INTRODUCTION

 

Many drugs that affect the central nervous system can maintain behavior that leads to their injection; the injection thus functions as a reinforcer (1,2,3). As with any environmental event, the suitability of a drug injection to function as a reinforcer depends on a number of factors, including the antecedent behavior, the injected drug, and the scheduled temporal relations between them, as well as the history of drug administration. In most experiments with animals, behavior has been maintained under simple fixed-ratio schedules where every nth response produced a drug injection (2-6). The patterns of responding maintained under simple fixed-ratio schedules can be used as components of more complex second-order schedules of drug injection (5,7-9). Under a second-order schedule, a pattern of responding resulting from the operation of one schedule is treated as a unitary response that is itself maintained according to a second schedule. Second-order schedules of drug injection are of particular interest because they introduce a higher order of intermittency between drug injections and responses and allow evaluation of the roles of both drug injections and environmental stimuli associated with drug injections in the maintenance of drug-seeking behavior. These schedules are analogous in many ways to the complex schedules that control drug-seeking behavior in humans. With human addicts, intermittent injections of drug maintain the long sequence of behavioral activities involved in procurement, preparation and injection. Environmental stimuli associated with this behavioral sequence play an important role in the control of the behavior (10-12).

 

In one series of experiments, squirrel monkeys with permanent venous catheters responded under a second-order schedule where every thirtieth key press during a five-minute interval of time produced a two-second flash of a light (30-response fixed-ratio component); the first fixed-ratio component completed after the five-minute interval elapsed produced both the light and an intravenous injection of 100 pg/kg cocaine (7-9). This sequence was repeated fifteen times during each daily session. Although drug injections were highly intermittent relative to key presses, repeated sequences of rapid key pressing were maintained during each daily session. From 400 to 1000 responses preceded each injection of cocaine, and mean response rates exceeded one response per second in all monkeys. Also, the brief lights controlled patterns of responding characteristic of fixed-ratio schedules; a pause in responding after each light presentation was followed by an abrupt change to rapid responding until the light was produced again. The rates and patterns of responding controlled by the brief lights under this second-order schedule were similar to those controlled by intravenous injections of cocaine under simple fixed-ratio schedules where completion of every 30-response fixed-ratio component resulted in an injection of cocaine. When the brief lights were omitted but the frequency of cocaine injections was unchanged, rates of responding decreased and patterns of responding were disrupted. Similarly, when saline injections were substituted for cocaine injections, rates of responding decreased and patterns of responding were disrupted. Thus, both drug injections and stimuli associated with drug injections were important in the maintenance of the behavior. Under second-order schedules of intravenous cocaine injection with fixed-ratio components, high rates of responding can be consistently maintained throughout experimental sessions; moreover, there is strong moment-to-moment control of characteristic fixed-ratio patterns of responding.

 

High rates of responding and fixed-ratio patterns of responding also can be maintained under second-order schedules by intravenous injections of morphine (13-15). The effects of morphine under these schedule conditions are of particular interest because morphine has usually failed to maintain high rates of responding under simple fixed-ratio schedules (2,16). This chapter reviews experiments by Goldberg (13,14) and Goldberg and Tang (15) that studied rhesus and squirrel monkeys under a second-order schedule in which intravenous injection of morphine occurred only at the end of each session and twenty-three hours or more elapsed between sessions, providing a separation between the reinforcing effects of morphine and its other pharmacological effects which may suppress behavior. Behavior of rhesus monkeys can be maintained under a similar second-order schedule by intramuscular injections of morphine (17). Since the route of administration may be a critical factor in using morphine to maintain behavior, this chapter also reviews these experiments.

