when it is remembered that all students in the analysis were nonusers at the time the predictor variables were measured. Nevertheless, 20 percent of the students in the analysis were misclassified, and those 20 percent reflect various aspects of individual uniqueness not captured in the analysis.

SUBSTANCE USE AND THE INTERACTION

BETWEEN EARLY PERSONALITY DEVELOPMENT
AND PEER-GROUP INFLUENCES

Friendship groups begin to form in the primary school grades, and it is likely that the behavioral predispositions of children comprising any given group tend to converge as members of the group share with each other the perceptions, experiences, values, beliefs, and lifeorienting conclusions that influence personality development. Children with similar values, attitudes, and other personal characteristics gravitate toward each other; and that association strengthens the very characteristics that brought them together in the first place.

Children with personality characteristics that promote rejection of adult demands and expectations exhibit that rejection in many ways: e.g., disparaging academic achievement, smoking cigarettes, breaking school rules, and engaging in other types of early childhood deviance. Such children tend to aggregate and form friendship groups, some members of which are precocious regarding both their motivation to use substances and their ability to find sources of supply.

Similarly, children with personality characteristics that facilitate acceptance of, and/or compliance with, the rules and expectations of adult authorities tend to become members of friendship groups that support further development of those characteristics; and such groups are likely to contain fewer members who are precocious regarding access to, and motivation for, substance use.

Thus, early in preadolescence, an interactive process begins that is influenced by (a) personality formation of individual children; (b) reinforcement of that formation through interaction with like-minded children; (c) differences among groups regarding attitudes toward, and the use of, substances; and (d) differential availability of substances to such groups. We believe this interactive process contributes substantially to the considerable success with which substance use can be predicted from personality characteristics and attitudes measured prior to initiation of use.

FACTORS CONTRIBUTING TO
CONTINUATION OF USE

The match between the needs of the user and the changes he or she attributes to the substance is important in determining whether or not use will continue. The individual who places high value on feeling strong, alert, decisive, and masterful is apt to find amphetamine or cocaine much more satisfying than a person who emphasizes peace, physical relaxation, and the contemplation of philosophical and

metaphysical issues. A person of the latter type would probably find drugs like marijuana and LSD far more enjoyable. The better the match between the perceived substance effects and the user's needs, the more likely use is to continue.

Future use is also influenced by the intensity of the needs that are perceived as being satisfied by use. The greater the importance ascribed by the user to these needs, the more likely it is that use will continue.

The mood and cognitive changes caused by use of certain substances can temporarily alter the user's concepts of Self and Ideal Self. If use reduces the discrepancy between the user's perceptions of Self and Ideal Self, continuation of use is likely--even if those changes last only as long as the drug effect itself.

It is also possible for substance use to produce changes in personality that are more or less enduring; e.g., increased sociability and improved social skills in an adolescent who previously was painfully shy. If such changes are highly valued by the user, the probability of continued use will be increased substantially.

During the relatively early phases of escalation toward compulsive use, it is possible for consciously recognized dangers that are associated with substance use to facilitate rather than inhibit use if those dangers are experienced as more exhilarating than anxiety-provoking; if the self-initiated risks bring status and social approval to the user; or if the user pits any perceived dangers against his or her competence and self-control, and then treats the matter as a contest which he or she is sure to win. As long as the user continues to perceive the overall gain as greater than the overall cost, use will continue; and the risk of escalation to more dangerous levels of use becomes more likely.

It should also be noted that some behavior that appears to be completely self-defeating might in fact be aimed at achieving objectives that simply are not easily recognized by an outside observer. For that matter, they might not be recognized by the actor. The adolescent who (for whatever reason) has a strong need to punish the Self, a parent, or some other significant person might find the agonizing costs of compulsive substance use more than offset by the benefits produced by the punishment inflicted.

FACTORS CONTRIBUTING TO
CESSATION OF USE

Although cessation itself is a single event, we assume that it reflects the outcome of a protracted process of assessment that has been ongoing (consciously and unconsciously) throughout most of the period of use. Factors that determine when (if ever) the advantages of cessation will be seen as outweighing the disadvantages include the following: changes in the user's life circumstances; increasing anxiety and concern regarding various potential losses associated with use; substitution of more cost-effective satisfactions for those previously obtained through substance use; increased attribution of importance to longer term costs and benefits associated with use; and a clearer recognition of the obstacles to achievement of important life goals posed by continuation of use.

Among children and young adults, examples of altered life circumstances that might facilitate cessation are moving from one neighborhood to another; changing friendship groups; graduating from high school; going to college; getting a full-time job; getting married; having children; and accepting new responsibilities associated with adulthood. Anxieties and concerns that might lead to cessation include conflicts with parents, school authorities, and police regarding substance use; having a severely frightening drug experience or series of such experiences; fear of losing a valued job or jeopardizing one's career advancement; concern over the possibility of having a serious accident or suffering impaired physical or psychological health; fear of losing the respect and esteem of loved ones and friends; reduced self-respect; and fear that an immediate choice must be made between cessation now or a lifelong dependency on substance use.

