The work of Abraham Wikler (1973) over the past three decades has called attention to the importance of conditioned responses in the addictive process. Drugs act as powerful forces in shaping behavior, both by their direct pleasant effects (positive reinforcement) and by their effects in relieving withdrawal symptoms (negative reinforcement). Wikler theorized that the environmental cues which have been repeatedly paired with drug-induced states may become conditioned stimuli. He observed that former addicts who are free of drugs often develop tearing and yawning (opiate withdrawal signs) when they discuss drugs in group therapy. He and others subsequently showed that withdrawal signs could become conditioned in animals (Wikler and Pescor, 1967; Goldberg and Schuster, 1970). More recently, conditioned withdrawal has been demonstrated in humans (O'Brien et al, 1975; O'Brien et al, 1977). These conditioned withdrawal responses are thought to be partly due to simple pairing of pharmacological withdrawal with environmental cues, and partly due to pairing of environmental stimuli with the body's homeostatic mechanisms adapting to the onset of drug effects (Wikler, 1973; Siegel, 1974). Eventually the environmental stimuli themselves can elicit the adaptative response and this can be perceived as withdrawal.

 

As a result of this conditioning former addicts may develop withdrawal symptoms when they return to the environment in which they had previously used drugs. Not only do they develop drug-craving, but actual physical signs of sickness may occur (tearing, yawning, nausea, vomiting). There is also some evidence (Whitehead, 1974; O'Brien, 1975) that conditioned withdrawal can occur in patients maintained on a steady dose of methadone.

 

There are other conditioning factors which help to maintain self-administration of drugs. Addicts tend to use drugs in a ritualistic fashion. When the drug is administered, the withdrawal discomfort (both pharmacological and conditioned) is eased. At times, depending on the level of tolerance and dose of the drug used, the patient will feel euphoric. There is some controversy over how often a typical addict obtains euphoria rather than just relief of withdrawal distress (McAulliffee and Gordon, 1974), but clearly the reward is intermittant. Since intermittant reinforcement is very effective in maintaining behavior, this may explain why self-administration can continue when street heroin potency is low or when saline is substituted for opiate (O'Brien, 1975). Addicts who report pleasure from self-injection of inactive substances have been termed "needle freaks" (Levine, 1974). This phenomenon is usually seen only in "blind" conditions when the user expects a drug effect, but the substance administered actually contains little or no active drug. If addicts are informed that a substance is a placebo or if they know they are on an antagonist which will block drug effects, self-injection will stop after a few trials (Altman et al, 1976).

 

The existence of these conditioned phenomena requires that they be considered in the treatment of addiction. A number of innovative methods have been attempted which might influence conditioning or enable the patient to cope with conditioned withdrawal or conditioned drug craving.

 

COUNTERCONDITIONING PROCEDURES

 

Numerous published studies have described procedures to combat drug-taking behavior by aversive conditioning. These have been critically reviewed by Callner (1975). The reports consist of single cases or groups of cases, some with excellent results. Electrical and chemical aversion techniques are difficult to apply to typical addicts, and the method of covert conditioning (Cautela, 1975) has gained some acceptance.

 

Covert conditioning or sensitization consists of the use of imagined scenes as aversive events and as rewarding events. Beginning with the initial craving for drugs the patient is asked to imagine as clearly as possible each link in the chain of events leading to drug-taking. Instead of pleasant drug effects, the patient is asked to imagine becoming severely ill in vivid terms as a consequence of the drug. When the patient imagines the avoidance of drugs he is told to imagine pleasant scenes. Covert sensitization has been successfully used in several published case reports or series of cases (Wisocki, 1973; Cautela and Rosenstein, 1975). It has been used in combination with other behavioral techniques (O'Brien et al, 1972) and it has been successfully augmented by hypnosis (Copeman, 1977) and by chemical aversion (Maletzky, 1974).

