An Addiction Science Network Resource

Reprinted from A.G. Phillips and H.C. Fibiger (1987), Anatomical and neurochemical substrates of drug reward determined by the conditioned place preference technique. In M.A. Bozarth (Ed.), Methods of assessing the reinforcing properties of abused drugs (pp. 275-290). New York: Springer-Verlag.

Brain Reward System
Back to book TOC
(Search book contents) (Search entire ASNet)


Chapter 15

Anatomical and Neurochemical Substrates of Drug Reward
Determined by the Conditioned Place Preference Technique

Anthony G. Phillips and Hans C. Fibiger

Department of Psychology and
Division of Neurological Sciences
Department of Psychiatry
University of British Columbia
Vancouver, British Columbia, Canada V6T 1W5

The conditioned place preference paradigm appears well suited to the analysis of neural pathways involved in drug reward. Current literature on this topic is reviewed with respect to three issues: (1) pharmacological blockade of drug reward, (2) mapping of effective brain loci where microinjection of drugs can be used to condition a place preference, and (3) attenuation of place preference conditioning obtained with systemic or intracerebral administration of drugs and by selective lesions of central dopaminergic pathways. Collectively, these data suggest that a mesotelencephalic dopamine system may serve as one important substrate for the rewarding effects of both psychostimulant and opiate drugs.



Animal models of drug reinforcement and drug dependence have employed both the voluntary and involuntary administration of drugs. For those interested in behavior related to the initial phases of drug acquisition and dependency, the intravenous self-administration procedure developed by Weeks (1962) has been the method of choice. Studies using this technique have demonstrated positive reinforcing effects of a wide variety of drugs including psychostimulants and opiates in many mammalian species (Schuster & Thompson, 1969). Despite the significant advances permitted by this paradigm, it is not without its drawbacks. For example, the implantation of intravenous catheters is laborious and the patency of the catheters can be problematic. In addition, the interpretation of data relating to neural substrates of drug reinforcement can sometimes be difficult as the animals are often required to perform operant responses after pharmacological treatments or brain lesions that may interfere with operant behavior. The rate of operant responding under simple schedules of drug delivery also can be an unreliable measure of the relative efficacy of reinforcement (Johanson & Aigner, 1981). Therefore the development of other procedures for the study of drug-induced reinforcement is desirable.

One alternative procedure, suitable for studying the relation between the rewarding stimulus properties of drugs and environmental stimuli, is the conditioned place preference paradigm (Rossi & Reid, 1976). This procedure is derived from the finding that the association of distinctive environmental stimuli with a primary reward such as food or a drug injection will result in an acquired preference for those specific environmental stimuli in the absence of the primary reward. Some of the advantages of the place preference paradigm include replacement of multiple drug injections by a single systemic or intracerebral injection at the beginning of each daily session. Furthermore, rapid conditioning in one to four trials eliminates the need for extensive behavioral testing. The purpose of the paper is to review recent research utilizing conditioned place preference as a procedure for the identification of anatomical and neurochemical substrates of drug reward.

It is now well established that many of the drugs used to maintain intravenous self-administration also can be used to establish conditioned place preference (Phillips, Spyraki, & Fibiger, 1982). Opiates (Rossi & Reid, 1976) and psychostimulants (Reicher & Holman, 1977) have been studied most frequently, and recent attempts to identify neurochemical correlates of drug reward have used these drugs in conjunction with specific opiate or catecholamine receptor antagonists.

Several laboratories have reported that naloxone disrupts conditioned place preference produced either by systemic administration of morphine (Sherman, Pickman, Rice, Liebeskind, & Holman, 1980) or by heroin (Bozarth & Wise, 1981b), thereby indicating an opiate receptor mediation of the rewarding effects of opiate drugs. This interpretation presumes that naloxone itself does not simply cancel the rewarding effect of opiates by having inherent aversive properties. One body of data indicates that peripheral administration of naloxone (1 to 3 mg/kg) via the intraperitoneal (i.p.) route does not produce a conditioned place aversion (Bozarth & Wise, 1981b; Phillips & LePiane, 1980). However, controversy surrounds this issue as Mucha and colleagues have demonstrated significant conditioned place aversion with both intravenous (0.1 to 4.5 mg/kg) and subcutaneous (0.1 to 0.5 mg/kg) administration of naloxone (Mucha & Iversen, 1984; Mucha, van der Kooy, O’Shaughnessy, & Bucenieks, 1982). In the latter study, a smaller but nevertheless significant place aversion was obtained with intraperitoneal administration of naloxone. The discrepancy between these experiments would appear to reflect both route of drug administration and, to a lesser degree, possible floor effects associated with pairing naloxone injections with confinement to the least preferred compartment of the shuttlebox. On the basis of these studies, it appears that intraperitoneal injections of naloxone can disrupt conditioned place preference produced by opiates by antagonizing the direct action of these drugs on the subclass of opiate receptor that mediates their rewarding effects on behavior. It should also be recognized that the more sensitive procedures employed by Mucha and his coworkers point to the existence of an endogenous opioid peptide system, the modulation of which can result either in reward or in punishment. The identification of such a system has been a high priority in this field of research, and the relevant experiments using intracerebral injections and brain lesions are reviewed below.

