decrease or eliminate the sources of stress. The belief that they are powerless to cope with stress is the major cognitive distortion of drug abusers. One consequence of this is the intense feeling of low selfesteem that is a well-known clinical entity among drug abusers (Krystal and Raskin 1970). Feelings of self-depreciation, which form the belief that one is powerless, represent the affective component of the CAP theory.

The experience of anxiety is, of course, uncomfortable, and a means of anxiety reduction is necessary. A primary pharmacogenic effect of heroin is anxiety reduction. Not only does the drug provide relief from anxiety, but the individual obtains a temporary ecstatic feeling--a "high." Under the influence of the drug the individual temporarily experiences an increased sense of power, control, and well being. The sense of powerlessness is replaced by an exaggerated sense of being all powerful--no task is too great and no feat impossible while "high." Thus, drugs can do for abusers what they believe they cannot do for themselves: get rid of anxiety, lead to good feeling about themselves, and make them believe they are competent, in control, and able to master their environment.

Unfortunately for the drug abuser, the drug effects are short lived and any temporary gains turn into long-term losses. Inevitably, after the high wears off some internal or external source of stress will rekindle the conflict and anxiety. Not only do the old feelings of lack of control return but they are likely to be even stronger than before. It is this increasing sense of powerlessness with increased drug use that leads the individual from drug use to abuse. Each time drug users rely on a drug to relieve tension and feel good about themselves, they become a little less capable of coping on their own. By using drugs to cope, the individual is cut off from learning other more adaptive coping mechanisms and becomes less tolerant of the pain of anxiety. The drug user now knows that anxiety does not have to be tolerated, as drug taking has been successful in the past in removing tension and producing good feelings. It is therefore expected that drug use will increase both in frequency and in the number of different situations in which it is employed. For example, arguments with parents may be a primary source of conflict and anxiety for the adolescent drug abuser. Drug taking will frequently follow such an argument. An adolescent experiencing school-related stress, having learned that drug taking is an effective means of anxiety reduction, may turn to additional drug taking to compensate for academic failures. The reliance on drugs to cope with stress therefore creates a vicious cycle; the more drugs are used, the more the individual believes they are necessary. Each drug experience serves to confirm for users the belief that they are powerless to function on their own.

The CAP model of drug abuse also makes several assumptions about the treatment of drug abuse. First, effective and lasting change is based on learning that behavior has consequences and that one can have an effect on his or her own life. To replace a sense of powerlessness with a sense of mastery, the abuser has to be taught alternative ways of responding to external or internal stress. These alternative ways cannot, however, be developed, practiced, and adopted as long as the individual continues to use drugs.

A second assumption is that an effective treatment plan must be multimodal (Lazarus 1976). A complete treatment plan must assess not only the overt behavior of drug taking but the negative emotions (e.g.,

anxiety), unpleasant physical sensations (e.g., aches and pains that accompany withdrawal), intrusive images (e.g., recollections of past failures), faulty cognitions (e.g., "nothing I do will ever be successful"), and interpersonal inadequacies (e.g., difficulty in making friends with non-drug-taking peers). Each of the individual's problem areas may require a specific treatment strategy. For example, systematic desensitization may be used to help the abuser cope with anxiety, while cognitive restructuring may be needed to correct the faulty cognitive processes.

The multimodal therapy approach is consistent with the CAP theory in that both stress the interaction between personality modalities, and both suggest that in complex human problems a lasting result depends upon addressing all relevant aspects of the individual's functioning. The high recidivism rate, characteristic of drug abuser treatment, may be due to treatment focusing on a limited aspect of the abuser's overall personality functioning and lifestyle (Platt and Labate 1976).