 

EXTINCTION OR DESENSITIZATION PROCEDURES

 

One approach to treatment is to attempt the systematic extinction of conditioned responses which have developed throughout the course of addiction. This method directly follows from the assumption that conditioned responses are helping to maintain drug-taking behavior. Thus an attempt to extinguish these responses would appear logical. To accomplish this, exposure to the situations associated with drug-taking should be accomplished, but drug effects must not follow. Kraft (1970) in a series of case reports of amphetamine-barbiturate users found certain social situations to provoke drug-taking. His treatment consisted of setting up hierarchies of such situations for each patient (Wolpe, 1958) and exposing the patient to each one beginning with the least evocative situation. Hypnosis and relaxation were used to prevent the patient from responding in the usual way with anxiety and drug-ingestion. Although definition of population and follow-up were minimal (Kraft, 1976), the results were encouraging.

 

Our group (O'Brien, 1975) has been able to obtain (by behavioral interview) hierarchies of stimuli which provoke craving or withdrawal responses among addicts. When slides or video tapes of these situations were shown to addict patients, the response was variable. Some reported strong drug-craving and others reported no effects (O'Brien et al, 1974). In subsequent studies we have thus far failed to find stimuli which are consistently effective. Negative affectual responses and slight feelings of withdrawal have been reported by Tea'sdale (4973) when drug related slides were shown to addicts. Physiological changes in response to video tapes of drug taking behavior have been reported by Sideroff and Jarvik (1977).

 

We have conducted several studies of detoxified opiate addicts and control subjects under baseline conditions in a controlled environment. While some addicts showed conditioned withdrawal responses to slides and video tapes of drug-related activities, more consistent responses occurred when the subjects handled bags of heroin or went through pre-injection rituals ("cook-up"). Conditioned withdrawal responses were: decreases in skin temperature, increases in heart rate, increases in pupil size, and increased scores on withdrawal and craving ratings (O'Brien et al, Note 1). If a simple technique could be found for producing conditioned craving or withdrawal in the clinic, patients could be systematically desensitized. Of course, it is not certain how much generalization there would be to the natural environment. For optimum effectiveness, it might be necessary for patients to be desensitized in situations that clearly resemble their own neighborhoods.

 

Narcotic antagonists have also been considered as an aid to extinction of drug conditioned responses (Wikler, 1974). While a patient is receiving an antagonist, such as naltrexone, he is effectively "protected" from most of the effects of injected opiates. Thus he can be exposed to stimuli which provoke opiate use and even use opiates with little or no reinforcement. However, most patients who are maintained on narcotic antagonists rarely test them by injecting heroin. This is confirmed by urine tests as well as by patients reports (Kleber et al, 1974; O'Brien et al, 1974). The patients report that once they are convinced that opiate effects will be blocked by the antagonist, they do not wish to waste their money by using heroin. Since the addicts do not perform the rituals which usually precede drug-taking, they do not actually extinguish the conditioned responses associated with this behavior.

 

Another problem is that duration of treatment with naltrexone is usually fairly brief (National Research Council, 1978) and thus opportunities for extinction are limited. The antagonist is lacking in the reinforcing properties found in agonist such as methadone. It produces no euphoria and no physical dependence. In the absence of pharmacological reinforcement, small monetary rewards have been recommended to induce patients to continue ingesting naltrexone (Meyer et al, 1976). We have found the scheduling of the monetary reward to be important (Grabowski et al, 1978, Note 2). Both response-based schedules (continuous reinforcement, fixed ratio, variable ratio) and a time-based schedule (fixed interval) were compared. The contingency payments (a total of $40 per month for all schedules) produced a significant lengthening of duration of naltrexone treatment. The fixed ratio condition ($3.34 every third visit) was found to be optimal. Although the contingency payments produced a clear effect on treatment duration, our preliminary follow-up data suggest that active extinction,of conditioned responses is necessary to improve overall treatment success rate.