In the context of pharmacological antagonism of opiate reward, there have been several independent reports of significant attenuation of opiate-induced place preference by neuroleptic drugs. Morphine place preference has been blocked by pretreatment with the catecholamine synthesis inhibitor alpha-methyl-p-tyrosine or by the dopamine (DA) receptor antagonist haloperidol (0.25 to 1.0 mg/kg) (Schwartz & Marchok, 1974). Similar results have been obtained with heroin induced place preference. In these studies pimozide (Bozarth & Wise, 1981b, 1982) and haloperidol (see Figure 1B) prevented conditioning of a place preference by heroin. Treatment with either neuroleptic alone did not produce a conditioned place aversion. Both pimozide and haloperidol have relatively selective effects on dopamine receptors. Therefore these data clearly implicate a dopaminergic mechanism in opiate reward processes.

Effect of haloperidol on amphetamine and heroin CPP
Figure 1: A: Effect of haloperidol (filled circles, 0.15 mg/kg; open triangles, 1.0 mg/kg) or vehicle injections (open circles) on conditioned place preference with amphetamine (1.5 mg/kg) reward. B: Effect of haloperidol (filled circles, 0.15 mg/kg) or vehicle (open circles) on conditioned place preference with heroin (2 mg/kg) reward. Lack of conditioned aversion to haloperidol (open triangles, 0.2 mg/kg) alone, paired with the initially preferred compartment also is illustrated. Adapted with permission from Spyraki, Fibiger, and Phillips, 1982a, 1983.

As indicated previously, psychostimulant drugs have been used successfully to induce conditioned place preference. Included in this category are d-amphetamine (Sherman, Roberts, Roskam, & Holman, 1980), apomorphine (Spyraki, Fibiger, & Phillips, 1982a), cocaine (Spyraki, Fibiger, & Phillips, 1982b), methylphenidate, and nomifensine (Martin-Iverson, Ortmann, & Fibiger, 1985). Interestingly, d-amphetamine induced place preference is attenuated by dopamine receptor blockade (see Figure 1A) but this treatment fails to block place preference obtained with cocaine (see Figure 2), methylphenidate, or nomifensine (Martin-Iverson et al., 1985). These data are important on two counts. First, they indicate that the neural substrates of the rewarding effect of d-amphetamine are to some extent distinct from those mediating the reinforcing effects of cocaine, methylphenidate, and nomifensine. Furthermore, these data point to possible differences between the intravenous self-administration and conditioned place preference paradigms in their sensitivity to different aspects of drug reward because DA receptor antagonists have similar effects on self-administration maintained by all of these stimulants.

Notwithstanding the useful data provided by the use of selective antagonists, it is clear that more direct and localized manipulation of brain activity is required for the successful identification of the neural substrates of drug reward. Some success has been obtained through mapping studies involving intracerebral micro-injection of opiates and psychostimulants. Selective lesions of central catecholamine pathways also have proved useful in this endeavor.

Effects of haloperidol and of pimozide on cocaine CPP
Figure 2: Effects of (A) pimozide (0.5 mg/kg), (B) pimozide (1.0 mg/kg), and (C) haloperidol (1.0 mg/kg) pretreatment on place preference conditioning produced by cocaine hydrochloride (5.0 mg/kg). Horizontal line indicates the no preference value (i.e., half of the 900-second session). Adapted with permission from Spyraki, Fibiger, and Phillips, 1982b.

Rewarding Effects of Localized
Intracerebral Microinjections of Drugs

A major advantage of intracerebral microinjection procedures for identifying neural substrates of drug reward is derived from the ability to place various compounds into specific subcortical loci in a relatively precise manner. When this procedure is coupled with the abundant information on the anatomical location of various receptor types, it becomes feasible to relate receptors for specific synaptic transmitters to reward produced by different drugs. Although intracerebral microinjection can also be used in conjunction with lever pressing to confirm the reinforcing properties of drugs (cf. Bozarth, 1983; Olds, 1979), the conditioned place preference paradigm has several features which recommend its use.

It is now well established that rewarding effects can be demonstrated by the intracerebral technique with as few as one to three individual pairings of drug and environmental stimuli (van der Kooy, Mucha, O’Shaughnessy, & Bucenieks, 1982). Therefore few intracerebral injections need to be made, thus minimizing nonspecific damage to brain tissue at the tip of the injection needle. Perhaps the greatest advantage comes from conducting the tests for conditioned reinforcement when the animals are in a drug free state. This fact along with the minimal response demands of the procedure insures that measures of reward are not confounded by changes in activity or performance.

Opiates and Opioid Peptides

The successful conditioning of a place preference with intraventricular injections of d-ala2-methionine enkephalin (D-Ala) (Katz & Gormezano, 1979; Stapleton, Lind, Merriman, Bozarth, & Reid, 1979), a long-acting enkephalin analogue (Pert, Pert, Chang, & Fong, 1976), provided the rationale for similar experiments with more localized microinjection of opiate drugs. Several factors influenced our initial selection of the ventral tegmental area (VTA) as a possible site at which the pharmacological action of opiates could lead to the reinforcement of behavior. A primary consideration was the facilitation of intracranial self-stimulation from electrodes implanted into the lateral hypothalamus by intracerebral injections of morphine at the diencephalic-mesencephalic border (Broekkamp, van den Bogaard, Heynen, Rops, Cools, & van Rossum, 1976). The most effective sites were subsequently localized to the VTA, and D-Ala was shown to mimic the facilitory effects of morphine on electrical self-stimulation (Broekkamp, Phillips, & Cools, 1979a). Locomotor activity, a behavior often linked to activation of dopaminergic neurons in the A nucleus of the VTA, also was observed after local application of morphine (Joyce & Iversen, 1979) or D-Ala (Broekkamp, Phillips, & Cools, 1979b) to this region of the brain. Other factors included the identification of enkephalinergic nerve terminals in this region of the brain (Johansson, Hokfelt, Elde, Schulzberg, & Terenius, 1978) and the excitation of single unit activity following microinjections of morphine into the region (Finnerty & Chan, 1979).

