of the permanence of brain damage.

As researchers began to look more and more carefully at both tolerance and recovery, some confusing observations began to appear. In some cases, clear recovery of function following brain damage could be observed on one task but not another. Likewise, drug tolerance could sometimes be observed in one measure but not another. There was something missing in the interpretation of these effects. Fortunately, most drug effects are considerably less permanent than brain damage (although some of the tolerance and rebound effects may be very long lasting) and can be administered repeatedly. This opens the door for experimental approaches that can better answer some of the questions posed above.

Pre-Post Design.

One of these approaches, known as the pre-post design, can be used to demonstrate the phenomenon known as behavioral tolerance. A series of experiments by Carlton and Wolgin (1971) exemplify this approach. The rats in these experiments received a restricted diet of food pellets and were given a daily period of access to a drinking tube that contained sweetened condensed milk. After several days, the amount of milk that was consumed during this period stabilized and served as a baseline for the drug and drug tolerance effects. If amphetamine was injected a few minutes before the daily test session, the milk consumption was greatly reduced. However, if rats received amphetamine injections before each daily session, the rats drank a little more milk each day until, by the end of the two-week experiment, the rats that received amphetamine were drinking as much milk as the control rats that received saline injections. Thus, tolerance had developed (see Figure 9.8).

The most obvious explanation of the increase in milk consumption is that one of the pharmacological tolerance mechanisms described above had taken place to reduce the effectiveness of the drug. Another possibility, however, is that the drug actions remained essentially the same, but that the animals made a behavioral compensation that allowed them to drink the milk despite the drug effects. The pre-post design allowed a test of these alternatives by comparing the milk consumption of the following treatment groups:

Group CON: Saline injections before each session

Group PRE: Amphetamine injections before each session

Group POST: Amphetamine injections after each session

The critical treatment group is the one that received the amphetamine injections after each session. Obviously, the drug cannot be influencing the milk consumption before it is given. (It could influence the consumption during the next session, 23 hr later, through residual effects of the drug or through conditioned aversion effects, but these possibilities were controlled for in the complete design of the study.) Even though the measure of milk consumption was taken deliberately in the "wrong" place, one would still expect that pharmacological tolerance would be developing to the amphetamine. Enzyme induction could occur, the number of receptors could be changed, the amount of neurotransmitters could be changed, or some combination of these could occur. But the measure of milk consumption would be blind to these effects because it was taken long after the drug was given (23 hr).

The critical test occurred after these mechanisms of tolerance were given an opportunity to develop. The rats that had been in Group POST now were given a test dosage of amphetamine BEFORE the milk consumption. If tolerance to the amphetamine had been developing, as it had in the rats in Group PRE, then the milk consumption should have remained at the baseline level (see Figure 9.8). Instead, the milk intake was substantially reduced! How could tolerance to amphetamine develop for one group of rats but not the other? The answer (although it does not specify a mechanism) is behavioral tolerance: the rats in Group PRE became tolerant to the effects of amphetamine on milk consumption. Although pharmacological tolerance may have developed over the treatment period, this was not sufficient to block its effects on milk consumption. The necessary component was behavioral experience while the drug was in effect. The rats, in some sense, had learned to consume milk despite the effects of amphetamine.

A comparable experiment has been done by Campbell and Seiden (1973) using performance of a drl task to assess the effects of amphetamine. The performance of this task, which requires low rates of responding, is severely impaired by the effects of amphetamine. However, tolerance develops with repeated injections, and the behavior returns to the normal baseline that was obtained without drug. Again, the pre-post test design showed that this return of normal behavior could not be attributed to pharmacological tolerance. Rats that had received repeated injections of amphetamine, but without the opportunity to perform the drl task while under the influence of the drug, still showed serious impairment when the drug was given before the drl test session.

The results of these experiments suggest some clinical considerations that probably have not received the attention they deserve. The amphetamines and related compounds are widely used by dieters in both prescription and over the counter formulations. The long term effectiveness of this therapy is marginal at best. It seems likely that humans, as well as rats, develop behavioral tolerance and learn to consume the good things in life (including sweetened condensed milk) despite the effects of the drug.

One possible explanation of behavioral tolerance is that it somehow blocks out the ability to perceive the effects of the drug. This is probably not the case. Bueno and Carlini (1972) showed that the ability of rats to climb a rope was impaired by THC (the active component of marijuana). After tolerance developed, the rats were able to climb the rope as well as control animals, but were nonetheless capable of discriminating (in a different task) the presence or absence of THC.