 

How can former addicts be induced to go through pre-injection rituals while protected by naltrexone? Meyer et al (1976) have reported that hospitalized former addicts on naltrexone usually did not continue to inject heroin spontaneously even when the drug is readily available at minimal cost. They did resume injecting when naltrexone was stopped. Our group (O'Brien et al, 1974; O'Brien, 1975) has attempted to actively extinguish pre-injection rituals and self-injection behavior as an experimental treatment. Former addicts maintained on antagonist (cyclazocine or naltrexone) were given opiates to self-inject on a regular prescribed basis in the laboratory. The behavior of these addicts changed remarkably over 5 to 25 self-injections and this occurred whether the injections contained saline or opiate (double-blind). The entire procedure (including pre- and post-injection rituals) was reported as mildly pleasant after the first several injections. The pre-injection rituals themselves produced craving and withdrawal as measured by rating scales and physiological responses. The injection itself relieved withdrawal and produced weak opioid effects on the first several occasions. As extinction progressed, the withdrawal responses increased and were worsened rather than relieved by the self-injection. Some patients became angry and many refused to continue the injections despite cash inducements.

 

Although the blocked self-injections resulted in the patients' no longer enjoying the drug-taking rituals, it was found that long-term outcome did not appear to have been affected. Recently, however, we have analyzed data from 118 patients who were followed-up at one month and six months after stopping naltrexone (O'Brien et al, Note 3). Ninety-seven patients received naltrexone and no behavioral treatment. Twenty-one were randomly assigned to self-injection ("extinction") using either an opiate (hydromorphone) or saline under double-blind conditions. There were no significant differences among treatment groups in age, education, sex, duration of addiction, criminal involvement, or prior treatment history. All patients received supportive counseling during treatment. The duration of treatment was approximately 60 days for all treatment groups except for those who self-injected saline. The saline group more rapidly developed craving and withdrawal responses and dropped out of treatment after a mean of only 31 days. Six months after stopping treatment, approximately 84 percent of patients were located for a follow-up interview, and urine test. Outcome was determined by employment, drug use, criminal activity, and social functioning measures. Patients who self-injected opiate in the laboratory while "blocked" by naltrexone had a significantly better outcome than controls who received no behavior therapy. Those who received saline had a significantly worse outcome than controls.

 

These results are only preliminary; they must be replicated by a larger study which eliminates some of the flaws in the initial study. The data suggest, however, that attempts to extinguish putative conditioned responses in opiate addicts can have beneficial effects or detrimental effects depending on how extinction procedures are applied. We are now testing a more gradual desensitization procedure which may lead to longer treatment duration and better outcome.

 

CONCLUSION

 

There is considerable evidence from animal and human studies that conditioning plays a role in maintaining drug-seeking behavior. Systematic efforts to modify drug conditioned bheavior as a therapeutic modality are still in their early stages. Although the initial results are promising, it must be remembered that human drug addiction is a complex process involving political and social as well as psychological and pharmacological factors. Behavioral treatments must be accompanied by broad rehabilitative measures.

 

NOTES

 

1. O'Brien CP, Ternes J and Greenstein R: Conditioned responses to drug-related stimuli in addicts and former addicts. (Manuscript in preparation)

2. Grabowski J, O'Brien CP, Greenstein R, Long M, Steinberg-Donato, S and Ternes J: Effects of contingent payment on compliance with a naltrexone regiman. (Manuscript in preparation)

3. O'Brien CP, Greenstein R, McLellan T, Ternes J and Woody G: Follow-up of opiate addicts treated by naltrexone and extinction of conditioned responses. (Manuscript in preparation)

 

REFERENCES

 

Altman JL, Meyer RE, Mirin SM and McNamee HB: Opiate antagonists and the modification of heroin self-administration behavior in man. Int J Addiction 11(3):485-499,1976

Caliner DA: Behavioral treatment approaches to drug abuse: A critical review of the research. Psycho! Bull 82:143-164, 1975

Cautela JR and Rosenstein AK: The use of covert conditioning in the treatment of drug abuse. Int J Addiction 10:277-303, 1975

Copemann CD: Treatment of polydrug abuse and addiction by covert sensitization: Some contraindications. Int J Addictions 12:17-23, 1977