 

MATERIALS AND METHODS

 

BEHAVIOR MAINTAINED BY INTRAMUSCULAR INJECTIONS OF MORPHINE IN RHESUS MONKEYS

 

Two male rhesus monkeys (Macaca mulatta) lived in primate cages enclosed in sound-attenuating isolation chambers. A response key and two 25-watt bulbs (white and red) were mounted on a transparent Lucite wall in the front wall of each primate cage. Each monkey wore a leather collar which was attached by a chain to the front of the cage. Once a day during initial training, the monkey's chain was drawn tight and his forearm pulled out through a small opening in the front of the cage; the experimenter then injected the monkey intramuscularly with 6 mg/kg of morphine. After several weeks, the monkey would immediately extend his arm out through the opening in the cage when the experimenter tapped on the cage.

 

Daily experimental sessions were subsequently conducted Monday through Friday using a second-order schedule of intramuscular morphine injection. A white light went on at the start of each session and every tenth key press during a five-minute interval of time produced a red light for two seconds, but had no other scheduled consequences (the 10-response fixed-ratio component of the second-order schedule; FR 10). The first FR 10 component completed after five minutes turned off the white light and produced a red light which remained on for two minutes while the chamber door was opened; in the presence of the red light, the monkey extended his arm and received an intramuscular injection of 6 mg/kg morphine (the five-minute fixed-interval component of the second-order schedule; FI 5 min). Once responding was well maintained, the interval was increased to a final value of sixty minutes. Using the nomenclature of Kelleher (18,19), the final schedule can be designated as FI 60 min (FR 10:S); that is, a visual stimulus (S) was briefly presented at the completion of each FR 10 component, and the first FR component completed after the 60-minute interval ended also resulted in injection of morphine.

 

After several months under the FI 60 min (FR (10:S) schedule of intramuscular morphine injection, the frequency of experimental sessions was reduced to three times a week, Monday, Wednesday and Friday. When response rates were stable, the dose of morphine per injection was decreased gradually from 6 mg/kg to 0 mg/kg (saline substitution). Each dose was studied for at least twelve sessions. For further details of these experiments see Goldberg et al. (17).

 

BEHAVIOR MAINTAINED BY INTRAVENOUS INJECTIONS OF MORPHINE IN RHESUS MONKEYS

 

Four male rhesus monkeys (Macaca mulatta) prepared with permanent venous catheters were kept in individual home cages with free access to food and water. During experimental sessions, each monkey was restrained in a Lucite chair and placed in a sound-attenuating isolation chamber. A response key and two 25 watt red bulbs were mounted on a transparent Lucite wall in front of the monkey; a 15-watt white bulb also was mounted on top of the chair. The monkey's venous catheter was connected by polyvinyl tubing to a motor-driven syringe located outside the isolation chamber. Operation of the motor-driven syringe was controlled by automatic programming equipment; duration of each injection was 200 msec and volume of each injection was 0.18 ml.

 

Daily experimental sessions were conducted Monday to Friday. Initially a white light went on at the start of each session and every third or 10th key press (FR 3 or FR 10) changed the light from white to red for two seconds and produced an intravenous injection of 0.1 mg/kg or 0.2 mg/kg morphine. Each session ended after about fifty injections or one hour. Once responding was maintained, the schedule was changed from the simple FR schedule to a second-order FI schedule with FR components. A white light went on at the start of each session and every FR 3 or FR 10 component completed during a 60-minute interval of time (FI 60 min) changed the light from white to red for two seconds; during brief presentations of the red light, responding had no scheduled consequences. The first FR component completed after sixty minutes turned off the white light and produced the red light which remained on during a series of five to ten injections spaced ten seconds apart. After the injections, all lights were turned off, and the monkey remained in the chamber for ten to fifteen minutes before being returned to its home cage. The total dose of morphine delivered at the end of each session was initially 1.0 mg/kg for monkeys AT and AW and 5.0 mg/kg for monkeys AR and AX. With these monkeys the FR requirement was increased gradually to thirty responses.

