Research reportNucleus accumbens shell and core dopamine: differential role in behavior and addiction
Introduction
The last quarter of the past century has seen a renewed interest in brain areas that belong to what we like to refer to as ‘the basement of the brain’. These ventrally and medially located areas include the ventral striatum and in particular the shell of the nucleus accumbens septi (NAc), as well as a number of nuclei that are part of the archistriatum [9], [122], [123]. Because of their strong homologies and reciprocal connections, these areas have been grouped into a complex, the extended amygdala, that includes the central amygdala, bed nucleus of stria terminalis, sublenticular substantia innnominata and intercalated grey masses [9], [122]. This nuclear complex corresponds to the oldest and most mysterious part of the forebrain, the one that, in contrast to the upper and more recent domains, has not undergone any major change throughout evolution. Also like any basement, the NAc shell/extended amygdala complex is full of those good-old-things that are kept, ‘just in case’ and turn out to be essential in exceptional, unpredictable circumstances. In the basement of the brain this dismissed merchandise corresponds to behavioral functions and response sets essential for the survival of the self and of the species. These functions can be grouped collectively under the heading of ‘motivation’, which refers to the ability, unique to living organisms, to respond to stimuli in relation to their individual needs and with the ultimate goal of the survival of their own species.
It must be admitted that a major impulse to the current interest on the areas of the ‘basement of the brain’ has been provided by their implication in drug addiction [70], [100], [167]. The reason why reinforcing drugs induce a condition of disturbed motivation like drug addiction is obscure and the subject of much speculation. However, to follow the basement analogy, one might consider the possibility that addiction is the result of the inadequate use of some of those basic ‘tools’ that are stored in the brain basement. This is, in a sense, the tenet of the present review. It will be argued that the abnormal activation of the mechanisms of the brain basement by certain drugs results in a disorder of learning that is expressed as drug addiction. It might be also argued that this mechanism could be utilized to explain other addictive behaviors motivated by non-drug stimuli (e.g. compulsive eating disorders).
This review can be schematically divided into six sections. The first section provides the terminological and theoretical background for the ensuing review, itself organized into four sections on the basis of two criteria: nature of the approach utilized (experimental vs. correlative) and nature of the reward (drug vs. non-drug) taken into consideration. Thus, the second section discusses the role of dopamine (DA) in responding motivated by conventional (non-drug) reinforcers as deduced from experimental manipulation of DA transmission by receptor-specific drugs and by lesions of DA neurons. In this section, the effect of local intracerbral drug infusion and of discrete lesions in relation to the role of the NAc and of its shell and core compartments is specifically discussed. The third section reviews the changes of in vivo DA transmission, estimated by microdialysis, in response to non-drug reinforcers. The fourth section analyzes the effect of the experimental manipulation of DA transmission on drug reward and reinforcement and the role of the NAc and its compartments. The fifth section deals with the effect of addictive drugs on DA transmission monitored in vivo. In the sixth section the data reviewed in the previous sections and in particular the effects of drug and non-drug reinforcers on DA transmission in the NAc are compared and this comparison is taken as a basis for an hypothesis of the role of DA in drug addiction and abnormal motivation.
Motivation is the process by which organisms emit responses to stimuli in relation to their predicted consequences in terms of survival of the self and of the species. Motivation consists therefore in learning of predictive relationships (contingencies) between salient stimuli and biologically meaningful ones and between responses and stimuli that follow (outcomes) or precede them. Learning of these contingencies enables the subject to actively promote by its actions the occurrence of biologically valuable events (instrumental action). The understanding of the mechanism by which goal-directed action is learned and maintained is central to the study of behavior and the subject of much debate and speculation. It is recognized that both cognitive, conscious (explicit/declarative), as well as associative, unconscious (implicit/procedural) mechanisms contribute to purposeful behavior [86], [266].
The basis for this form of motivated responding is hard-wired by evolution in the brain of organisms, including man. Organisms are provided with the innate ability of coding the intrinsic biological value of objects and organisms on the basis of the signals they emit and to respond to these signals in a manner consistent with that code [108]. Thus, certain stimuli such as the taste of sweet, the smell of a female, the cry of a predator, evoke behaviors that, depending on the stimulus, consist in approach or avoidance of the object or organism from which they originate. These responses are not the result of learning by experience of the consequences of the stimuli or of sheer imitation of the behavior of conspecifics. They are in fact primary, that is, unconditioned.
Pavlovian learning provides the opportunity for extending the biological code of a primary stimulus (reward or punisher) to other stimuli by learning of stimulus-reward contingencies (Pavlovian learning). By this mechanism novel salient stimuli (CS's) that reliably predict the occurrence of an unconditioned stimulus (US) acquire conditioned response-eliciting properties (CR) consistent with the valence of the US.