Successful conditioning of a place preference was established after three daily microinjections of morphine (0.2 to 1.0 mg/0.5 ml) injected bilaterally into the VTA were paired with the less preferred compartment of a shuttlebox (see Figure 3; Phillips & LePiane, 1980). Independent confirmation of this effect was provided by Bozarth and Wise (1982) using unilateral injections of morphine (0.25 mg/0.5 ml). Subsequently, a similar effect was reported with unilateral injections of D-Ala (0.1 to 0.25 mg/0.5 ml) into the VTA (Phillips & LePiane, 1982). Naloxone (2 mg/kg, i.p.) successfully antagonized the rewarding effect of a unilateral injection of 0.1 mg D-Ala. The failure to obtain significant place preference with injection sites located dorsal (Phillips & LePiane, 1980, 1982) or caudal (Bozarth & Wise, 1982) to the VTA suggests that important elements of opiate reward are located in the VTA.

These place preference experiments with intracerebral injections of opiate drugs and enkephalin analogues have been complemented by a study using an enkephalinase inhibitor which protects endogenous opioid peptides from enzymatic degradation (Glimcher, Giovino, Margolin, & Hoebel, 1984). Microinjection of the enkephalinase inhibitor thiorphan (Roques, Fournie-Zaluski, Soroca, Lecomte, Malfroy, Llorens, & Schwartz, 1980) was used to elevate endogenous levels of enkephalin in the VTA by inhibiting dipeptidyl carboxypeptidase, the enkephalinase which normally cleaves enkephalin at the gly -phe -amide bond. Successful conditioning of a place preference following such injections suggested that the release of endogenous enkephalin in the VTA has primary rewarding effects.

It must be emphasized that the localization of a substrate for opiate reward in the VTA does not preclude other critical sites elsewhere in the brain. In fact, several positive sites have been identified using bilateral injections of morphine (5 g per side). Significant place preference was conditioned in rats with cannulae placements into the lateral hypothalamus, nucleus accumbens, and periaqueductal gray; no effect was obtained following injections into the central nucleus of the amygdala, caudate-putamen, or nucleus ambiguus (van der Kooy, Mucha, O’Shaughnessy, & Bucenieks, 1982). The most pronounced behavioral effects of morphine observed during these conditioning trials were increased locomotion and excessive rearing and grooming. It is noteworthy that hyperactivity also accompanied morphine injections into the VTA (Joyce & Iversen, 1979). An important control in the study by van der Kooy et al. (1982) was the use of the active (-) and inactive (+) isomers of morphine (Jacquet, Klee, Rice, & Minamikawa, 1977). The establishment of place preference only with (-) morphine provides further evidence for the involvement of specific brain opiate receptors in the rewarding effect of morphine. To date, the four brain loci noted above have been identified as active sites at which opiates produce primary rewarding effects. Given the limited number of sites tested, it is obviously premature to conclude that these positive loci constitute the neural substrates of opiate reward. As noted by van der Kooy et al. (1982), it is possible that greater diffusion of a drug within a given nucleus or that refinement of the place conditioning procedure may be used to condition significant place preference at loci that so far have failed to yield positive results. It also should be stated that identification of positive sites need not infer an action of morphine in that specific area of the brain. Intraventricular injections of morphine at a dose of 10 mg into the lateral ventricle can be used to obtain conditioned place preference, whereas a dose of 0.5 mg is below threshold (cf. van der Kooy et al., 1982). Consequently, care must be exercised to avoid penetration of the ventricles during stereotaxic implantation into regions such as the nucleus accumbens and lateral hypothalamus. Inadvertent damage to the ventricles in conjunction with relatively high doses of morphine (i.e., 10 mg) would preclude localization of effects to specific brain nuclei. Such factors may account for some of the positive results reported by van der Kooy et al. (1982).

CPP from intracerebral morphine microinjections
Figure 3: Preference scores expressed as time spent on one side of the test chamber before and after intracerebral microinjection of morphine or saline. The preconditioning scores indicate that the "conditioned" side was the less preferred prior to drug treatment. The amount of time spent on the preferred side may be obtained by subtracting a given group score from the session length of 900 seconds. Postconditioning scores represent the amount of time spent on the same side after association between this environment and microinjection of morphine or saline. Groups: (filled triangles) 0.2 mg morphine, VTA; (filled squares) 1.0 mg morphine, VTA; (open circles) 1.0 mg morphine, dorsal placements; (open triangles) 0.9% saline, VTA. Adapted with permission from Phillips & LePiane, 1980.


The conditioned place preference paradigm is now used routinely in conjunction with peripheral routes of drug administration to confirm the rewarding effects of both opiates and psychostimulants. In contrast, to date there has been only one report of conditioned place preference with intracerebral injections of a stimulant drug. A significant preference was produced by d-amphetamine sulphate (10 mg/0.5 ml) for the drug environment after six daily injections of the drug unilaterally into the nucleus accumbens (Carr & White, 1983). These cannulae were implanted on an angle to avoid penetration of the ventricles. Comparable injections into the head of the striatum did not have a significant effect. The results of this study along with the earlier report of conditioned place preference after morphine injections into the nucleus accumbens suggest that a common mechanism in this region of the brain may mediate the rewarding effects of both psychostimulants and opiates. As already alluded to above in reviewing the effects of neuroleptics on conditioned place preference, the common denominator may well involve activation of certain mesotelencephalic dopaminergic neurons.