Certain patterns of heavy substance use can cause hobbies, sports activities, and other previously enjoyable ways of spending time to become less rewarding. Success in rekindling those earlier interests, or in developing new ones, is apt to increase the likelihood that use will cease.

The probability of cessation is increased by any shift in orientation away from the present toward the future, or by any increased capacity to forego immediate gratifications to achieve more important subsequent ones. That probability is also increased if the user views continuation as being incompatible with achievement of long-term, significant life goals, especially if those goals are part of a clearly defined, carefully considered career plan that seems both achievable and likely to bring important future occupational, financial, social, and personal satisfactions.

IMPAIRED REALITY TESTING,
COMPULSIVE SUBSTANCE USE,
ADDICTION, AND READDICTION

Whatever its amount, frequency, and pattern, substance use will continue until the user perceives the disadvantages of use as outweighing its benefits. The subjective character of this cost-benefit relationship is emphasized once again because in many (perhaps most) instances of compulsive use, the user perceives use as having a net positive effect long after most outside observers would have concluded that the cost-benefit relationship had shifted from positive to negative.

As escalation progresses, cognitive functions (perception, memory, and judgment) tend to be altered in a manner that restricts and vitiates the feedback available to the user regarding the benefits and costs of use. This undermines the reality testing processes that might otherwise alert the user to his or her increasing vulnerability to addiction. Convictions based on early evidence that the aggregate net effect of substance use is positive may cause new and contradictory evidence to be discounted, misinterpreted, or denied altogether.

It is well known that memory is highly selective. This may be important in explaining why addicts fight and win the agonizing battle to become free of addiction, only to become readdicted after a period of

abstinence. Perhaps the suffering is remembered as being less intense than it actually was. The likelihood of readdiction is increased by such retrospective cognitive distortions or by any other failures in reality testing that cause the recollection of past negative consequences to appear diminished in importance or that cause the recollection of past positive consequences to appear enhanced in importance.

Impaired reality testing might also promote readdiction by enabling the user to believe, erroneously, that the factors accounting for his or her previous addiction no longer apply. For example, the user might believe that he or she is now clearly aware of the warning signs that appear prior to the stage of compulsive use, will vigilantly heed any such warnings, and, in that manner, can achieve the pleasure of occasional, well-regulated, noncompulsive use without running the risk of readdiction. Or, if the individual's abstinent periods are themselves psychologically distressing (due to depression, anxiety, guilt, anger, etc.) and substance use reduces those discomforts, it might be quite easy for the user to misjudge the risks of readdiction and conclude that just enough substance can be taken to control those distressing mood states without returning to the level of compulsive use. Impaired reality testing may also play a role in allowing the user to accord undue importance to immediate gratifications at the expense of more distant ones. Continued use is facilitated by ambiguity of longterm goals; by undervaluing either their importance or their likelihood of attainment; and by failing to recognize the relationship between continued use and the likelihood of achieving those goals. If support for the belief that substance use has a net positive effect becomes sufficiently weak, then the defenses that previously permitted the user to discount, misinterpret, or deny the true costs of substance use become harder and harder to sustain. The self-deception may then be recognized, and use may cease.

SUBSTANCE-INDUCED CHANGES IN MOOD
AND SOMATIC FEELING STATES

Although most aspects of this theory concern mood states rather than somatic feelings, the latter are very important in determining usage patterns. Present information concerning the separate and interactive roles of mood and somatic feeling states in sustaining substance use is meager--even with a substance as widely used and as frequently studied as cigarettes. This is one of many issues regarding substance use that will require further investigation.

WITHDRAWAL DISTRESS AND THE
SELF-PERPETUATION OF USE

Substances differ regarding the production of negative mood and somatic feeling states after their use. They also differ regarding the success with which such effects can be reduced by readministration of the original substance. The nervousness and jittery feelings that result from excessive consumption of caffeine are increased, not reduced, by ingesting additional caffeine; but those and other symptoms of

excessive alcohol consumption can be reduced by taking additional alcohol. Escalation to compulsive use is a danger with any substance that can be ingested to alleviate withdrawal distress resulting from previous ingestion--particularly if the substance is one for which tolerance develops rapidly, with a resultant need for higher and higher dose levels to produce a given effect. It is well known, for example, that the aversiveness of withdrawal distress is powerfully important in driving the heroin addict to readminister.

Of course, the amount of substance used (and other factors, such as the route of administration) influences the likelihood that a user will be drawn into a cycle of self-perpetuating compulsive use. Cocaine, as presently used in the United States, rarely generates compulsive use, but it has been reported that in Peru and other South American countries, where coca paste is inexpensive and is smoked in large quantities, some users are catapulted to levels of intensely compulsive use with frightening rapidity (Jeri et al. 1978).