Environment and ritual.

One of the most dramatic demonstrations of the power of behavioral tolerance has been demonstrated by Siegel and coworkers (1982). The repeated administration of opiate drugs produces a remarkable degree of tolerance. In order to maintain the same level of analgesia over a period of days, the drug must be administered in ever increasing dosages and can reach levels that may be several times higher than the LD-50 established for naive animals. These investigators administered a schedule of increasing dosages of heroin working the rats up to a dose that they could not have initially tolerated. Half of the rats had received this series of injections in the colony room. The other half of the rats were removed from the colony to a test room that differed both visually and by the presence of a 60 dB white noise, where they received their injection of heroin. As expected, all of the rats withstood the increasing dosages of heroin.

After completing this series of increasing dosages, the rats were given a single test dosage of 15 mg/kg of heroin. This is a very large dosage of the drug, being close to the LD-100 (96% mortality) for rats that have had no experience with the drug. Rats that had received the series of heroin injections showed a substantial increase in the ability to withstand the drug, with only a 32% mortality rate. However, a large portion of this protective effect was attributable to behavioral rather than pharmacological tolerance. If the rats received the same injection, but simply in a different room (colony rats in noise room; noise room rats in colony), they were twice as likely to die (64%). The association of a particular environment with the administration of a drug adds to the ability to compensate for the effects of the drug. In this case, the learned aspects of the tolerance can literally mean the difference between life and death.

Siegel proposes that many of the deaths that occur through drug overdose have a behavioral component. Addicts frequently develop ritualistic behavior associated with the administration of a drug (same place, same people, etc.) When this ritual is changed, for example, by purchasing the drugs on the streets of another city, the likelihood of death through overdose is increased. The common explanation that the drug obtained was more potent than that usually used may be true in many cases, but the behavioral component may be a major factor as well.

How is it possible for an animal to behaviorally reduce a drug effect? Some of Siegel's earlier work provides a possible answer to this question (Siegel, 1975). The work centered on the possibility of the Pavlovian conditioning of drug effects. Suppose, for example that the injection procedure (the room, the handling, the insertion of the needle) is always conducted in the same manner. This set of stimuli could serve as a conditioned stimulus (CS) to predict the physiological changes (UR) that would follow as a result of the drug injection (the unconditioned stimulus, or US). What would happen after a number of such pairings if saline were substituted for the drug? The CS (injection procedure) would be the same as always, but would a conditioned response (the physiological change) be observed?

The drug under investigation was insulin, which lowers the blood sugar levels. If insulin injections are given in the same manner, is it possible to get a conditioned change in blood sugar levels that parallels the conditioned salivation that occurs when a bell has signaled the presentation of food powder? The answer is yes, but the direction of the effect is opposite that which one might first expect. Instead of getting a conditioned lowering of blood sugar, Siegel observed a conditioned increase in blood sugar. This makes perfect sense if the conditioned response is viewed as an attempt to compensate for a predicted change in the environment. Normally, the amount of insulin produced by the animal is controlled within rather narrow limits to regulate the level of glucose utilization by the cells, prepare for digestive loads, etc. The injection of an outside source of insulin disturbs this balance, and the animal must reduce its own production of insulin in an attempt to counteract this effect. According to Siegel's results, this compensatory response can be learned, and a sham injection procedure causes a reduction in insulin and a corresponding increase in blood sugar. Of course, little or none of this is cognitive learning (try to imagine how you would voluntarily reduce your own insulin levels), but the mere fact of association is sufficient to trigger these processes according to the laws of Pavlovian conditioning.

Once again, the importance of these phenomena in the clinic should not be overlooked. When drugs that cause pronounced physiological effects are given over long periods of time, there is a very real possibility that Pavlovian learning processes may take place to counteract the effects of the drug.

Opponent Process Theory

Solomon and Corbit's theory of opponent processes is modeled after well established events that occur in the sensory systems. The most familiar of these are the negative after images that occur in the visual system. If an individual stares steadily at a relatively bright object, say a television screen in a dimly lighted room, the absence of that stimulus produces a curious illusion. When the vision is shifted to a neutral part of the room, a ghostlike image of the screen is projected onto the surface and this image is dark rather than light. Hence, the term negative after image. The negative after image is also familiar in the case of indoor photography. The bright flash of light is followed by after images (usually negative, but sometimes alternating positive and negative) of small dots that are projected onto the "real" visual world. These after images even extend into the realm of color vision, with the after images being of the complementary color (red objects produce a green after image, blue objects produce yellow, and vice versa). Comparable illusions can appear with the motor system, as evidenced by the "light-footed" feeling that occurs when a pair of heavy boots or roller skates is removed. But does it make sense to apply these principles to something as complicated as emotions? Probably yes.