One of the most significant findings of all of the mapping studies conducted to date with the conditioned place preference procedure is the consistent overlap between reward placements and the trajectory of the dopaminergic neurons arising from the VTA. Positive results have been obtained with all compounds known to influence the activity of these neurons, be they opiate drugs, opioid peptides, or stimulants. Given this pattern of results, it follows that other endogenous neuropeptides with direct excitatory effects on dopaminergic neurons in the VTA may have rewarding effects on behavior. Neurotensin, a tridecapeptide, has been shown to have excitatory effects on the neurophysiological activity of DA neurons (Andrade & Aghajanian, 1982), and neurotensin-like immunoreactivity has been seen in the VTA (Uhl, Goodman, & Snyder, 1979). These characteristics of neurotensin led to the hypothesis that it may serve as a novel "reward" peptide. This conjecture has received initial support from the demonstration that neurotensin can be used to induce a place preference when injected into the VTA (Glimcher, Margolin, Giovino, & Hoebel, 1984).

Brain sites where intracerebral microinjections produce a CPP
Figure 4: Sagittal section of the rat brain indicating subcortical regions from which intracerebral microinjection of drugs can be used to induce a conditioned place preference. Abbreviations: Amph = d-amphetamine sulphate; D-Ala = D-Alanine2-Met5-enkephalinamide; Mor = Morphine sulphate; NT = Neurotensin; Thiorp = Thiorphan, an enkephalinase inhibitor; ACB = Nucleus Accumbens; BC = brachium conjunctivum; CA = anterior commissure; CC = corpus callosum; FX = fornix; HPC = hippocampus; LHA = lateral hypothalamus; LM = medial lemniscus; NR = red nucleus; PVG = periventricular gray; VTA = ventral tegmental area.

Blockade of Conditioned Place Preference
with Selective Lesions

Major advances in identifying neurochemical substrates of drug reward have come from the use of selective lesions to transmitter-specific pathways in the brain. The neurotoxin 6-hydroxydopamine (6-OHDA) has been used routinely to lesion various catecholamine terminal areas and axons. Bilateral injections of 6-OHDA into the dopaminergic terminal field of the nucleus accumbens produced extinction-like responding in animals with a stable history of intravenous self-administration of cocaine (Roberts, Koob, Klonoff, & Fibiger, 1980). In contrast, lesions to ascending noradrenergic pathways did not affect cocaine self-administration (Roberts, Corcoran, & Fibiger, 1977). In a related study, rats in which dopaminergic nerve terminals in the nucleus accumbens had been destroyed by 6-OHDA failed to initiate self-administration of d-amphetamine despite nearly three weeks of post-lesion testing (Lyness, Friedle, & Moore, 1979). Together these data point to an involvement of central DA neurons in the rewarding properties of amphetamine and cocaine. Similar studies now have been conducted with these drugs as primary reinforcers in the conditioned place preference paradigm, and these results have identified possible differences in the substrates of amphetamine and cocaine reward.

Amphetamine, Cocaine, and Apomorphine

The attenuation of amphetamine-induced place preference by neuroleptic treatment (Spyraki et al., 1982a) and the lack of effect when cocaine is used as the primary reward (Spyraki et al., 1982b) provided the initial indication of a dissociation between the mechanisms by which these two drugs can be used to reinforce behavior. A similar pattern of results again emerged with animals receiving 6-OHDA lesions of the nucleus accumbens. In the experiment with amphetamine-induced place preference, there was a significant correlation (r = 0.76) between the absence of conditioned place preference and the magnitude of DA depletion in the nucleus accumbens (Spyraki et al., 1982a). No significant correlation was observed with the DA content in the striatum. Depletion of peripheral catecholamines by systemic injections of 6-OHDA did not affect d-amphetamine-induced place preference conditioning. In contrast to the effects obtained with amphetamine, neither 6-OHDA lesions of DA terminals in the nucleus accumbens nor 6-OHDA-induced destruction of central and/or peripheral noradrenergic systems affected cocaine-induced place preference conditioning (Spyraki et al., 1982b).

Resolution of these anomalous results may have been provided by the demonstration of significant place preference conditioning with the local anesthetic procaine at doses that did not affect locomotor activity (Spyraki et al., 1982b). Cocaine too has significant local anesthetic properties; and for reasons yet to be determined, the local anesthetic or other, non-dopaminergic effects of both procaine and cocaine after intraperitoneal administration may be sufficient to produce place preference conditioning. On the other hand, if it were possible to block selectively the local anesthetic properties of cocaine, then conceivably the drug could still produce place preference conditioning solely on the basis of the facilitation of dopaminergic neurotransmission. That is, cocaine may produce place preference by two independent mechanisms, with only one being dopaminergic.