 

After performance of monkeys AT and AW stabilized under the FI 60 min (FR 30:S) schedule, saline was substituted for the 1.0 mg/kg dose of morphine for three (AT) or seven (AW) sessions, and responding declined to low rates. A low dose of 0.3 mg/kg (monkey AW) or 0.5 mg/kg (monkey AT) of morphine was then studied for eight to twelve sessions. With monkey AT, the schedule then was changed so that no stimulus change occurred at completion of each FR component; the red light was presented only in association with morphine injections at the end of each session. This schedule could be designated FI 60 min (FR 30). After twelve sessions with the brief light changes omitted, an additional four sessions were conducted with the brief light changes restored. Subsequently, the dose of morphine injected at the end of each session was increased to 5.0 mg/kg for ten sessions and the effects of omitting the brief light changes were again studied. For further details of these experiments see Goldberg and Tang (15).

 

BEHAVIOR MAINTAINED BY INTRAVENOUS INJECTIONS OF MORPHINE IN SQUIRREL MONKEYS

 

Four male squirrel monkeys (Saimiri sciurea) prepared with,-permanent venous catheters were kept in individual home cages with free access to food and water. During experimental sessions, each monkey was restrained in a Lucite chair and placed in a sound-attenuating isolation chamber. A response key and two green and two amber 6-watt bulbs were mounted on a transparent Lucite wall in front of the monkey. The catheter connections, motor-driven syringe system, and injection volume and duration were the same as those used with the rhesus monkeys.

 

Before the present experiments, the squirrel monkeys had been studied under various schedules of intravenous cocaine injection; training techniques were generally similar to those described by Goldberg (5). At the start of the present experiments, morphine injections were substituted for cocaine injections and the schedule of drug injection was changed to a second-order FI schedule with FR components. Each monkey was tested once a day, Monday to Friday. Initially, a green light went on at the start of the session and every 30th key press (FR 30) during a five-minute interval of time changed the light from green to amber for two seconds; during brief presentations of the amber light, responding had no scheduled consequences. The first FR 30 component completed after five minutes turned off the green light and produced fifteen consecutive injections of 0.1 mg/kg morphine; each 200 msec injection occurred at the onset of a two-second amber light. These light-injection pairings were spaced ten seconds apart so that over a period of 140 seconds the monkey received a total dose of 1.5 mg/kg morphine. After the injections, all lights were turned off, and the monkey remained in the chamber for five to ten minutes before being returned to its home cage. Once responding was well maintained, the interval was increased to a final value of sixty minutes.

After performance stabilized under the FI 60 min (FR 30:S) schedule, the total dose of morphine injected at the end of each session was increased and decreased over a range of 0 (saline) to 6.0 mg/kg. Each dose was studied for ten to sixteen sessions. Subsequently, the morphine dose was returned to 1.5 mg/kg, and after ten sessions the schedule was changed so that no stimulus change occurred at completion of each FR component; two-second amber light presentations occurred only in association with morphine injections at the end of each session. After six sessions (monkey S-405) or fifteen sessions (monkey S-369) with the brief light changes omitted, an additional four to six sessions were conducted with the brief light changes restored. For further details of these experiments see Goldberg and Tang (15).

 

Analysis of results

 

Mean rates of responding under the second-order schedules, with and without brief stimulus changes, were computed for each session by dividing total responses in the presence of the green or white light by total time the green or white light was present; responses during the two-second stimulus change to an amber or red light and total time the amber or red light was present were not included in computations.

 

Drugs 

 

Morphine sulfate was dissolved in saline (0.9% NaCl). All doses are expressed as the salt.

 

RESULTS

 

BEHAVIOR MAINTAINED BY INTRAMUSCULAR INJECTIONS OF MORPHINE IN RHESUS MONKEYS

Responding increased and then stabilized at a mean rate of approximately 0.10 response/second with monkey M-7 and 0.23 response/second with monkey M-9 under the final FI 60 min (FR 10:S) schedule of intramuscular morphine injection. A cumulative-response record, representative of final performance of monkey M-7 is shown in Figure 1. The briefly presented lights controlled patterns of responding characteristic of FR schedules (20); a pause in responding after each two-second red light was usually followed by an abrupt change to rapid responding until the light was produced again. Pauses in responding were longest at the start of the 60-minute interval and became shorter as time elapsed in the interval. An overall pattern of positively-accelerated responding is characteristic of FI schedules (20). Monkey M-9 responded more rapidly than M-7, but there was occasional deterioration in the FR patterns of responding, and the positively-accelerated FI pattern of responding was less evident.

Fig. 1. Representative performance of rhesus monkey M-7 under the zecond-order schedule of intramuscular morphine injection. Ordinates, cumulative number of key-pressing responses; abscissae, time. Short downward deflections on the cumulative record indicate two-second presentations of a red light after every 10th response; during the two-second light presentations the recorder did not operate and responses had no scheduled consequences. The session ended with an intramuscular injection of 3 mg/kg morphine sulfate accompanied by a red light, which is indicated by the recording pen resetting to the bottom of the cumulative record. From Goldberg (14) with permission.

 

Mean response rates maintained by different doses of morphine or by saline are shown in Figure 2. Responding was well maintained by doses of morphine from 1.5 mg/kg to 6 mg/kg. Over this range there was no clear relationship between morphine dose and rates or patterns of responding. Before experimental sessions on Monday, when three days had elapsed since the last morphine injection, both monkeys often showed some signs of hyperactivity and hyperirritability, but a clear withdrawal syndrome was never observed. When the dose of morphine was below 1.5 mg/kg, responding decreased. When saline injections were substituted for injections of these low doses of morphine, responding of both monkeys further decreased and FR patterns of responding were lost. When 3 mg/kg injections of morphine were reinstated after twelve saline-substitution sessions, the previous rates and patterns of responding were restored within six sessions.

Fig. 2. Mean response rates of rhesus monkeys M-7 and M-9 under the second-order schedule of intramuscular morphine injection as a function of the dose of morphine ()) sulfate injected at the end of the session. Ordinates: mean response rate; abscissae: dose. Each point represents the mean and the brackets the standard error from the last six sessions of saline (NaC1) substitution are shown by the open symbols. From Goldberg e permission.

 

BEHAVIOR MAINTAINED BY INTRAVENOUS INJECTIONS OF MORPHINE IN RHESUS MONKEYS

 

Rates and patterns of responding appeared stable after three to four weeks under the FI 60 min (FR 30:S) schedule of intravenous morphine injection. Cumulative-response records, representative of final performance of three monkeys at a dose of 5.0 mg/kg of morphine, are shown in Figure 3. Ain the previous studies with intramuscular injections of morphine, characteristic FR patterns of responding were controlled by the briefly presented lights; a pause in responding after each two-second red light was followed by an abrupt change to rapid responding until the light was produced again. Also, the pauses in responding generally were longest at the start of the 60-minute interval and become shorter as time elapsed in the interval, a pattern of responding characteristic of FI schedules.

 

Although similar patterns of responding were maintained under second-order schedules by intramuscular and intravenous injections of morphine, much higher rates of responding were maintained by intravenous injections of morphine. At a dose of 5.0 mg/kg of morphine, given intravenously, mean response rates ranged from 0.4 to 0.7 response/second in the different monkeys. As shown in Figure 4, mean response rates as high as 1.0 response/second were maintained in two monkeys at 1.0 mg/kg morphine. When the dose of morphine was below 1.0 mg/kg, rates of responding decreased. When injections of saline were substituted for injections of morphine, rates of responding decreased further and FR patterns of responding were lost within five sessions. Previous rates of responding and characteristic FR patterns of responding could be restored within three sessions by reinstating injections of morphine. Although some signs of hyperactivity and hyperirritability were occasionally noted during saline substitution, a clear withdrawal syndrome was never observed.

 

 

Fig. 3. Representative performances of rhesus monkeys AR, AT and AX under a second-order schedule of intravenous morphine injection. Ordinates, cumulative number of key-pressing responses; abscissae, time. Short downward deflections on the cumulative records indicate two-second presentations of a red light after every 30th response; during the two-second light presentations the recorder continued to operate but responses had no scheduled consequences. The recording pen reset to the bottom of the cumulative record whenever 1000 responses cumulated and when morphine was injected at the end of the session. The red light remained on at the end of the session until five or ten injections of morphine, spaced ten seconds apart, were delivered (a total dose of 5 mg/kg morphine sulfate).

Fig. 4. Mean response rates of rhesus monkeys AW and AT under the second-order schedule of intravenous morphine injections as a function of the total dose of morphine sulfate injected at the end of the session. Ordinates: mean response rate; abscissae: dose. Each bar represents the mean and the brackets the range from the last five sessions at each dose of morphine and of the last two sessions of saline (0 mg/kg) substitution. Modified from Goldberg and Tang (15).

 

Cumulative-response records representative of the effects of omitting the brief light changes with rhesus monkey AT are shown in Figure 5. A morphine dose as high as 5.0 mg/kg maintained mean rates of responding exceeding 1.0 response/second during the last thirty minutes of each session with monkey AT.

Fig. 5. Performance of rhesus monkey AT before, during and after omitting the brief light presentations under the second-order schedule of intravenous morphine injection. Recordings as in figure 3. Each record shows a complete session which ended with injection of a total w dose of 5 mg/kg morphine sulfate. 2 The records shown are from: the 2 last session (A) before omitting Fci the brief presentations of the red light; the 13th session (B) o with no scheduled stimulus change at completion of each fixed-ratio component (the red light occurred only in association with morphine injections at the end of the session); and the second session (C) with the brief light presentations reinstated. Modified from Goldberg and Tang (15).

 

When the brief changes were omitted, but morphine was still injected at the end of each session, rates of responding decreased markedly in four to eight sessions and FR patterns of responding were lost. When the brief light change again occurred at completion of each FR component, the higher rates and characteristic FR patterns of responding were restored. When responding was maintained by 0.5 mg/kg of morphine with monkey AT, rates of responding fell to such low levels after six sessions with the brief light changes omitted, that during some sessions no responses occurred during the 60-minute interval (sufficient responding occurred several minutes after the end of the interval to produce injection of morphine and end the session). In order to restore responding when the brief light changes were reinstated at the 0.5 mg/kg dose of morphine, the FR response requirement was reduced to one and three during the first two sessions with the brief light changes reinstated. Responding increased during these sessions during subsequent sessions, the FR response requirement was returned to thirty and higher rates of responding were again maintained by the 0.5 mg/kg dose of morphine.

 

BEHAVIOR MAINTAINED BY INTRAVENOUS INJECTIONS OF MORPHINE IN SQUIRREL MONKEYS

 

High rates and characteristic FR patterns of responding also were maintained under the FI 60 min (FR 30:S) schedule of intravenous morphine injection in squirrel monkeys. Figure 6 shows cumulative-response records representative of final performances of four squirrel monkeys at a dose of 1.5 mg/kg of morphine. High rates of responding were maintained throughout most of the session even though morphine was injected only at the end of the session; characteristic FR patterns of responding were controlled by the briefly presented lights.

 

Fig. 6. Representative performances of four squirrel monkeys under a second-order schedule of intravenous morphine injection. Ordinates, cumulative number of key-pressing responses; abscissae, time. Short downward deflections on the cumulative records indicate

two-second presentations of an amber light after every 30th response; during the two-second light presentations the recorder LiT, continued to operate but responses o had no scheduled consequences. The session ended with fifteen consecutive injections of 0.1 mg/kg morphine, spaced ten seconds apart (a total dose of 1.5 mg/kg morphine sulfate); each injection occurred at the onset of a two-second amber light. The downward deflection on the horizontal event line in each record indicates the period of morphine injection. The recording pen reset to the bottom of the cumulative record whenever 1100 responses cumulated and when the session ended.

 

For each of the squirrel monkeys, a mean rate of responding exceeding 1.0 response/second was maintained at a dose of 0.75 or 1.5 mg/kg of morphine. At lower doses of morphine, ranging from 0.3 to 0.75 mg/kg in different animals, mean rates of responding were considerably lower but were still above saline-substitution levels. In squirrel monkey S-369, a dose of morphine as high as 6.0 mg/kg continued to maintain responding, although at somewhat reduced rates. When saline injection was substituted for injection of 1.5 mg/kg morphine, mean response rates decreased markedly and FR patterns of responding were lost within seven sessions, but rates and patterns of responding could be restored within a few sessions by reinstating injections of morphine. Withdrawal signs were never observed in squirrel monkeys when saline was substituted for morphine.

 

Representative cumulative-response records of monkeys S-369 and S-405 from sessions before, during and after six to fifteen consecutive sessions with the brief light changes omitted are shown in Figure 7. When the brief light changes were omitted, the resulting FI 60 min (FR 30) schedule closely resembled a simple 60-minute FI schedule of morphine injection. Although the same dose of 1.5 mg/kg morphine continued to be injected at the end of each session, omitting the brief light changes resulted in a marked decrease in mean response rates within two sessions. There was also a loss of the FR patterns of responding and the appearance of more pronounced progressively-accelerated patterns of responding. The final rates and patterns of responding under the FI 60 min (FR 30)

schedule were similar to what would be expected under a long FI schedule (20). When a brief light change again occurred at completion of each FR component, the high mean rates and FR patterns of responding were immediately restored.

Fig. 7. Performances of squirrel monkeys S-369 and S-405 before, during and after omitting the brief light presentations under the second-order schedule of intravenous morphine injection. Recordings as in figure 6. Each record shows a complete session which ended with injection of a total dose of 1.5 mg/kg morphine sulfate. The records shown are from: the last session (A) before omitting the brief presentations of the amber light; the first session (B) and the last session (C) with no scheduled stimulus change at completion of each fixed-ration component (the amber light occurred only in association with morphine injections at the end of the session); and the first session (D) with the brief light presentations reinstated. Modified from Goldberg and Tang (15).

 

DISCUSSION

 

Schedules of drug injection which restrict drug injection to the end of each session provide a means of studying the reinforcing effects of large doses of drugs such as morphine, which have pronounced suppressant effects on behavior. Pretreatment with doses of morphine ranging from 0.3 to 5.6 mg/kg has been shown to markedly suppress responding maintained under FR schedules of intravenous codeine or cocaine injection in rhesus monkeys (2,21) and under FR or FI schedules of food presentation in rhesus monkeys (22,23) and squirrel monkeys (24,25). Under simple FR schedules of drug injection, where intravenous injection of morphine follows completion of each FR component and repeated injections can be obtained throughout and experimental session, doses of morphine above 0.1 mg/kg/injection have usually failed to maintain rates of responding above those maintained by saline injections (2,16). Under the present second-order schedules, however, morphine was injected only at the end of each session and doses of morphine ranging from 0.75 to 6 mg/kg functioned effectively as reinforcers and maintained rates of responding much higher than those maintained by saline injections. Schedules of drug injection in which the drug is injected only at the end of each experimental session allow one to study the reinforcing effects of a drug independently of its other effects on behavior.

 

A critical factor in using a drug to maintain behavior may be its route of administration. Intravenous injections seem particularly effective. In experimental animals prepared with permanent venous catheters, drug can be injected immediately and without trauma after some response of the subject, and the effects of the drug usually occurs quickly. When responding was maintained by intravenous injections of morphine in the present experiments, final rates of responding were much higher than when responding was maintained under a similar condition by intramuscular injection of morphine. Although higher mean rates of responding were maintained by intravenous injections of morphine, intramuscular injections of morphine were able to consistently maintain responding over long periods of time and brief light changes associated with the injections modulated the control of the behavior.

 

Morphine is often presumed to be particularly effective as a reinforcer when it is given intravenously to a subject tolerant to large doses of morphine and currently in withdrawal. In the present experiments, however, intravenous injection of morphine maintained mean rates of responding exceeding one response per second even though the monkeys showed no clear signs of physiological dependence development. Other investigators have also demonstrated that intravenous injection of moprhine can maintain responding in non-dependent rhesus monkeys independently of its efficacy in terminating the withdrawal syndrome (16,22,26,27).

 

In clinical situations of drug abuse and relapse to drug use, environmental cues and stimuli associated with the drug experience are believed to be contributory factors to such behaviors (10-12). The present second-order schedules of morphine injection provide a convenient experimental tool for determining the conditions under which environmental stimuli associated with extended sequences of drug-seeking behavior come to play an important role in the maintenance of and relapse to drug-seeking behavior. Under the present second-order schedules, drug was injected only at the end of the session and behavior during the session was controlled by intermittent changes in the color of the stimulus lights. The brief two-second light changes controlled local patterns of responding characteristic of FR schedules and appeared necessary for the maintenance of high mean rates of responding by intravenous injections of morphine. When the brief light changes were omitted during the session, but morphine continued to be injected intravenously at the end of the session mean rates of responding decreased markedly in both rhesus and squirrel monkeys. Further experiments are being conducted to determine the conditions necessary for brief environmental stimuli to facilitate drug-seeking behavior under second-order schedules. In the present experiments, for example, the light change always occurred during the series of morphine injections at the end of each session. Experiments are now being conducted to determine whether or not high rates and FR patterns of responding can be maintained during the session if the brief light change follows completion of all but the final FR component and is never directly associated with morphine injection. This will allow assessment of the importance of temporally pairing environmental stimuli with the actual injection of drug.

 

SUMMARY

 

Under second-order schedules of intravenous morphine injections, high rates of responding were maintained when morphine was injected only at the end of each experimental session. During each session every 30th response produced a two-second light change. These briefly presented stimuli exerted important control over behavior as evidenced by a pause in responding after each light change followed by rapid responding until the light changed again, and by a marked decrement in rates of responding when the brief light changes were omitted during the session. Responding could be consistently maintained under a similar second-order schedule by intramuscular, rather than intravenous, injections of morphine, although rates of responding were lower.

 

REFERENCES

 

1. Deneau, G.A., Yanagita, T. and Seevers, M.H.: Self-administration of psychoactive substances by the monkey: A measure of psychological dependence. Psychopharmacologia 16:30-48 (1969).

2. Hoffmeister, F. and Schlichting, U.U.: Reinforcing properties of some opiates and opioids in rhesus monkeys with histories of cocaine and codeine self-administration. Psychopharmacologia 23:55-74 (1972).

3. Wilson, M.C., Hitomi, M. and Schuster, C.R.: Psychomotor stimulant self-administration as a function of dosage per injection in the rhesus monkey. Psychopharmacologia 22: 271-281 (1971).

4. Downs, D.A. and Woods, J.H.: Codeine- and cocaine-reinforced responding in rhesus monkeys: Effects of dose on response rates under a fixed-ratio schedule. J. Pharm. exp. Ther. 191:179-188 (1974).

5. Goldberg, S.R.: Comparable behavior maintained under fixed-ratio and second-order schedules of food presentation, cocaine injection or d-amphetamine injection in the squirrel monkey. J. Pharm. exp. Ther. 186:18-30 (1973).

6. Pickens, R. and Thompson, T.: Cocaine-reinforced behavior in rats. J. Pharm. exp. Ther. 161:122-129 (1968).

7. Goldberg, S.R.: Sequences of rapid responding maintained by cocaine self-injection in squirrel monkeys. Pharmacologist 13:503 (1971).

8. Goldberg, S.R.: Control of behavior by stimuli associated with drug injections, Psychic Dependence. Edited by Goldberg, L. and Hoffmeister, F., Springer-Verlag, Berlin, P. 106-109 (1973).

9. Goldberg, S.R., Kelleher, R.T. and Morse, W.H.: Second-order schedules of drug injection. Fed. Proc. 34:17711776 (1975).

10. Wikler, A.: Conditioning factors in opiate addiction and relapse, Narcotics. Edited by Wilner, D.M. and Kassenbaum, C.G., McGraw-Hill, New York, p. 85-100 (1965).

11. Vaillant, G.E.: The natural history of urban narcotic drug addiction - some determinants, Scientific Basis of Drug Dependence. Edited by Steinberg, H., J & A Church-hill, Ltd., London, p. 341-361 (1969).

12. Goldberg, S.R.: Relapse to opioid dependence: The role of conditioning, Drug Dependence (Advances in Mental Science, Vol. 2. Edited by Harris, R.T., Mclsaac, W.M. and Schuster, C.R., Jr., The University of Texas Press, Austin and London, p. 170-197 (1970).

13. Goldberg, S.R.: Stimuli associated with drug injections as events that control behavior. Pharmacol. Rev. 27,:325340 (1975).

14. Goldberg, S.R.: Second-order schedules of morphine or cocaine injection. Pharmacologist 18:197 (1976).

15. Goldberg, S.R. and Tang, A.H.: Behavior maintained under second-order schedules of intravenous morphine injection in squirrel and rhesus monkeys. Psychopharmacology, in press (1976).

16. Hoffmeister, F. and Goldberg, S.R.: A comparison of chlorpromazine, imipramine, morphine and d-amphetamine self-administration in cocaine-dependent rhesus monkeys. J. Pharm. exp. Ther. 187:8-14 (1973).

17. Goldberg, S.R., Morse, W.H. and Goldberg, D.M.: Behavior maintained under a second-order schedule by intramuscular injection of morphine or cocaine in rhesus monkeys. J. Pharm. exp. Ther. in press (1976).

18. Kelleher, R.T.: Chaining and conditioned reinforcement, Operant Behavior: Areas of Research and Application. Edited by Honig, W.K., Appleton-Century-Crofts, New York, p. 160-212 (1966).

19. Kelleher, R.T.: Conditioned reinforcement in second-order schedules. J. Exp. Anal. Behay. 9:475-485 (1966).

20. Ferster, C.B. and Skinner, B.F.: Schedules of Reinforcement. Appleton-Century-Crofts, New York (1957).

21. Wilson, M.C. and Schuster, C.R.: The effects of stimulants and depressants on cocaine self-administration behavior in the rhesus monkey. Psychopharmacologia 31: 291-304 (1973).

22. Woods, J.H. and Schuster, C.R.: Opiates as reinforcing stimuli, Stimulus Properties of Drugs. Edited by Thompson, T. and Pickens, R., Appleton-Century-Crofts, New York, p. 163-175 (1971).

23. Downs, D.A. and Woods, J.H.: Morphine, pentazocine and naloxone effects on responding under a multiple schedule of reinforcement in rhesus monkeys and pigeons. J. Pharm. exp. Ther. 196:298-306 (1976).

24. McKearney, J.W.: Effects of d-amphetamine, morphine and chlorpromazine on responding under fixed-interval schedules of food presentation or electric shock presentation. J. Pharm. exp. Ther. 190:141-153 (1974).

25. Goldberg, S.R., Morse, W.H. and Goldberg, D.M.: Some behavioral effects of morphine, naloxone and nalorphine in the squirrel monkey and the pigeon. J. Pharm. exp. Ther. 196:625-636 (1976).

26. Schuster, C.R.: Variables affecting the self-administration of drugs by rhesus monkeys, Use of Nonhuman Primates in Drug Evaluation. Edited by Vagtborg, H., The University of Texas Press, Austin, p. 283-299 (1968).

27. Woods, J.H. and Schuster, C.R.: Reinforcement properties of morphine, cocaine and SPA as a function of unit dose. Int. J. Addict. 3:231-237 (1968).