Dopamine (3,4-dihydroxyphenylethylamine) is the predominant catecholamine neurotransmitter in the mammalian brain. Catecholamines are so called because they are amines (compounds derived by replacing the hydrogen atoms in ammonia with organic groups) that contain the aromatic alcohol catechol. Other catecholamine neurotransmitters and hormones are adrenaline (epinephrine) and "noradrenaline (norepinephrine). The catecholamines in the body are synthesized from the amino acid tyrosine. In fact, till the 1950s, dopamine has been considered merely as an intermediate metabolite in the pathway leading to the synthesis of noradrenaline and adrenaline. Only later was it discovered that dopamine itself is a neurotransmitter (reviewed in Cooper etal. 1996). In peripheral tissues dopamine regulates renal, cardiovascular, gastrointestinal, and other visceral functions. To the general public dopamine is known mainly because of its role in Parkinson's disease. This disease is caused by the degeneration of dopamine producing nerve cells in the brain. As dopamine does not penetrate the blood-brain barrier, it cannot be administered to the patient to replenish the brain with the missing chemical. Rather, the disease is treated with L-DOPA (l-dihy-droxyphenylalanine), which penetrates the brain and is converted there to dopamine.
A convenient "taxonomy of the dopamine systems in the brain is based on the length of their efferent fibres (ibid.). Local, or short projectional systems, exist in the retina and in the olfactory bulb. Intermediate length systems include the tuberohypophysial, incertohypo-thalamic, and medullary periventricular group. Longrange systems originate in dopamine neurons in the ventral tegmentum and substantia nigra and innervate striatal, "limbic, and "cortical areas (Figure 26). They comprise the mid-brain dopaminergic system, which is further differentiated into three pathways. The projection from the substantia nigra to the basal ganglia is termed the 'nigrostriatal pathway'. The projection to the septum, "amygdala, and the piriform cortex, is termed the 'mesolimbic dopaminergic system'. The projection to the prefrontal, cingulate, and entorhinal cortex is termed the 'mesocortical dopaminergic system'. Much has been learned in recent years about the cellular and molecular properties of the dopaminergic systems in the brain, including the metabolism of dopamine, its
Fig. 26 A schematic diagram of the long-range central dopaminergic systems in the mammalian brain. The nigrostriatal pathway (ns) projects from the substantia nigra (SN) to the basal ganglia (BG). The mesolimbic pathway (ml) projects from the mid-brain ventral tegmental area to some so-called '*limbic structures', e.g. the nucleus accumbens (NAC). The mesocortical pathway (mc) projects from the ventral tegmental area to the *cerebral cortex, e.g. frontal cortex. (Adapted from Stahl 1996.)
release and re-uptake, and its membrane "receptor families (Cooper et al. 1996; Missale et al. 1998).
Ample data indicate that dopamine plays a part in several facets of the 'reactivity' of the organism, including arousal, "attention, emotion, motivation, motor control, food intake, and endocrine regulation (Mason 1984; Wise and Rompre 1989; Robbins and Everitt 1996; Spanagel and Weiss 1999). It is also well established that dopamine is important in several types of learning. As the mid-brain dopaminergic systems are anatomically and pharmacologically heterogeneous, they are expected to subserve a number of physiological functions, which could contribute differentially to learning and memory. We will focus only on a few postulated functions.
An hypothesis that has guided the field for many years now is the 'dopamine hypothesis of reward'. It considers the mid-brain dopaminergic system as a common pathway for encoding the "reinforcing attributes of reward (Wise and Rompre 1989; Robbins and Everitt 1996; Nader et al. 1997). The dopamine hypothesis of reward rests on multiple lines of evidence. Its roots could be traced to a "classic set of experiments, in which Olds and Milner (1954) demonstrated that rats can be made to work hard, even to exhaustion, to self-stimulate certain centres in their brain via implanted electrodes.1 These 'reward centres' include the mid-brain dopaminergic system. It was later shown that dopaminergic blockers inhibit self-stimulation as well as food-elicited reward (Rolls et al. 1974), and addictive drugs, such as morphine, cocaine, amphetamine, and nicotine, increase the activity of mid-brain dopaminergic neurons (Pontieri et al. 1996; Nestler and Aghajanian
1997; Robbins and Everitt 1999; Berke and Hyman 2000). Furthermore, recording of neuronal activity from the brain of behaving animals has shown that dopaminergic neurons respond preferentially to rewarding stimuli (Schultz 1998).
But do mid-brain dopaminergic neurons actually encode the reward, or do they encode something else, which is required for the actualization of the neuronal and behavioural effect of the "reinforcer? Cellular analysis in the behaving "monkey, engaged in the "acquisition and the performance of a variety of "delay tasks, has shown that dopamine neurons respond to the salient stimuli whose detection is crucial for the learning, but do not themselves encode the reward. During learning, the response of these neurons transfers from the primary reward to the conditioned, reward-predicting stimuli. Moreover, the cellular response to the reward is highly characteristic: activation in response to rewarding events that are better than predicted, no response to events that are as predicted, and depression in response to events that are worse than predicted (Schultz 1998). It seems, hence, that the dopamine neurons encode the prediction error of the reward. Such prediction error drives learning in some formal "models (see the Rescorla-Wagner model in "algorithm; "surprise; Schultz et al. 1997). This is definitely an impressive example of cross-"level "reductive research, which binds learning theory, behavioural phenomena, brain neuroanatomy, and cellular and molecular mechanisms. It provides learning theorists with biological tools to test their predictions, and neurobiologists with conceptual frameworks to accommodate their findings.
Intriguing and influential as it is, the dopamine prediction-error hypothesis is still only one type of interpretation of the data. An alternative interpretation proposes that the fast dopamine response is not a teaching signal but rather an "attentional switch (Redgrave et al. 1999).
Much attention has been focused in recent years on the function of dopamine in the prefrontal cortex, including its role in "working memory (Goldman-Rakic 1995). The attentional-switch hypothesis of dopamine action fits well the proposed dopaminergic role in working memory, in which attention must be shifted quickly on and between the on-line and off-line "internal representations that are required to perform the ongoing memory task. Malfunction of the dopamine system, especially in the prefrontal cortex, is postulated to contribute to cognitive pathologies in which attention is impaired, particularly schizophrenia. This assumption, called the 'dopamine hypothesis of schizophrenia', is guiding the search for schizophrenia-linked defects in dopaminergic receptors, and for specific dopaminergic drugs to ameliorate the disease (e.g. Okubo et al. 1997; Lidow et al. 1998). The dopaminergic hypothesis, however, is only one of the 'theories' of schizophrenia (Willner 1997; Harrison 1999). By the way, those who find it difficult to think and learn under a distracting loud noise, should note that this could be related as well to dopaminergic prefrontal malfunction: the music in a modest discotheque suffices to push the dopaminergic system in the prefrontal cortex off balance (Arnsten and Goldman-Rakic 1998).
Dopamine could have also played a part in shaping one of the hallmarks of human behaviour, namely, novelty seeking. It has been suggested by some authors (Benjamin et al. 1996; Ebstein et al. 1996), although questioned by others (Gelernter et al 1997), that individuals who persistently seek novelty, owe this personality trait to a certain genetically determined composition of subtypes of dopaminergic receptors. Even if at the end of the day the role of dopamine in novelty seeking will be proven true, we should not expect to become daring adventurers just by swallowing dopaminergic drugs; as any other complex behavioural trait ("neurogenetics), novelty seeking is probably determined by many genes.
Selected associations: Algorithm, Attention, Neurotransmitter, Reduction, Reinforcer
'The original observation was fortuitous; see *reinforcer. The self-stimulation studies were replicated in many species, ranging from fish and chick, via goats, dogs, *monkeys, and dolphins, to humans (Olds 1969). Luckily, most humans do not normally have the opportunity to self-administer electrical pulses to their own mid-brain dopaminergic centres, but video games, which release striatal dopamine, could provide a safe substitute (Koepp et al. 1998).
Was this article helpful?