The place-preference paradigm has been used to confirm the rewarding effects of apomorphine (Spyraki et al., 1982a; van der Kooy, Swerdlow, & Koob, 1983), reported previously with the intravenous self-administration procedure (Baxter, Gluckman, Stein, & Scerni, 1974). A dose of 0.5 mg/kg subcutaneously produced a clear preference (i.e., > 50% of time in compartment paired with the drug) when paired explicitly with a nonpreferred compartment (Spyraki et al., 1982a). Ambiguous results were obtained when apomorphine injections (0.01 to 10.0 mg/kg) were paired randomly with two distinct environments regardless of initial preference. Control rats with or without prior experience of apomorphine did not spend more time in the apomorphine as compared to the vehicle-paired side of the test chamber (van der Kooy et al., 1983). Despite the absence of a clear place preference in sham-operated controls, rats with bilateral 6-OHDA lesions of the nucleus accumbens did show a significant preference for the drug compartment over the environment associated with vehicle injections. This effect could be attributed to the development of DA receptor supersensitivity (Creese & Snyder, 1978) following destruction of the dopaminergic innervation of the nucleus accumbens. The fact that this result paralleled the potentiation of apomorphine-induced locomotion in the same subjects may again point to a common dopaminergic substrate for both the rewarding and locomotor effects of psychostimulant drugs.

Opiates and Opioid Peptides

Given the attenuation of heroin-induced place preference by pretreatment with neuroleptics (Spyraki et al., 1983), an obvious complement was to examine the effect of 6-OHDA lesions of the mesolimbic dopaminergic pathway at the level of the nucleus accumbens. Pairing of heroin (2.0 mg/kg) with the initially "nonpreferred" side of the shuttlebox produced a strong conditioned preference for the environmental cues associated with this drug. The percentage of time in the "drug" compartment switched from a 30% preconditioning score to 72% postconditioning. Bilateral lesions caused a 73% reduction of DA in the nucleus accumbens and olfactory tubercle. These depletions were accompanied by a significant reduction in time spent in the drug-paired compartment to approximately 50% of the test session (Spyraki et al., 1983).

Although the 6-OHDA lesions in the nucleus accumbens caused a partial lesion of the noradrenergic projection to the cortex and hippocampus, damage to these projections could not account for the attenuation of place preference. In a separate experiment neonatal injections of 6-OHDA were used to produce far more severe depletions of noradrenaline in cortex and hippocampus, and yet this treatment had no effect on heroin-induced place preference. These data are consistent with the lack of effect of noradrenergic lesions on cocaine self-administration (Roberts et al., 1977) and add to the growing skepticism about the role of noradrenaline in reward (Fibiger, 1978; Wise, 1978).

Successful conditioning of place preference with unilateral injections of the enkephalin analogue D-Ala (Phillips & LePiane, 1982) sets the stage for a more direct assessment of the role of the mesotelencephalic DA neurons in opioid induced reward (Phillips, LePiane, & Fibiger, 1983b). Two groups of rats, each with unilateral cannulae aimed at the VTA, were prepared with 6-OHDA lesions of the DA pathways either ipsilateral or contralateral to the injection site. Upon completion of the place preference conditioning with unilateral injections of D-Ala, the extent of each lesion was assessed biochemically. Eight of the 15 animals with ipsilateral lesions had greater than 95% depletion of forebrain DA levels. Seven animals had partial lesions (mean = 74.5% of control values). A similar dichotomy was observed with contralateral lesions. The behavioral data for the four groups revealed successful place preference conditioning in both of the contralateral lesion groups and the ipsilateral group with partial lesions. Conditioned place preference was not observed with animals receiving nearly complete lesions of the ipsilateral DA system (see Figure 5).

Comparison of Difference Procedures
for Identifying Neural Substrates of Drug Reward

Three different approaches currently are being used to identify the neural mechanism of drug reward. These include the use of the various mapping and lesion techniques in conjunction with the conditioned place preference paradigm as described in the present chapter, studies of lesion effects on intravenous self-administration (see Roberts & Zito, this volume), and intracranial self-administration procedures (see Bozarth, this volume). Although each procedure has been discussed separately and in detail in this volume, there are a number of important similarities and differences in the conclusions drawn from the use of these various procedures. These points will be discussed briefly with respect to an accurate description of the neural substrates of drug reward.

Effects of 6-OHDA lesions on morphine CPP
Figure 5: Upper: Schematic representation of unilateral 6-OHDA lesions of forebrain dopamine pathways contralateral or ipsilateral to microinjection sites in the VTA. Lower: Effects of partial (< 90%) or complete (> 90%) ipsilateral or contralateral depletion of forebrain dopamine on conditioned place preference induced by unilateral injections of D-Ala (0.2 mg) into the VTA. Adapted with permission from Phillips, LePiane, and Fibiger, 1983b.

There appears to be a clear consensus from studies using each of the strategies described above that a dopaminergic projection to the nucleus accumbens is important for the rewarding effects of amphetamine. Lesions of these dopaminergic neurons block acquisition of amphetamine self-administration (Lyness et al., 1979); amphetamine is self-administered directly into the nucleus accumbens (Monaco, Hernandez, & Hoebel, 1981); and local injections of amphetamine at this site can be used to induce a place preference (Carr & White, 1983).