According to Konorski [166], CS's, depending on their nature, can elicit conditioned consummatory or preparatory responses. Conditioned consummatory responses are phenomenologically similar to the correspondent unconditioned response (UR) and can be understood as the result of the excitation by the CS of a representation of the US. Conditioned preparatory responses instead are not specific to a given US since, irrespective of it, consist of flexible patterns of orienting, approaching and exploring the CS. These typically incentive responses, in contrast to consummatory CR, are quite different from the response to the US (UR). For this reason, it is difficult to explain their emission as the result of the direct excitation by the CS of the representation of the US. According to Konorski [166], these preparatory (incentive) responses to the CS can be explained as due to excitation of a motivational system common to different US's. Thus, as a result of the association with the US, the representation of the CS establishes a connection with this motivational system (Pavlovian incentive learning) [88] thus acquiring the ability of inducing preparatory-incentive responses [166].
Incentives are commonly attributed many properties: (1) a directional property that promotes responses directed towards the incentive itself and, through it, towards the reward to which the incentive has been conditioned; (2) an activational property consisting of a state of motivational arousal (incentive arousal) that increases in a non-specific manner the incentive properties of other stimuli present in the environment but not necessarily related to the reward to which the triggering incentive has been conditioned. This arousing property of incentives can explain their ability to trigger, under appropriate conditions, the repetitive and excessive emission of behaviors that are part of the species repertoire (adjunctive or displacement behavior) [101], [160].
An important property of incentives is that of increasing the emission of responses instrumental to the presentation of the reward to which they have been conditioned. Thus, incentives acquired through Pavlovian (stimulus-reward) associations are capable of energizing primary reinforcement [41], [93], [175], [191], [271]. This property has been termed ‘transfer from Pavlovian to instrumental (PIT) responding’ [95].
Incentives acquired through Pavlovian contingency learning are also capable of acting as secondary reinforcers, promoting responding instrumental to their own presentation in the absence (extinction) of response reinforcement by the reward.
Pavlovian incentive learning, apart from increasing, through its activational effects, the probability of encountering a reward present in the environment, does not provide per se organisms with the ability of controlling by their actions the occurrence of biologically significant events. That is instead what instrumental learning does and what instrumental responding is about.
In the past, instrumental responding was explained in a rather mechanistic way as strengthening of the tendency to emit a response to a situational stimulus by its satisfying consequences [264] or as strengthening of the association between an arbitrary stimulus (S) and a response (R) by its consequences (e.g. feeding) [142], [195], [274].
This modality of instrumental responding might be approximated to the current notion of ‘habit responding’. In this instrumental modality, response is mainly controlled by stimuli that precede rather than follow it (outcomes) [85]. As a result of this, devaluation of response outcome or degradation of the instrumental act–outcome contingency fails to impair habit responding. Habit responding takes place as a result of exhaustive training on high ratio schedules or under variable interval schedules where reinforcement is loosely related to response [85].
However, under short-trials of continuous reinforcement schedules, where every response is reinforced, responding is tightly controlled by an act–outcome contingency and by the value of the outcome. The dependence of responding from a tight response-reward contingency would indicate that this form of instrumental responding (incentive instrumental responding) is controlled by the establishment of a declarative (conscious) representation of the cause–effect relationship between each act and its outcome [86], [267] or by a procedural (unconscious) response–outcome association [62]; on the other hand, the circumstance that responding is controlled by the current value of the reward would indicate that incentive instrumental responding is truly goal-directed in nature [86].
With practice, incentive (act–outcome) instrumental responding is transformed into habit responding based on S–R associations [85]; this modality ensures responding to stimuli at a speed that would be unattainable by incentive instrumental responding, due to its dependence on outcome. Habit responding, although automatic, is not impervious to adaptive control by its outcome. Thus, repeated failure to meet the requirements of a situational change, results in switching back from the habit modality to the incentive instrumental modality and than, after stabilization and practice, in the acquisition of a new habit. In this manner intentional act–outcome modalities (incentive instrumental responding) alternate with automatic habit modalities of responding in relation to the changing needs of the external world. Such interplay among different modalities of instrumental responding seems to apply also to addictive behavior [265].
Section snippets
Non-drug reward: experimental studies
Brain DA has been traditionally viewed as involved in the acquisition and expression of motivation and reinforcement. Here, the role attributed to DA in psychostimulant reinforcement has been simply extended to non-drug reinforcement. As told by Wise (1982, p. 39): “Because direct activation of dopaminergic synaptic activity by amphetamine, cocaine, and apomorphine is reinforcing in its own right… and because selective dopaminergic lesions or receptor blockade attenuates the reinforcing actions
Non-drug reward: correlative studies
Studies involving manipulation of DA transmission by drugs or lesions, while essential for providing experimental evidence for a role of DA in behaviour, are nonetheless unable to clarify the quantitative and temporal relationship between the activity of DA transmission and behaviour. To this end, correlative evidence obtained by monitoring the activity of DA neurons or the extracellular levels of DA are necessary.
A starting point of the correlative approach has been the study of the
Drug reward: experimental studies
From the point of view of the mechanism of action of reinforcing drugs it is useful to distinguish them into two broad categories, psychostimulants (including opiates, cannabinoids, ethanol, barbiturates, GHB and nicotine) and non-psychostimulants, including cocaine and amphetamine-like drugs (amphetamine, methamphetamine, phencyclidine, ecstasy, khat).
Psychostimulants increase the concentration of DA in the EC by acting directly on DA mechanisms either blocking DA reuptake (cocaine) [239] or
Drug reward and dependence: correlative studies
Transcerebral brain microdialysis studies of the effects of addictive drugs on DA transmission in the dorsal caudate–putamen and in the NAc have shown that, not only psychostimulants like cocaine [171] and amphetamine [49] but also narcotic analgetics [72], nicotine [149], ethanol [148] and phencyclidine [49] stimulate DA transmission preferentially in the NAc [71], as compared with the dorsal caudate–putamen. If one considers the NAc as the main area of the ventral striatum [123] and the
Drug addiction as abnormal motivation
The definition of ‘dependence’ (i.e. addiction) provided by the DSM-IIIR (American Psychiatric Association [10]) and DSM-IV (American Psychiatric Association [11]) consists of a list of seven criteria or conditions, at least three of which should be present at the same time to allow a diagnosis of dependence. Two of these criteria correspond to physiological adaptive changes (1) tolerance; (2) physical dependence; three of them correspond to loss of control over drug taking (3) persistent
Tolerance and dependence of dopamine transmission in the n.accumbens
Abstinence from chronic exposure to addictive drugs of different classes has been reported to induce a state of ‘anhedonia’ and dysforia that is expressed by a reduction in the reinforcing properties of natural rewards and electrical brain stimulation [68], [168]. Associated to this state is a reduction of in vivo DA transmission in the NAc and in the activity of DA units in the ventral tegmentum that appears dissociated from the physical signs of abstinence [4], [83], [84], [221], [238].
A natural history of drug-addiction
From the above discussion one gets the impression that each one of the available theories is unlikely to provide a comprehensive interpretation of drug addiction as only accounts for specific aspects of the whole process. The reason for this is that drug addiction is not a unitary process, related to a simple basic mechanism, but is made up of multiple processes. These processes are all related to the action of the drug but are complicated both by the multiple nature of drug action and by the
Acknowledgements
The advice of Adelaide Ciuti with the references is acknowledged. The studies made in the author laboratory and reproduced in this review have been made by Valentina Bassareo, Cristina Cadoni and Sandro Fenu. Their collaboration is gratefully acknowledged. Funds for these studies have been obtained from many sources including the Consiglio nazionale delle ricerche (CNR), Ministero dell'Università e della Ricerca (MURST/MIUR), European Community (EC), Regione Autonoma della Sardegna, Institute
References (292)
- et al.
Nucleus accumbens dopamine depletions make rats more sensitive to high ratio requirements but do not impair primary food reinforcement
Neuroscience
(1999) - et al.
Reward and reinforcement produced by drinking water: role of opioids and dopamine receptor subtypes
Pharmacol. Biochem. Behav.
(1993) - et al.
Reward and reinforcement produced by drinking sucrose: two processes that may depend on different neurotransmitters
Pharmacol. Biochem. Behav.
(1995) - et al.
New perspectives in basal forebrain organization of special relevance for neuropsychiatric disorders: the striatopallidal, amygdaloid and corticopetal components of substantia innominata
Neuroscience
(1988) - et al.
Effects of ibotenic acid lesions of the nucleus accumbens on instrumental action
Behav. Brain Res.
(1994) - et al.
Differential responsiveness of dopamine transmission to food-stimuli in nucleus accumbens shell/core compartments
Neuroscience
(1999) - et al.
The use of conditioned defensive burying to test the effects of pimozide on associative learning
Pharmacol. Biochem. Behav.
(1980) Dissociating the effects of altered dopaminergic function on performance and learning
Brain Res. Bull.
(1989)The role of dopamine in locomotor activity and learning
Brain Res.
(1983)- et al.
Dopamine D1-like receptors and reward-related incentive learning
Neurosci. Biobehav. Rev.
(1998)