Goldberg SR and Schuster CR: Conditioned nalorphine-induced abstinence changes; persistence in post-morphine dependent monkeys. J Exper Anal Behav 14:33-46, 1970

Kleber H, Kinsella JK, Riordan C, Greaves S and Sweney D: The use of cyclozocine in treating narcotic addicts in a low-intervention setting. Arch Gen Psychiatry 30:37-42, 1974

Kraft T: Treatment of drinamyl addiction. J Nery Ment Disorder 150:138-144,1970 Kraft T: Long-term effects of behavior therapy. Br J Psychiatry 510-511, 1976

Lal H, Miksic S and Smith N: Naloxone antagonism of conditioned hyperthermia: An evidence for release of endogenous opioid. Life Science 18:971-976, 1976 Levine, DG: Needle freaks: Compulsive self-injection by drug users. Am J Psychiatry 131:297-299, 1974

Maletzky BM: Assisted covert sensitization for drug abuse. Int J Addictions 9:411-429, 1974

McAulliffee WE and Gordon RA: A test of Lindesmith's theory of addiction: The frequency of euphoria among long-term addicts. Am J Sociol 79:795-840, 1974

Meyer R, Randall M, Barrington C, Mirin S and Greenberg I: Limitations of an extinction approach to narcotic antagonist treatment. Nat Inst Drug Abuse, Res Mon Series (9):123-135, 1976

National Research Committee: Clinical evaluation of naltrexone treatment of opiate dependent individuals. Arch Gen Psychiatry 35:355-340, 1978

O'Brien CP: Experimental analysis of conditioning factors in human narcotic addiction. Pharmacl Rev 27:533-543, 1975

O'Brien CP, Chaddock B, Woody G and Greenstein R: Systematic extinction of narcotic drug use using narcotic antagonists. Pro NAS/NRC Corn Probl Drug Depend 216-222, Washington, D.C., 1974

O'Brien CP, O'Brien TJ, Mintz J and Brady JP: Conditioning of narcotic abstinence symptoms in human subjects. Drug Alco Depend 115-123, 1975

O'Brien CP, Testa T, O'Brien TJ, Brady JP and Wells B: Narcotic withdrawal in humans. Science 195:1000-1002, 1977

O'Brien JS, Raynes AE and Patch VD: Treatment of heroin addiction with aversion therapy relaxation training and systematic desitization. Behav Res Ther 10:77-80, 1972

Sideroff SI and Jarvik ME: Conditioned responses to a video tape showing heroin

related stimuli. Pro Nat Drug Con Drug Abuse May 5-9, 1977, (in press) Siegal S: Evidence from rats that morphine tolerance is a learned response. J Compar

Physiol Psych 89(5):498-506, 1975

Teasdale JD: Conditioned abstinence in narcotic addicts. Int J Addictions 8:274-292, 1974

Whitehead C: Methadone pseudowithdrawal syndrome: Paradign for a psycho-

pharmacological model of opiate addiction. Psychosom Med 35:189-198, 1974 Wikler A: Dynamics of drug dependence. Arch General Psychiatry 28:611-616, 1973 Wikler A: Conditioning of successive adaptive responses to the initial effects of drugs.

Conditional Reflex 8:193-210, 1973

Wikler A: Requirements for extinction of relapse-facilitating variables and for rehabilitation in a narcotic antagonist treatment program. In: Narcotic Antagonists, MC Braude, et al (eds), New York: Raven Press, 339-414, 1974

Wikler A and Prescor F: Classical conditioning of a morphine abstinence phenomenon, reinforcement of opioid-drinking behavior and "relapse" in morphine-addicted rats. Psychopharmacologia 10:255-284, 1967

Wisocki PA: The successful treatment of a heroin addict by covert conditioning techniques. J Behav Ther Experi Psychiatry 4:55-61, 1973

Wolpe J: Psychotherapy by Reciprocal Inhibition, pp 139-165. Stanford, California: Stanford University Press, 1958