Much more uncertainty surrounds the critical involvement of the nucleus accumbens in cocaine reward. On one hand studies with the place preference procedure and neuroleptics or denervation of DA neurons in the nucleus accumbens found no evidence for an exclusive role for DA in mediating the rewarding effects of cocaine (Spyraki et al., 1982b). These data contrast with those showing an attenuation of cocaine self-administration after similar treatments (Roberts et al., 1977, 1980) and may reflect the fact that several mechanisms underlie cocaine reward. One effect may not lend itself to an association with lever pressing in a response contingent manner and as a consequence is not detected by the intravenous self-administration paradigm. The failure to observe intracranial self-administration of cocaine into the nucleus accumbens, despite a clear indication of self-administration into the medial prefrontal cortex (Goeders & Smith, 1983), is also problematic for the nucleus accumbens hypothesis. It is likely that 6-OHDA lesions of the nucleus accumbens would also damage the dopaminergic projection to the prefrontal cortex; this damage may contribute to the blockade of cocaine self-administration following such lesions. Studies with selective lesions of the medial and sulcal prefrontal cortex would resolve this issue. The frontal cortex may contribute to the rewarding effects of both cocaine and amphetamine because rhesus monkeys have been shown to self-administer amphetamine directly into the orbitofrontal cortex (Phillips, Mora, & Rolls, 1981). It still remains to be determined if DA is essential for these effects.

Although a strong case can be presented in support of a dopaminergic substrate for opiate reward processes, this too is still subject to debate. Bilateral 6-OHDA lesions of the dopaminergic axons and terminals in the nucleus accumbens did not attenuate intravenous self-administration of heroin, although similar lesions blocked cocaine self-administration (Pettit, Ettenberg, Bloom, & Koob, 1984). Similarly, pretreatment with high doses of the DA antagonist alpha-flupenthixol blocked responding for cocaine but only produced a slight reduction in heroin self-administration (Ettenberg, Pettit, Bloom, & Koob, 1982). One possible explanation for the discrepancy between these data and the results from place conditioning studies (Phillips et al., 1983b) may be the existence of multiple neural substrates of opiate reward, only one of which is located in the VTA. Systemic administration of opiates would influence many different substrates and would therefore be relatively unaffected by damage to only one subsystem. In contrast, reward produced by direct intracranial activation of a specific substrate would be critically dependent upon the integrity of that single element.

All of the opiate studies with both the conditioned place preference paradigm and intracranial self-administration procedure (cf. Bozarth, 1983) provide strong complementary evidence for a ventral tegmental-nucleus accumbens (i.e., mesolimbic) dopaminergic substrate of opiate reward. An opiate drug presumably gains access to this system through an action at either the nucleus accumbens (Olds, 1982) or the VTA (Bozarth & Wise, 1981a; Phillips & LePiane, 1980). Given the failure to obtain intracranial self-administration of morphine in the lateral hypothalamus (Bozarth & Wise, 1982) despite initial reports of success (Olds, 1979), it seems unlikely that opiate reward is mediated by an action in the hypothalamus. The lack of effect of kainic acid lesions of hypothalamic neurons on heroin self-administration via the intravenous route (Britt & Wise, 1980) is consistent with this conclusion.

In summary, data from many laboratories using a variety of different procedures appear to support a dopaminergic involvement in some of the rewarding effects of both psychostimulants and opiate drugs. Part of the normal physiological function of this system may involve mediation of positive affective states in both humans and animals (Phillips, Broekkamp, & Fibiger, 1983a). If these two premises are correct, the acquisition and maintenance of drug self-administration is best attributed to the artificial induction of euphoria as distinct from the alleviation of withdrawal stress. This conjecture has received strong support from the recent demonstration that anatomically distinct opiate receptor fields mediate reward and physical dependence (Bozarth & Wise, 1984). Withdrawal symptoms appear to involve opiate action in the periventricular gray region independently from reward. In contrast, the rewarding effects of opiate injections into the VTA are not accompanied by any signs of physical dependence. Together these results confirm that dependence is not a prerequisite for opiate reward and that the latter is probably mediated to an important degree by a mesotelencephalic DA system.


The fruitful collaboration of F. G. LePiane, M. Martin-Iverson, R. Ortmann, and C. Spyraki on various phases of our experiments with conditioned place preference is acknowledged with pleasure. This research was funded by a program grant (PG-23) from the Medical Research Council of Canada.


Andrade, R., & Aghajanian, G. K. (1982). Neurotensin selectively activates dopaminergic neurons of the substantia nigra. Society for Neuroscience Abstracts, 7, 753.

Baxter, B. L., Gluckman, I. I., Stein, L., & Scerni, R. A. (1974). Self-injection of apomorphine in the rat: Positive reinforcement by a dopamine receptor stimulant. Pharmacology Biochemistry & Behavior, 2, 387-391.

Bozarth, M. A. (1983). Opiate reward mechanisms mapped by intracranial self-administration. In J. E. Smith & J. D. Lane (Eds.), The neurobiology of opiate reward processes (pp. 331-359). Amsterdam: Elsevier.

Bozarth, M. A., & Wise, R. A. (1981a). Intracranial self-administration of morphine into the ventral tegmental area in rats. Life Sciences, 28, 551-555.

Bozarth, M. A., & Wise, R. A. (1981b). Heroin reward is dependent on a dopaminergic substrate. Life Sciences, 29, 1881-1886.

Bozarth, M. A., & Wise, R. A. (1982). Localization of the reward-relevant opiate receptors. In L. S. Harris (Ed.), Problems of drug dependence, 1981 (National Institute on Drug Abuse Research Monograph 41, pp. 158-164). Washington, DC: U.S. Government Printing Office.

Bozarth, M. A., & Wise, R. A. (1984). Anatomically distinct opiate receptor fields mediate reward and physical dependence.Science, 224, 516-517.

Britt, M. D., & Wise, R. A. (1980). Effects of hypothalamic kainic acid lesions on self-administration of heroin and cocaine. Society for Neuroscience Abstracts, 6, 105.