There are three major components to the opponent process theory:

Affective contrast is the most fundamental of these, and closely parallels the response of the visual system to light. The presentation of a bright light produces a peak response followed by rapid adaptation to a stable level (the A-response in Figure 9.9). When the light is turned off, a negative after image occurs and gradually dissipates with time. The magnitude of these effects is related to the intensity of the stimulus. Several observations can be cited to relate this to emotions. An infant may be lying quietly in its crib, exhibiting no particular emotion. If a nipple containing a sugar solution is offered, a positive response is obtained (the A-response). Withdrawal of the nipple results in vigorous crying (the B-response), an effect which would not have been observed if the positive stimulus had not been presented. Comparable effects can be observed in the case of initially negative stimuli. Electric shock administered to dogs can produce an increase in heart rate. When the shock is terminated, there is a dramatic decline in heart rate and the dogs may show behavioral excitement. This is very likely the laboratory equivalent of the frenzied play activity that sometimes follows the administration of a bath to a dog (or a child, for that matter!). A more familiar example to students may be the excited chatter that frequently fills the hallways after a major examination.

The story gets more complicated with affective habituation. If a bright light is presented for a long period of time, habituation occurs and the perceived brightness is greatly diminished (the A'-response in Figure 9.9). But when the light is terminated, the negative after image is both stronger and more enduring than it was following a brief, initial exposure (the B'-response). In the laboratory, this can be observed with repeated presentation of shock to dogs. After a time, the shock no longer produces a change in the heart rate, but the "after-image" (the decrease in heart rate) becomes very pronounced. Again, the same pattern emerges in the case of human emotions: The heart throbbing, adrenergic effects of a new amour might well become a health hazard if they continued; but in the words of the songwriter, "...after 16 years of marriage, the fires don't burn so hot!" (Harry Chapin). Returning to the theory, the A'-response takes over, but the stage is set for a tremendous B'-response if the stimulus should be terminated, e.g, the grief response that follows the loss of a loved one.

The third aspect of the theory, affective withdrawal, is really just a sharpening of the concepts described by the first two. We will describe two of the numerous examples put forth by the theorists. One of these involves the sport (?) of skydiving. For the naive jumper, the period before the jump is filled with anxiety. This anxiety is galvanized into terror with the actual jump, and relief follows a safe landing. Should the individual continue this pastime, the emotions that color the experience undergo the pattern of affective habituation and contrast described above. Anxiety is replaced by eagerness, the terror is downgraded to a thrill, and relief is transformed into intense exhilaration--the raison d'etre for what would otherwise be a silly thing to do. A similar pattern can be applied to the abuse of a drug, for example, heroin. The initial presentation is preceded by a state of rest, the drug's actions produce a "rush", and the aftereffect is one of craving. With veteran users, there is a shift in the emotions, and the drug's actions produce a state of contentment (rather than a rush). This contentment (the A'-response) is followed by abstinence agony (the B'-response), which turns into an intense craving for the drug, and the drug now has only the capacity to relieve the craving rather than reproducing the initial rush-- and the circle continues (see Figure 9.10).

The opponent process model is also relevant to many of the paradoxical effects that accompany goal attainment. Reinforcement in an operant schedule does not necessarily spur the pigeon into immediate further action, but rather may be followed by a post-reinforcement pause. The attainment of a long-sought goal such as a college degree is frequently followed by a bout of depression, and the postpartum blues are almost unavoidable. Solomon and Corbit emphasize the view that all of these effects are noncognitive in nature. That is, they are not the result of a logical, cognitive analysis of the present environment, but rather are the result of a previous environment that no longer applies.

These changes may well be noncognitive, but they cannot-- if we continue our attempts to view the brain and behavior in a lawful relationship-- be nonphysical. Powerful stimuli produce powerful changes in the neurotransmitter systems, and these in turn trigger the processes of neuromodulation, changes in receptor sensitivity, and even transmitter depletion. These reactions, like tolerance to a drug, alter the responses to standard stimuli and set the stage for withdrawal reactions.

Although the opponent process theory has not gained universal acceptance, we shall risk pushing it one step further in terms of the noncognitive aspects of emotions. There will not be many students of this book who will recall the Mary Tyler Moore show, but one of the episodes provided a poignant example of opponent processes in action. A dear friend of Mary's, Chuckles the Clown, died. Of course, she was stricken with grief, and she decried all references to the lighter side of his life and career. At the funeral, however, she was overcome by an uncontrollable urge to laugh; not hysterical, unfeeling laughter, but true, euphoric, high spirited laughter. Why? The grief reaction is understandable in the framework presented above. However, the grief itself is a powerful stimulus that can set up its own opponent processes, and as grief subsides, periods of unexplainable high spirits may penetrate the prevailing negative mood (bringing with it a certain burden of guilt).

The point of all this is that the brain is a dynamic system that can respond quickly in terms of neuronal action potentials, but more slowly in terms of the chemical adjustment of the overall tonus of a transmitter system. These changes occur in direct response to the changes in the environment, but the properties of inertia and momentum do not always allow the changes to reflect, in a veridical manner, what is happening at a particular moment. An appreciation of these facts can make the emotional responses to a loss (or for that matter, a gain) a lot more understandable. Clinicians now expect a recurring cycle of mood changes following a loss such as a serious knee injury in an athlete. The first phase is one of denial that the injury is serious or that the loss will have a major impact on the individual's loss. This is followed by anger. The anger is followed by depression. The depression, in turn, may be followed by denial that may even take on a flavor of a high spirited, can-do attitude about coping with the injury. Most of these changes can be characterized as noncognitive in that they bear little relationship to the current changes in the outside environment. They are, almost literally, drug effects.



The terms drug addiction and drug abuse once seemed like eminently reasonable descriptive terms. The use of certain drugs produced a physical dependence, creating a situation in which the body required the presence of the drug to maintain normal physiological functions. This physical dependence on the drug was the basis for the individual's profound need for the drug, the addiction. These definitions worked fairly well for certain classes of drugs and drug users, but there was a growing list of instances in which the definitions seemed inappropriate. Tobacco use, for example, certainly involves craving, but the degree of actual dependence (i.e., physiological need) is much less dramatic than in the case of morphine or barbiturates. There is virtually no danger of death or severe symptoms of withdrawal even with complete abstinence. A distinction has sometimes been made between a drug habit and a drug addiction to reflect, in a rather rough manner, the differing physiological bases that control the use of the drug. These distinctions blur, however, with differing patterns of use, and the three-pack-a-day smoker may well have a greater physiological need than the individual who manages to limit the use of morphine. The distinction is equally blurry when one tries to draw the lines between moderate drinking, heavy drinking, and alcohol addiction.

As it became apparent that the physiological measures of dependence or the amount of drug used could not clearly define addiction, new terminology began to arise. The term addiction began to give way to the term drug abuse, which suggests a greater behavioral contribution. If an individual's use of a drug is extensive enough to interfere with work, family, or lifestyle, then the drug is being abused. There are still fuzzy edges in this definition, but the term is somewhat more realistic than the term addiction, because it reflects the pattern of use as well as the amount of drug used.

There also can be some argument concerning the term drug. Almost everyone will agree that morphine is a drug, but some will balk at considering coffee as a drug, and consensus becomes even more difficult in the case of chocolate bars, nutmeg or peanut butter. This dilemma was met with yet another evolution of the terminology, and researchers now speak of substance abuse. This too shall pass: There is a growing recognition (especially with the burgeoning business of state lotteries) that behavior itself can be the object of abuse. There is a commonality among the heroin junkies, smokers, coffee drinkers, beer drinkers, gamblers, overeaters, workaholics, and maybe even runners. They are neither inherently evil nor necessarily burdens on society, but they are all caught, to some degree, in a behavioral and pharmacological trap. We turn now to an examination of this trap.

Self Administration

Although physiological dependence may not be essential for addiction or abuse (cf., Bozarth & Wise, 1984), it certainly can be an important contributor. This is most easily seen in laboratory models of addiction in which animals are given the opportunity to self administer drugs. In some cases, the animals may simply be given access to a solution that contains the drug and allowed to freely ingest the substance. More commonly, the drug is used as a reinforcer in an operant conditioning situation as shown in Figure 9.11. A catheter may be permanently implanted into a blood vessel (e.g., the jugular vein or carotid artery), with provision made to connect the catheter to an outside source via a small tube. When the animal has fulfilled the requirements of the schedule of reinforcement, a small amount of drug is injected as a reinforcer.

In general, there is a fairly close correspondence between the list of drugs that are abused by humans and the list of drugs that animals will self administer in the laboratory. Among these are the opiates, the barbiturates, amphetamines, and some hallucinogens. The drugs that can be used as reinforcers appear to have three characteristics in common:

Reinforcement restructured.

The behavior of a rat in a self administration experiment shows many parallels to drug use in humans. In the case of morphine, for example, the initial rate of pressing the lever to obtain the drug may be very low. Gradually, over a period of days and weeks, tolerance to the morphine begins to develop (this can be demonstrated by independent testing of pain thresholds) and the lever pressing shows correspondingly greater rates in order to inject the greater amount of drug that is required to produce the "desired" effects. If the rat is given a dosage of morphine via a standard injection procedure, the amount of lever pressing is greatly decreased for the duration of the drug effect. If the rat is removed from the apparatus and withdrawn from the morphine for a period of time, physical withdrawal symptoms will be seen, and if the rat is returned to the apparatus, very high rates of lever pressing may be observed as the animal restores the morphine levels.

The administration of morphine under laboratory settings not only parallels some of the features of human drug use, but also parallels some of the aspects of conventional drives and reinforcers. As in the case of food and water, the drug can serve as a reinforcer for operant behavior. If the reinforcer is given outside of the operant setting, there will be a corresponding decrease in lever pressing, while withdrawal from the reinforcer will result in higher rates of responding when the subject is returned to the operant chamber. But there is an important difference. The drug not only acts as a reinforcer, but sets the stage for the development of the motivation to obtain the drug. Presumably, the rat has not had a lifelong yearning to obtain morphine, nor can we attribute the initial lever pressing to peer pressure or the ills of society. The first dosage of morphine appears to have some immediate reinforcing value, but more importantly, it initiates a chain of physiological events that now result in a deprivation state that was not there before: The absence of morphine is aversive. Eventually, the positive rewarding effects of morphine may pale in comparison to the aversive effects of not having the drug, and the resulting behavior may be more akin to avoidance behavior than responding for reward.

Not all drugs that have abuse potential show such close parallels between human usage and laboratory models. Researchers have found that it is almost embarrassingly difficult to get laboratory animals to consume alcohol. The taste is sufficiently aversive to the naive palate to prevent consumption in amounts that lead to tolerance, rebound effects, etc. It is usually necessary to coerce the animals to consume the alcohol by making it a part of their required food or water supply. However, once the alcohol consumption has been established, the animals will readily and voluntarily maintain the "habit". But why should it be so difficult to establish alcohol abuse in the rat while it is so difficult to prevent it in man? There is no simple answer to this question, but a more careful analysis of the drug administration procedures may provide some important clues.

Environmental bridges.

Goldberg and associates (1981) performed an experiment in which monkeys were given the opportunity to press a lever to obtain a small intravenous injection of nicotine. Although the monkeys pressed the lever enough to receive a few injections (thereby having the opportunity to experience the drug effects), the rate of pressing did not increase, but rather remained at the low level that was shown by a control group that received saline injections. Again, the failure to demonstrate self administration was curious in view of the ability of nicotine to reinforce behavior in humans. These investigators made a clever extension of their results in a second experiment: Whenever the rats earned a reinforcement, it resulted in both the drug injection and the change of a green light into amber as the subjects entered a 3-min period of darkness during which time the drug effect developed. This additional stimulus had no effect on animals that were receiving saline injections, but greatly enhanced the self administration of nicotine.

Why should the addition of an external stimulus aid the establishment of a nicotine habit? And even if it works, does it not belittle the results somewhat to have to resort to this sort of a crutch to demonstrate the drug administration? The answer to both questions may be found in a series of experiments performed by Snowden (1969). There was some controversy about whether the regulation of the amount of food eaten is controlled by the acts of chewing, tasting and swallowing, or by monitoring the caloric feedback from the food in the stomach. One way of testing these alternatives was to place the rat in a situation in which all nutrients were obtained by pressing a lever to inject liquid diet directly into the stomach. This procedure is directly comparable to that used for the self administration of drugs, and most experiments demonstrate that the rats are remarkably accurate in controlling the overall calories that are ingested in this manner. But Snowden showed that the results were not as clear-cut as they seemed. The liquid diet is prone to spoilage in these long term experiments, but the problem can easily be avoided by keeping the reservoir in an ice bath. This prevents the spoilage, but when the rat earns a reinforcement, it receives not only a small amount of diet in the stomach, but also experiences a cool tactile sensation as the liquid passes through the tube under the skin of the head and neck en route to the stomach. When Snowden warmed the liquid to body temperature before it reached the skin, the ability of the liquid diet injections to serve as a reward was greatly diminished. Why should this happen?

In both cases, the reinforcing value of a substance was enhanced by the addition of some external stimulus. The most likely explanation of these results is that the external stimulus helps to bridge the gap in time between the physical delivery of the reinforcer and the actual physiological change that results. In the real world and most laboratory situations the presence of these external mediators are the rule rather than the exception. The sight, smell, taste, and texture of food are all powerful reinforcers that signal the ultimate physiological reward, caloric energy. In terms of the immediate ability to control behavior, these harbingers of physiological change are more important than the real change. When the situation is so tightly controlled that only the physiological change can be experienced, the reinforcing value is greatly diminished. The situation becomes, in a sense, a Pavlovian delayed conditioning procedure which is successful only after many trials, if at all.

The picture that emerges is that environmental cues and behavior are an inextricable part of drug effects and of drug abuse. These behaviors become an important part of the overall pattern of abuse, even when the drug action is so fast that an external stimulus is not essential to bridge the gap. Consider, for example, the administration of nicotine by cigarette smoking. How does a drug that requires an environmental bridge in the laboratory gain such control over so many people in the natural environment? One reason is that the route of administration is ideal in terms of the speed of the effect. The inhalation of nicotine in tobacco smoke produces very rapid effects, reaching the brain within 8 seconds (Jaffe, 1980). This is even faster than an intravenous injection into the arm, and is fast enough that each puff of the cigarette can produce a discrete, detectable drug effect! This rapid delivery of distinct reinforcements serves not only to maintain the behavior, but also provides an excellent environment for the development of secondary reinforcers of associated behaviors such as manipulation of the cigarettes, oral contact, the smell of the smoke, and specific times and places (e.g., after meals, while driving, while reading the paper, etc.)

Environmental and behavioral cues are not only important contributors to the rewarding effects of drugs, but also to the motivational states that direct the organism toward specific drug effects. Certainly, a major aspect of these motivational states can be attributed directly to the physiological actions of the drug. The effects of enzyme induction, neuromodulation and rebound phenomena all contribute to an internal environment that can be "corrected" by an additional dosage of the drug. But certain aspects of these physiological changes can be influenced by learning, as we have already seen in the cases of insulin or morphine injections. Stressful environments may be especially potent in this regard because of previous situations in which engaging in the rewarded behavior (e.g., smoking a cigarette) has led to a rapid, rewarding effect. The rewarding effects may be even more pronounced with a drug such as alcohol, which has some inherent properties of anxiety reduction.

It is even possible to go a step further in the analysis of environmental cues and show that these are important not only in helping to mediate the motivational state and the rewarding effects, but also in contributing to the behavioral outcome of the drug use. A particularly intriguing example of this has been shown in a clever experimental design that was developed by John Carpenter (cf., Marlatt & Rohsenow, 1981). This design unveiled the phenomenon that has come to be known as the Think-Drink effect. The critical feature of the design was the development of a cocktail that tasted the same with or without alcohol. The recipe was four parts tonic water, one part vodka, and one part lime juice. The nonalcoholic version of this was simply five parts tonic water and one part lime juice. Preliminary tests showed that the identification of these two recipes was at chance levels, the protestations of the seasoned drinkers' palates notwithstanding. The design of the experiment, shown in Figure 9.12, was a 2 X 2 design in which the subject either received alcohol or not and were told that they were receiving alcohol or not. Thus, some of the subjects expected the effects of alcohol when it was not present, while other were not expecting the effects of alcohol when it was present. The behavioral measures (including social aggressiveness, talkativeness, motor coordination, and others) showed that the behavior was more closely related to what the participants thought they were drinking than to what they actually were drinking! Obviously, alcohol is a real drug, and with large dosages there is no way to think one's way to normalcy. However, the results of these experiments suggest that much of the stereotyped behavior associated with alcohol use may occur before the physiological effects are present or at doses which would not be sufficient to produce the behavior directly, and there is also evidence for behavioral tolerance when specific behaviors are practiced under the influence of the drug (e.g., Wenger et al, 1981).

The environmental factors become especially important when a distinction is made between use and abuse of drugs. Alcoholism is certainly one of the most costly of society's ills. The obvious solutions of voluntary abstinence or legal prohibition seem not to work. Accordingly, many researchers have turned their attention to the causes of alcoholism. One of the most interesting set of findings is that there are some subcultures that use a fairly large amount of alcohol, but have very low incidences of abuse (e.g., Aronow, 1980). Several features of alcohol use seem to be common among these groups. Children are exposed to alcohol and use alcohol at an early age, usually in the form of wine or beer as part of the meal. The parents do not become inebriated and there are strong sanctions against those who do. Inebriation is never viewed as something humorous or daring. The use of alcohol in moderation is simply taken for granted, with neither positive nor negative attributes attached. A glass of wine can be accepted or refused with the same impunity as one accepts or refuses a slice of bread. This contrasts sharply with the more traditional Middle America pattern that prohibits the use of alcohol in the young, while viewing inebriation as a source of humor ("Did you hear the one about the drunk who...") and the ability to drink as a sign of adulthood, authority and power. As children approach adulthood (or as they want to approach adulthood), they surreptitiously obtain and consume alcohol, usually in excess and almost always under conditions of stress--the perfect conditions for establishing the use of the drug for the purposes of aggrandizement and stress reduction.

Breaking the Cycle

One of the greatest ironies of humanity is that almost everyone assumes free will and control over behavior while, at the same time, ruing the fact that they cannot stop smoking, overeating, drinking coffee, gambling, drinking alcohol, taking tranquilizers, or biting fingernails. Breaking these so-called habits is one of the most difficult areas of behavior. There are those who claim it is simply a matter of will power and that they could stop at any time they really wanted to. Indeed, some do, but the rate of recidivism is high. One of the main reasons for the high rate of returning to the habit is that it has been linked to stressful situations. Abstinence is itself a stressful event, and only serves to increase the likelihood of the behavior, especially during the early stages. Schachter has claimed (on the basis of informal surveys; 1982) that the statistics are unnecessarily pessimistic because they are based upon individuals who seek professional help. His observations suggest that there are many individuals who lose weight or give up smoking without professional help and with considerably lower chances for returning to the original patterns of behavior. Whether or not these observations will hold up under more rigorous scrutiny, it is clear that there are many cases in which the behavior is especially intractable. There is no clear formulation that can guarantee success in the attempt to break drug abuse patterns, but several suggestions can be made, based upon the way in which the abuse pattern has been established and maintained.

1. Change the US effects of the drug

If drug use is viewed as a straightforward example of conditioning, then the rewarding drug effects can be considered as the unconditioned stimulus or US. One of the more obvious ways to interfere with this chain of events is to change the effects of the drug. Perhaps the best known example of this is the drug known as Antabuse, which interferes with the metabolism of alcohol. The ingestion of alcohol causes severe gastrointestinal illness when this drug is present, and it is necessary to refrain from taking Antabuse for about three days before alcohol can be consumed without experiencing these ill effects. This drug has been used with some success in clinical settings, but one of the obvious drawbacks is that the individual must take the Antabuse on a regular basis.

Another example of interference with the US effect is the substitution of methadone for heroine abuse. The methadone is not without its own potential for abuse, but the cravings for the drug and the withdrawal effects appear to be somewhat less severe and (perhaps most importantly) it is usually prescribed under careful conditions in known quantities and purity.

A third example is somewhat akin to fighting fire with fire. One of the ways in which smokers have been aided in breaking the cycle of smoking cigarettes is to produce the drug effects via a different route. Chewing gum that contains nicotine can produce the drug effects without engaging in the sequence of behaviors that has been established by smoking. Others may turn to what is by most standards (baseball players excepted) the even less socially accepted habit of taking snuff or chewing tobacco (Will restaurants adopt spitting and non-spitting sections?) Social customs aside, these are valid methods of interrupting the cycle, because it provides for the first time, a separation between the behavioral patterns and the drug effect. If it is properly guided (and there is always the danger of substituting one habit for another, or even adding one habit to the other), the individual can eliminate many of the behavior patterns without suffering the physical symptoms that might accompany abstinence.

2. Change the reward structure

To some extent, this category overlaps with the previous one, but there are ways in which the reward structure can be changed to aid in reducing some of the behavioral components of the pattern. One way which has been moderately successful (although care must be taken to avoid nicotine poisoning) is forced smoking. The individual is forced to rapidly smoke one cigarette after another, rather than leisurely puffing away in the normal manner. This has two consequences: It usually causes some degree of discomfort due to the rapid effects of the high dosage (thereby associating aversive consequences with the behavior of smoking), and it again allows a way in which the drug effects can be obtained under unusual circumstances.

This general type of technique has also been used in the case of food abuse (overeating) by attempting to limit eating strictly to mealtimes under rigid conditions.

3. Change the environmental cues

One of the hallmarks of substance abuse (it is unfortunate that the phrase abusive behavior has the wrong connotation) is that it involves a high degree of ritualization-- the cigarette after breakfast, the drink or two after work, the pretzels while watching TV. An important part of any program to eliminate habitual behaviors is to change these environmental cues whenever possible. This may involve a new schedule, such as skipping breakfast or eating breakfast at a later hour, avoiding certain locations, moving the TV to a different room, changing a work schedule, etc. Usually, this is not easy. There are too many restrictions in most lifestyles to allow very major changes. One of the major advantages of a formal clinical setting is also one of the major disadvantages: On the one hand, the new, controlled environment is a tremendous aid in helping to interrupt the patterns of behavior that have maintained the pattern of abuse. On the other hand, once the behavior has been changed and the patient leaves, it is very likely that the return to the previous environment will re-trigger all sorts of cues that have supported the habit in the past. Clinical psychology (and medical practice, for that matter) would be much simpler if the patients did not have to return to the causes of their disorders when they left the couch.

4. Avoid using the drug

This is an obvious truism, but it is mentioned here because of some recent controversy concerning alcohol consumption. Several different support groups, most notably Alcoholics Anonymous, have long advocated complete abstinence once the drinking behavior had been disrupted. Their view is that there is never a cure, and that any drinking will reestablish all of the behavioral patterns of abuse. It has been suggested that this requirement may be a bit too Spartan, and that controlled drinking might be allowable (cf., Sobell & Sobell, 1978). This seems not to be the case: In a follow-up study of 20 alcoholics who participated in the controlled drinking experiment (Pendery et al, 1982), the following dismal results were found: 1 was successful, 8 continued to have drinking problems, 6 returned to abstinence voluntarily, 4 died, and 1 was missing.

The presence of the drug in the body not only serves to reactivate some of the metabolic systems that were changed through previous exposure, but also recreates internal conditions that have been strongly associated with the relevant behavior patterns that supported the pattern of abuse. Apparently, truisms prevail.



1. The effect of a drug can be decreased or increased by changing the access of the drug molecules to the receptors, or by setting up a physiological opposition to the drug.

2. The mechanisms of tolerance (or sensitization) can change nearly any feature of the drug's actions.

3. Indirect acting drugs may lose their effectiveness rapidly through depletion of the transmitter stores.

4. As neuromodulatory processes take place, the changes in receptor number or sensitivity is reflected by a gradual change in response to the drug.

5. The presence of a drug may induce the formation of enzymes that will inactivate the drug more quickly when it is administered in the future.

6. If a drug that has been present for some time is abruptly withdrawn from the system, it unmasks compensatory reactions and opposite, rebound effects may be observed.

7. The opponent process theory applies many features of tolerance to emotional responses and to some of the phenomena of addictive behaviors.

8. In some cases, the development of tolerance requires that specific behaviors occur while the drug is in effect, a phenomenon known as behavioral tolerance.

9. The pre-post design has been used to separate the effects of pharmacological tolerance from behavioral tolerance.

10. Some aspects of behavioral tolerance may be the result of the Pavlovian conditioning of compensatory responses.

11. Drug addiction, drug abuse, and substance abuse are all terms that apply to behavior that is maintained by acquired motives.

12. Self administration procedures are used as animal models of drug abuse in humans.

13. Environmental stimuli associated with drug use may serve as important bridges for the development and maintenance of habitual drug use.

14. The effectiveness of a drug may be significantly changed by the user's expectations, as in the think-drink effect.

15. The development of substance abuse may interact with normally occurring states, especially those involving stress.

16. The drug abuse cycle may be interrupted at several points that are specified by the laws of learning.



Acquired motivation

Affective habituation

Affective contrast

Affective withdrawal


Behavioral tolerance


Compensatory response

Delirium tremens

Denervation supersensitivity

Drug effect

Drug action

Enzyme induction



Indirect action


Negative after image


Opponent process

Pavlovian conditioning

Pharmacological tolerance

Physical dependence

Pre-post design

Rebound effects

Ritualistic behaviors

Secondary reinforcer

Self administration


Substance abuse


Think-drink effect


Withdrawal effects