RESEARCH SUPPORT FOR THE CAP THEORY

The CAP theory of drug abuse was developed primarily on experiences gained working with adolescent drug abusers at Holy Cross Campus, a coed residential treatment center in Rhinecliff, New York (Coghlan et al. 1973). To evaluate the effectiveness of the treatment program and the CAP model, adolescents completed two personality tests, once approximately 30 days after admission and again six months later (Gold and Coghlan 1976). The Rotter Locus of Control (1-E) Scale (Rotter 1966) was used to assess whether an individual believed reinforcement to be contingent on personal efforts and behavior (internal control) or a result of luck, fate, chance, or more powerful others (external control). A second scale, the Self-Esteem Survey (SES) was also used as a measure of self-evaluation (Coopersmith 1967). It was predicted that after six months in residential treatment the adolescents would move toward more internal control and greater self-esteem. Data based on 32 males and 21 females provided some support for the hypotheses. Females became significantly more internally oriented. Both males' and females' scores on the SES reflected higher self-esteem, though the change was not statistically significant. A second important finding was a significant correlation for the females between low self-esteem and both running away and self-destructive acts (Gold and Coghlan 1976).

The role of perceived control has been examined in a series of studies by Seligman and his associates (Seligman 1975; Maier and Seligman 1976). A belief in external causation or control may dramatically impair learning and functioning. The research paradigm is as follows: One group of subjects is exposed to a situation in which their behavior can control the occurrence of an aversive event, while another group experiences the same situation except that the aversive event is beyond their control. When both groups are next presented with a new situation in which learning is required, the typical finding is that people who previously experienced control learn faster in the new situation. Moreover, some subjects, after experiencing the lack of control, may not learn at all even though the task is often quite simple. Seligman (1975) interprets such findings as indicating that, when an organism's behavior has no effect on its environment, "learned helplessness" is the result. The learned-helplessness theory has been suggested as a

model for the development of reactive depression. It also points out a way in which the sense of helplessness or powerlessness may be a characteristic of drug abusers. Individuals prone to drug abuse may be those who have a history of lack of relationship between their responses and consequences--a series of learning experiences which teach them they are not effective in altering or influencing their environment. For example, studying may have no effect on grades received; behaving as demanded by parents may not lead to being loved; hard work may not lead to a promotion or better job; etc. The similarities between a model of reactive depression and drug abuse are not surprising, as there are aspects of drug abuse that parallel depression. Drug abuse can be described as a self-destructive activity and often is clinically viewed as a form of "slow suicide." Gold and Coghlan (1976) found a relationship between adolescent female abusers' belief in external control and low self-esteem with overt self-destructive behavior. Wetzel (1976) studied 154 suicide attempters, threateners, and psychiatric controls and found that a sense of hopelessness was highly correlated with suicidal behavior, even more so than depth of depression.

The effects of perceived control have also been studied with reference to coping with aversive stimulation. For example, Geer et al. (1970) found that college students who falsely believed they had control over the duration of shocks received displayed less physiological response to the shock. The finding of less arousal suggests that the shocks were becoming less stressful for them. Turk (1975) trained volunteers to develop different coping strategies to deal with pain to encourage them to believe they could successfully manage it. Cognitively trained subjects were able to tolerate the pain for almost twice as long as untrained subjects.

In summary, the CAP theory of drug abuse emphasizes the interaction of cognitive-affective-pharmacogenic effects of drug taking. The belief that one is powerless to affect the environment and cope with stress plays a central role in the theory. The CAP theory is seen as being consistent with newer cognitive models which emphasize the role of internal thoughts and beliefs in the development of maladaptive behavior. Research findings support the hypothesis that an individual's belief in the ability to control a situation strongly influences behavior. Successful treatment of the drug abuser requires a multimodal approach which alters faulty thinking, teaches new interpersonal skills, helps the abuser cope with pain and anxiety, and encourages the development of a positive self-image.

The Bad-Habit Theory of Drug Abuse

Donald W. Goodwin, M.D.

INTRODUCTION

By "bad habit" I refer to repetitious, harmful, semireflexive behavior resulting from classical conditioning in "susceptible" individuals. With regard to drugs, "susceptibility" may be specific for certain drugs or nonspecific, i.e., the individual may be susceptible to abusing a number of drugs, perhaps only in certain classes (e.g., the sedativehypnotics) or perhaps across classes (e.g., opiates, sedative-hypnotics, nicotine, etc.). Susceptibility may be partly inherited (under some degree of genetic control), or it may reflect purely psychosocial influences, or both. These issues are complicated, and a global theory of addiction may be premature. My theory is limited to alcoholism, but I have included a brief discussion of the possibility that theories of alcoholism may help to explain other forms of substance abuse.

WHAT IS INHERITED?

Perhaps the strongest evidence for a genetic factor in alcoholism is the evidence that alcoholism strongly runs in families (Cotton 1979). This, combined with findings from twin and adoption studies, at least suggests the possibility of a hereditary factor (Goodwin 1979). If so, what is inherited?

Certain behaviors associated with drinking must be explained before it is known why serious drinking problems develop in perhaps one of 12 or 15 drinkers in Western countries. These core features must be explained: (1) loss of control, (2) tendency to relapse, and (3) tolerance. The following explanations blend possible genetic and nongenetic factors.

Indisputably, there is a wide range of innate variations in response to alcohol. This is true in humans and every species studied. There are not only strain and species differences but also differences between

individuals. It is difficult to account for this variation other than to ascribe it to innate, probably genetically controlled influences.

In humans, the most conspicuous example of innate variation in alcohol response has been shown in Orientals, whose low alcoholism rates have usually been attributed to social factors. However, three studies have now shown that small amounts of alcohol cause a cutaneous flush and unpleasant reactions in about three-quarters of Orientals (Wolff 1973; Ewing et al. 1974; Seto et al. 1978), indicating that a large number of Orientals are physiologically intolerant of alcohol. The biochemical basis for these adverse reactions has not been determined, but recent data indicate a high frequency of atypical liver alcohol dehydrogenase among Japanese (Stamatoyannopoulas et al. 1975). This coenzyme may alter the metabolism of alcohol, leading to increased formation of acetaldehyde, and this may explain the flush and other ill effects (such as nausea).

Other groups with relatively low alcoholism rates may be similarly protected by an innate sensitivity to alcohol. For example, fewer women than men are alcoholic, and one study reports that women have higher blood alcohol levels after ingesting a given amount of alcohol than do men (Jones and Jones 1976). Informal surveys suggest that a substantial proportion of women experience unpleasant physical effects after modest amounts of alcohol (e.g., nausea and headache). Anecdotal evidence also suggests that more Jews than non-Jews have adverse physical reactions to modest amounts of alcohol, which may contribute to the low prevalence of alcoholism among Jews.

It is obviously essential to be able to drink large quantities of alcohol to be alcoholic. Many people are prevented from this because of innate cutoff points almost certainly under genetic control. That genetic control is an important factor in drug metabolism in general has been demonstrated by numerous studies showing that identical twins metabolize a wide variety of drugs (including alcohol) at almost identical rates, while fraternal twins have widely disparate rates of metabolism (Vesell et al. 1971). Whether the development of alcoholism is also subject to some genetic control remains conjectural.

It is widely believed that tolerance to alcohol is acquired mainly from "practice"; the more a person drinks, the more he or she needs to drink to get the same effect. With opiates, this clearly is true; with alcohol, it is not so clearly true. Animals fail to show much tolerance to alcohol, even after repeated exposure. Also, young men with almost no prior drinking experience vary widely in their response to alcohol in experimental studies (Goodwin et al. 1969). Some show almost no effect, while others are quite easily intoxicated. Since this variability does not correlate with prior drinking history, the only other explanation is that innate biological factors are responsible. To summarize, large numbers of people are more or less "protected" from becoming alcoholic because of genetically determined adverse physical reactions to alcohol. If anything is inherited in alcoholism, it is probably the lack of intolerance for alcohol. (Parenthetically, it is interesting that Alcoholics Anonymous often refers to allergy as a factor in alcoholism, usually properly bracketing "allergy" in quotation marks. It now seems that this is indeed true, but it is the nonalcoholics, not the alcoholics, who are allergic!)