Broekkamp, C. L., Phillips, A. G., & Cools, A. R. (1979a). Facilitation of self-stimulation behavior following intracerebral microinjection of opioids into the ventral tegmental area. Pharmacology Biochemistry & Behavior, 11, 289-295.

Broekkamp, C. L., Phillips, A. G., & Cools, A. R. (1979b). Stimulant effects of enkephalin microinjection into the dopaminergic A10 area. Nature, 278, 560-562.

Broekkamp, C. L. E., van den Bogaard, J. H., Heynen, H. J., Rops, R. H., Cools, A. R., & van Rossum, J. M. (1976). Separation of inhibiting and stimulating effects of morphine on self-stimulation behavior by intracerebral microinjections. European Journal of Pharmacology, 36, 443-450.

Carr, G. D., & White, N. M. (1983). Conditioned place preference from intra-accumbens but not intra-caudate amphetamine injections. Life Sciences, 33, 2551-2557.

Creese, I., & Snyder, S. H. (1978). Behavioral and biochemical properties of the dopamine receptor. In M. A. Lipton, A. Di Mascio, and K. F. Killam (Eds.), Psychopharmacology: A generation of progress (pp. 377-388). New York: Raven Press.

Ettenberg, A., Pettit, H. O., Bloom, F. E., & Koob, G. F. (1983). Heroin and cocaine intravenous self-administration in rats: Mediation by separate neural systems. Psychopharmacology, 78, 204-209.

Fibiger, H. C. (1978). Drugs and reinforcement mechanisms: A critical review of the catecholamine theory. Annual Review of Pharmacology and Toxicology, 18, 37-56.

Finnerty, E. P., & Chan, S. H. H. (1979). Morphine suppression of substantia nigra zona reticulata neurons in the rat: Implicated role for a novel striato-nigral feedback mechanism. European Journal of Pharmacology, 59, 307-310.

Glimcher, P. W., Giovino, A. A., Margolin, D. H., & Hoebel, B. G. (1984). Endogenous opiate reward induced by an enkephalin inhibitor, thiorphan, injected into the ventral midbrain. Behavioral Neuroscience, 98, 262-268.

Glimcher, P. W., Margolin, D. H., Giovino, A. A., & Hoebel, B. G. (1984). Neurotensin: A new ‘reward peptide.’ Brain Research, 291, 119-124.

Goeders, N. E., & Smith, J. E. (1983). Cortical dopaminergic involvement in cocaine reinforcement. Science, 221, 773-775.

Jacquet, Y. F., Klee, W. A., Rice, K. C., & Minamikawa, J. (1977). Stereospecific and nonstereospecific effects of (+) and (-) morphine: Evidence for a new class of receptors? Science, 198, 842-845.

Johanson, C. E., & Aigner, T. (1981). Comparison of reinforcing properties of cocaine and procaine in rhesus monkeys. Pharmacology Biochemistry & Behavior, 15, 49-53.

Johansson, D., Hokfelt, T., Elde, R. P., Schulzberg, M., & Terenius, L. (1978). Histochemical distribution of enkephalin neurons. Advances in Biochemical Psychopharmacology, 18, 51-70.

Joyce, E. M., & Iversen, S. D. (1979). The effect of morphine applied locally to mesencephalic dopamine cell bodies on spontaneous motor activity in the rat. Neuroscience Letters, 14, 207-212.

Katz, R. J., & Gormezano, G. (1979). A rapid and inexpensive technique for assessing the reinforcing effects of opiate drugs. Pharmacology Biochemistry & Behavior, 11, 231-234.

Lyness, W. H., Friedle, N. M., & Moore, K. E. (1979). Destruction of dopaminergic nerve terminals in nucleus accumbens: Effect on d-amphetamine self-administration. Pharmacology Biochemistry & Behavior, 11, 553-556.

Martin-Iverson, M. T., Ortmann, R., & Fibiger, H. C. (1985). Place preference conditioning with methylphenidate and nomifensine. Brain Research, 332, 59-67.Monaco, A. P., Hernandez, L., & Hoebel, B. G. (1981). Nucleus accumbens: Site of amphetamine self-injection: Comparison with the lateral ventricle. In R. B. Chronister and J. F. De France (Eds.), The neurobiology of the nucleus accumbens (pp. 338-343). Brunswick, ME: Haer Institute.

Mucha, R. F., & Iversen, S. D. (1984). Reinforcing properties of morphine and naloxone revealed by conditioned place preference: A procedural examination. Psychopharmacology, 82, 241-247.

Mucha, R. F., van der Kooy, D., O’Shaughnessy, M., & Bucenieks, P. (1983). Drug reinforcement studied by use of place conditioning in rat. Brain Research, 243, 91-105.

Olds, M. E. (1979). Hypothalamic substrates for the positive reinforcing properties of morphine in the rat. Brain Research, 168, 351-360.

Olds, M. E. (1982). Reinforcing effects of morphine in the nucleus accumbens. Brain Research, 237, 429-440.

Pert, C. B., Pert, A., Chang, J. K., & Fong, B. T. W. (1976). (D-Ala ) met-enkephalinamide: A potent long-lasting synthetic pentapeptide analgesic. Science, 194, 330-332.

Pettit, H. O., Ettenberg, A., Bloom, F. E., & Koob, G. F. (1984). Destruction of dopamine in the nucleus accumbens selectively attenuates cocaine but not heroin self-administration in rats. Psychopharmacology, 84, 167-173.

Phillips, A. G., Broekkamp, C. L., & Fibiger, H. C. (1983a). Strategies for studying the neurochemical substrates of drug reinforcement in rodents. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 7, 585-590.

Phillips, A. G., & LePiane, F. G. (1980). Reinforcing effects of morphine micro-injection into the ventral tegmental area. Pharmacology Biochemistry & Behavior, 12, 965-968.

Phillips, A. G., & LePiane, F. G. (1982). Reward produced by microinjection of (D-Ala ) Met -enkephalinamide into the ventral tegmental area. Behavioural Brain Research, 5, 225-229.

Phillips, A. G., LePiane, F. G., & Fibiger, H. C. (1983b). Dopaminergic mediation of reward produced by direct injection of enkephalin into the ventral tegmental area of the rat. Life Sciences, 33, 2505-2511.

Phillips, A. G., Mora, F., & Rolls, E. T. (1981). Intracerebral self-administration of amphetamine by rhesus monkeys. Neuroscience Letters, 24, 81-86.

Phillips, A. G., Spyraki, C., & Fibiger, H. C. (1982). Conditioned place preference with amphetamine and opiates as reward stimuli: Attenuation by haloperidol. In B. G. Hoebel and D. Novin (Eds.), The neural basis of feeding and reward (pp. 455-464). Brunswick, ME: Haer Institute.

Reicher, M. A, & Holman, E. W. (1977). Location preference and flavor aversion reinforced by amphetamine in rats. Animal Learning and Behavior, 5, 343-346.

Roberts, D. C. S., Corcoran, M. E., & Fibiger, H. C. (1977). On the role of ascending catecholamine systems in intravenous self-administration of cocaine. Pharmacology Biochemistry & Behavior, 6, 615-620.

Roberts, D. C. S., Koob, F. G., Klonoff, P., & Fibiger, H. C. (1980). Extinction and recovery of cocaine self-administration following 6-hydroxydopamine lesions of the nucleus accumbens. Pharmacology Biochemistry & Behavior, 12, 781-787.

Roques, B., Fournie-Zaluski, M., Soroca, E., Lecomte, J., Malfroy, B., Llorens, C., & Schwartz, J. C. (1980). Thiorphan shows antinociceptive activity in mice. Nature, 288, 286-288.

Rossi, N. A., & Reid, L. D. (1976). Affective states associated with morphine injections. Physiological Psychology, 4, 269-274.

Schuster, C. R., & Thompson, T. (1969). Self-administration of and behavioral dependence on drugs. Annual Review of Pharmacology, 9, 483-502.Schwartz, A. S., & Marchok, P. L. (1974). Depression of morphine seeking behavior by dopamine inhibition. Nature, 248, 257-258.

Sherman, J. E., Roberts, T., Roskam, S. E., & Holman, E. W. (1980). Temporal properties of rewarding and aversive behaviors of amphetamine in rats. Pharmacology Biochemistry & Behavior, 13, 597-599.

Spyraki, C., Fibiger, H. C., & Phillips, A. G. (1982a). Dopaminergic substrates of amphetamine-induced place preference conditioning. Brain Research, 253, 185-193.

Spyraki, C., Fibiger, H. C., & Phillips, A. G. (1982b). Cocaine-induced place preference conditioning: Lack of effects of neuroleptics and 6-hydroxydopamine lesions. Brain Research, 253, 195-203.

Spyraki, C., Fibiger, H. C., & Phillips, A. G. (1983). Attenuation of heroin reward in rats by disruption of the mesolimbic dopamine system. Psychopharmacology, 79, 278-283.

Stapleton, J. M., Lind, M. D., Merriman, V. J., Bozarth, M. A., & Reid, L. D. (1979). Affective consequences and subsequent effects on morphine self-administration of d-Ala2 methionine enkephalin. Physiological Psychology, 7, 146-152.

Uhl, G. R., Goodman, R. R., & Snyder, S. H. (1979). Neurotensin containing cell bodies, fibers, and nerve terminals in the brain stem of the rat: Immunohistochemical mapping. Brain Research, 167, 77-92.

van der Kooy, D., Mucha, R. F., O’Shaughnessy, M., & Bucenieks, P. (1982). Reinforcing effects of brain microinjections of morphine revealed by conditioned place preference. Brain Research, 243, 107-117.

van der Kooy, D., Swerdlow, N. R., & Koob, G. F. (1983). Paradoxical reinforcing properties of apomorphine: Effects of nucleus accumbens and area postrema lesions. Brain Research, 259, 111-118.

Weeks, J. R. (1962). Experimental morphine addiction: Method for automatic intravenous injections in unrestrained rats. Science, 138, 143-144.

Wise, R. A. (1978). Catecholamine theories of reward: A critical review. Brain Research, 152, 315-347.

©1987 Springer-Verlag (printed version)
©2000-2009 Addiction Science Network (web-enhanced version)

ASNet Home

ASNet Profile

Illicit Drug Index

Virtual Lab Tour

Research Reports

Drug Classification

A Primer on Drug Addiction

Experimental Methods

Treatment Resources

Biological Basis of Addiction

Before Prohibition: Early Psychoactive Medicines

Links to Other Websites

Click here to enter the Addiction Science Network Discussion Forum

Brain Reward System©1999-2009 Addiction Science Network
This page was last revised 05 April 2009 11:22 EDT.
Send comments to:
Report technical problems to: