The basal ganglia are a group of interconnected subcortical nuclei that represent one of the brain's fundamental processing units.
The basal ganglia comprise two principal input nuclei, the striatum and the subthalamic nucleus (STN), and two principal output nuclei, the substantia nigra pars reticulata (SNr) and the internal globus pallidus (GPi) (primates) which in cats and rodents is known as the entopeduncular nucleus ( Figure 1). The external globus pallidus (GPe) is principally an intrinsic structure that receives most of its afferents from, and provides efferent connections to other basal ganglia nuclei. Finally, dopaminergic neurones in substantia nigra (pars compacta) (SNc) and the adjacent ventral tegmental area (VTA) provide other basal ganglia nuclei, principally the striatum, with important modulatory signals.
Striatum is the largest nucleus of the basal ganglia. In primates the striatum comprises the caudate nucleus and the putamen, and in all mammals, the ventral striatum or nucleus accumbens (Gerfen and Wilson 1996, Voorn et al. 2004). It receives direct input from most regions of the cerebral cortex and limbic structures including the amygdala and hippocampus. Additional input from sensorimotor and motivational regions of the brainstem arrives indirectly via relays in the thalamus. Finally, important modulatory afferents come from substantia nigra pars compacta (dopamine) and the raphe nuclei (serotonin) in the midbrain. The striatum is subdivided into sectors along a ventromedial-dorsolateral continuum largely on the basis of external connectivity (Voorn et al. 2004). All regions of the striatum are divided further into regions of patch/striosomes and matrix, again on the basis of differential connectivity and distribution of neurochemical markers (Gerfen and Wilson 1996). The principal cell type (representing >90% of all neurones) in all striatal regions is the GABAergic medium spiny neurone. Spiny neurones have been separated into two further populations according to which neuroactive peptide they contain (Substance P and dynorphin or Enkephalin) and the relative proportions of D1- and D2-type dopamine receptors they express. Striatal medium spiny neurones are GABAergic providing inhibitory inputs to adjacent spiny neurones via local axon collaterals, to the globus pallidus (external), and to both basal ganglia output nuclei. The remaining 5-10% of neurones in the striatum (fewer in rodents, more in primates) are either GABAergic or cholinergic interneurones, which can be distinguished according to neurochemical and in some cases morphological characteristics (for precise details of relative numbers see - Tepper and Bolam 2004).
Subthalamic nucleus was considered an important relay in the "indirect output pathway" from the striatum via the external globus pallidus (Albin et al. 1989). While still serving this function, it is now also considered a second important input nucleus of the basal ganglia (Nambu et al. 2002). Inputs external to the basal ganglia derive not only from large parts of frontal cortex, but also from various thalamic and brainstem structures. The subthalamic nucleus has a predominant cell type that is immunoreactive for glutamate that sends excitatory projections to both basal ganglia output nuclei and the external globus pallidus.
Globus pallidus (internal)/entopeduncular nucleus
Globus pallidus (internal)/entopeduncular nucleus is one of the two output nuclei that receive inputs from other basal ganglia nuclei and provides output to external targets in the thalamus and brainstem. Thus, it receives inhibitory GABAergic afferents from the striatum and external globus pallidus, and excitatory glutamatergic input from the subthalamic nucleus. Neurones of the internal globus pallidus are GABAergic and exert powerful inhibitory effects on targets in the thalamus the lateral habenula and the brainstem (Parent et al. 1999).
Substantia nigra pars reticulata
Substantia nigra pars reticulata is the second principal output nucleus also receiving afferents from other basal ganglia nuclei and providing efferent connections to the thalamus and brainstem. Inhibitory (GABAergic) inputs come from the striatum and globus pallidus (external) and excitatory input from the subthalamus (Gerfen and Wilson 1996). Pars reticulata neurones are also GABAergic and impose strong inhibitory control over parts of the thalamus and brainstem, including the superior colliculus, pedunculopontine nucleus and medullary reticular formation (Chevalier and Deniau 1990).
Globus pallidus (external)
Globus pallidus (external) is the principal "intrinsic" structure of the basal ganglia since it's major connections are with other basal ganglia nuclei. Thus, it receives inhibitory input from the striatum, excitatory input from the subthalamus, and provides GABAergic inhibitory efferent connections to all the basal ganglia's input and output nuclei (Chan et al. 2005). It also provides inhibitory input to the SNc (Parent et al. 1999).
Substantia nigra pars compacta/ventral tegmental area
Substantia nigra pars compacta/ventral tegmental area are regions of the ventral midbrain that contain the dopaminergic neurones that give rise to the nigrostriatal and mesolimbic/mesocortical projections (Lindvall and Bjorklund 1974). These pathways provide important modulatory signals both to other basal ganglia nuclei and to external structures (frontal cortex, septal area, amygdala, habenula). The highest concentration of dopaminergic terminals is in the striatum where they make synaptic and non-synaptic contacts with both medium spiny and interneurones (Sulzer 2005). Both pars compacta and the ventral tegmental area contain variable proportions of GABAergic neurones which make contact with nearby dopaminergic neurones (White 1996, Omelchenko and Sesack 2006). The main inputs to dopaminergic containing regions of the ventral midbrain come from other basal ganglia nuclei (Haber et al. 2000) and the brainstem (Kitai et al. 1999, Comoli et al. 2003); other afferent connections are from the frontal cortex (Frankle et al. 2006) and the amygdala (Gonzales and Chesselet 1990, Fudge and Haber 2000).
Direct and Indirect pathways
An influential view of the intrinsic organisation of the basal ganglia was proposed by Albin and colleagues (1989) ( Figure 2A). In their scheme, signals originating in cerebral cortex are distributed to the two populations of striatal medium spiny output neurones. Neurones containing Substance P and a preponderance of D1-type dopamine receptors make "direct" contact with the basal ganglia output nuclei -- the direct pathway. While, striatal neurones containing Enkephalin and express mainly D2-type dopamine receptors make "indirect" contact with the output nuclei via relays in the globus pallidus and subthalamus -- the indirect pathway. Basal ganglia output was thought to reflect a balance between these two projections.
Additional anatomical observations ( Figure 2B) have, however, revealed a more complex organisation. The main findings are as follows:
- both populations of striatal output neurones project to the globus pallidus (external), one exclusively (Enkephalin/D2 neurones), the other via collaterals from the fibres innervating the output nuclei (Substance P/D1 neurones) (Parent et al. 2000);
- globus pallidus neurones make direct contact with the output nuclei as well as to the subthalamus, often with branching collaterals to all three structures (Smith et al 1998);
- the globus pallidus also projects back to the striatum (Bevan et al. 1998);
- major input to the subthalamic nucleus originates from both cortical and sub-cortical structures external to the basal ganglia, consequently, it is now considered a major input structure, rather than a simple relay in the "indirect-pathway" (Nambu et al. 2002).
Although the overall pattern of intrinsic circuitry is complex ( Figure 2B), connections between components of the basal ganglia are topographically ordered. Some of these projections are comparatively focused (e.g. the striato-nigral projection), others more diffuse (e.g. the subthalamo-nigral projection) (Mink 1996). Differences in the comparative numbers of neurones in the striatum and the output nuclei suggest a dramatic compression of information as it is processed within the basal ganglia (Oorschot 1996).
Input to the striatum from all major sources, the cerebral cortex, limbic structures and the thalamus are also topographically ordered (Voorn et al. 2004). Terminals from some sources (cerebral cortex and central lateral thalamic nucleus) appear to make few contacts with many striatal neurones while inputs from other regions (parafasicular thalamic nucleus) have many contacts with fewer individual striatal neurones (Gerfen and Wilson 1996). Afferent connections to the subthalamic nucleus, at least from cerebral cortex, are also topographically organised (Nambu et al. 2002).
Basal ganglia outputs contact regions of the thalamus (the intralaminar and ventromedial nuclei) that project directly back to basal ganglia input nuclei (Kimura et al. 2004, Smith et al. 1998) but also back to those regions of cortex providing original inputs to the striatum (Alexander et al. 1986). Similarly, basal ganglia outputs to the brainstem tend to target those regions that provide indirect input to the striatum via the thalamic midline and intralaminar nuclei (McHaffie et al. 2005). Projections from the basal ganglia output nuclei to the thalamus (Mengual et al. 1999) and brainstem (Mana and Chevalier 2001) are also topographically ordered. In addition, many output projections of the basal ganglia are extensively collateralised (Cebrian et al. 2005) suggesting that divergent targets in the thalamus, midbrain and hindbrain may be influenced simultaneously.
Manifest topographies associated with input projections, intrinsic connections and outputs of the basal ganglia provided a basis for the influential organisational principle suggested by Alexander and colleagues (1986). Connections between the cerebral cortex and basal ganglia can be viewed as a series of parallel projecting, largely segregated cortico-striato-nigro-thalamo-cortical loops or channels ( Figure 3). Thus, an important component of the projections from different functional territories of cerebral cortex (e.g. limbic, associative, sensorimotor) project to exclusive functional territories in the basal ganglia input nuclei, which are then maintained in the internal circuitry. Output signals from functional territories represented in the output nuclei are returned, via appropriate thalamic relays, to the cortical regions providing the original input signals (Middleton and Strick 2000).
The concept of potentially segregated parallel projecting loops through the basal ganglia has been extended to their connections with sensorimotor and motivational structures in the brainstem, including the superior colliculus, periaqueductal grey, pedunculopontine and parabrachial nuclei (McHaffie et al. 2005). An important difference is that, in the case of cortical loops, the thalamic relay is on the output side of the loop, whereas for the sub-cortical loops the thalamic relay is on the input side ( Figure 4). Much work will be required to test whether projections from different brainstem structures, as they pass through the thalamic and basal ganglia relays, represent functionally segregated channels.
Input signals to the striatum
Signals received by the striatum from the cerebral cortex and thalamus are mediated by excitatory glutamatergic neurotransmission (Gerfen and Wilson 1996). These fast, phasically active excitatory inputs are mediated predominantly by AMPA and kainate receptor subtypes when the medium spiny neuronal membranes are near resting potential, with NMDA receptors playing a great role when the membranes are depolarised. Glutamatergic inputs from both cerebral cortex and thalamus also impinge on striatal interneurones (Tepper and Bolam 2004, Smith et al. 1998). The effects of dopaminergic inputs on striatal neuronal activity are complicated with many conflicting results (Nicola et al. 2000). Problems undoubtedly arise because it is difficult to evoke in slice and anaesthetised prepartions the appropriately timed cortically and thalamically based inputs with which dopaminergic signals will interact (see below). However, the current weight of evidence suggests dopamine can increase signal-to-noise ratios in the striatum -- enhancing the effects of strong inputs while suppressing weak ones (Nicola et al. 2004). The actions of dopamine on GABAergic and cholinergic interneurones (Tepper and Bolam 2004, Nicola et al. 2000) may also contribute. Although anatomically significant (Soubrie et al. 1984), much less is known about the role(s) of serotoninergic inputs to the basal ganglia.
Input signals to the subthalamic nucleus
The main external sources of input to the striatum also provide parallel inputs to the subthalamic nucleus. The subthalamus, therefore, receives phasic excitatory glutamatergic signals both from cerebral cortex (Nambu et al. 2002) and the thalamus (Mouroux and Feger 1993). Following cortical stimulation short-latency excitatory effects in the subthalamus are thought to be mediated via these "hyperdirect" pathways while longer latency suppressive effects more likely come from indirect inhibitory inputs from other basal ganglia nuclei, principally the external globus pallidus (Nambu et al. 2002). Modulatory dopaminergic and serotoninergic inputs appear to produce local excitation in the subthalamus (Ni et al. 2001, Xiang et al. 2005). Finally, and unlike the striatum, the subthalamus is modulated by additional cholinergic signals from the tegmental pedunculopontine nucleus (Mena-Segovia et al. 2004).
The manner by which the basal ganglia exert influence over target structures is by a fundamental process of disinhibition (Chevalier and Deniau 1990) ( Figure 5). GABAergic neurones in the basal ganglia output nuclei have high tonic firing rates (40-80 Hz). This activity ensures that target regions of the thalamus and brainstem are maintained under a tight and relatively constant inhibitory control. Focused excitatory inputs from external structures to the striatum can impose a focused suppression, (mediated via "direct" GABAergic inhibitory connections), on sub-populations of output nuclei neurones. This focused reduction of inhibitory output activity effectively releases or disinhibits associated target regions in the thalamus (e.g. ventromedial nucleus) and brainstem (e.g. superior colliculus) from normal inhibitory control.
In humans, basal ganglia dysfunction has been associated with numerous debilitating conditions including Parkinson's disease, Huntington's disease, Tourette's syndrome, schizophrenia, attention-deficit disorder, obsessive-compulsive disorder, and many of the addictions. To understand and correctly interpret how a complicated system such as the basal ganglia can malfunction, it is useful to appreciate how the network works normally. What are the normal functions of basal ganglia circuitry? Two recurring themes in basal ganglia literature point to their involvement in action selection and reinforcement learning.
Despite numerous suggestions that the basal ganglia are involved in a wide range of functions including perception, learning, memory, attention, many aspects of motor function, even analgesia and seizure suppression, increasingly evidence points to an underlying role in basic selection processes (Mink 1996, Redgrave et al. 1999).
- Selection is an old problem: The anatomical connections and neurotransmitters systems of the basal ganglia in vertebrate species are remarkably similar, suggesting that the evolution of these structures has been very conservative (Medina and Reiner 1995). Consequently, whatever computational problems the basal ganglia evolved to solve, they were likely to be as much problems for early vertebrates as they are for us today. A likely possibility is that multifunctional agents typically have to express different functional outputs through a shared motor resource - the final common motor path. A fundamental requirement is to determine which functional system should be allowed control of the motor output at any time. This selection problem is one shared by all vertebrates and has not changed materially over the course of evolution, despite great changes in the range, power and sophistication of systems competing for expression.
- The basal ganglia can select: The macro-architecture of the basal ganglia appears to be configured for selection ( Figure 6). The parallel loops originating from and returning to diverse cortically and sub-cortically based functional systems (Alexander et al. 1986, McHaffie et al. 2005) convey phasic excitatory signals (bids for selection) to the input nuclei. Depending on comparative magnitudes of "input saliences", channels returning to structures providing the most "salient" inputs would be selectively disinhibited. Returning disinhibitory signals may permit the sensory/cognitive inputs to the targeted functional system access to the shared motor resource. Maintained or increased levels of tonic inhibitory signals in non-selected channels would prevent the output of non-selected target structures accessing the common motor path. Independent of any biological considerations, a similar "central-selection" control architecture was devised to select the actions of an autonomous mobile robot (Snaith and Holland 1990). Subsequently, it has been confirmed that a biologically constrained model of basal ganglia architecture can do likewise (Prescott et al. 2006).
- Open and closed loops: A fundamental requirement for selection is that activity within the functionally segregated loops should interact. Consequently, at each major relay point within each of the basal ganglia loops (input nuclei, output nuclei, and the thalamus) signals flowing in the parallel channels can be subjected to influences originating outside the loop (Joel and Weiner 1994). With the selection hypothesis in mind, mechanisms within the internal circuitry can be identified that would promote "selection", in part by permitting different channels to influence each other ( Figure 7):
- Excitatory inputs to an individual spiny neurone must be sufficiently synchronised to depolarise the membrane of medium spiny neurones to an "up-state" where it can fire action potentials (Gerfen and Wilson 1996). This mechanism might represent an initial filter to exclude "weak" competitors.
- Local inhibitory collaterals between striatal spiny neurones (Plenz 2003) and longer range inhibitory effects of interneurones (Tepper and Bolam 2004) should cause highly activated striatal elements to suppress activity in more weakly activated channels.
- At the level of basal ganglia output nuclei, the imposition of focused inhibition from the striatum onto a more diffuse excitation from the subthalamic nucleus should cause an inhibited (selected) centre with an excitatory (non-selected) surround (Mink 1996).
- Local inhibitory collaterals between output nuclei neurones (Mailly et al. 2003) should further "sharpen" the difference between inhibited and non-inhibited channels. Together these mechanisms can be viewed as a sequence of mechanisms for selection.
- Canonical micro-architectures: The internal micro-architecture of each basal ganglia structure is retained across the representations of different functional territories. Insofar as function is an emergent property of connectivity, the presence of common architectures suggests that similar computational processes are applied to inputs from drastically different functional origins. It is noteworthy that goal directed behaviour can be conceived as a three tier hierarchy with selections required at each level (Redgrave et al. 1999):
- selections of overall goal;
- selections of actions to achieve a selected goal; and
- selections of movements to achieve a selected action.
Thus, the same micro-circuitry shared by the different functional territories of the basal ganglia could, in principle, select between competing "goals", "actions" and "movements". It is therefore relevant that across the ventromedial-dorsolateral gradient proposed for the striatum (Voorn et al. 2004), inputs to ventromedial sectors come from structures in which competing behavioural goals may be represented (prefrontal cortex, amygdala, hippocampus), while the connections of dorsolateral sectors are from regions that guide movements (e.g. sensory and motor cortex). Consequently, it is not difficult, to map the conceptual framework of goal, action and movement selections onto the "spiral architecture" proposed by Suzanne Haber (Haber et al. 2000) for successive connections between the limbic, associative and sensorimotor territories of the basal ganglia.
The basal ganglia have long been associated with the processes of reinforcement learning (Schultz 2006; see also Reward Signals). This should not be surprising since instrumental or operant conditioning (the class of learning most commonly linked to the basal ganglia) can be viewed as the biasing of future action selections by past action outcomes. One of the strongest lines of evidence supporting the involvement of the basal ganglia in reinforcement learning is the electrophysiological data obtained from behaving monkeys. Typically, unexpected biologically significant events including sudden novel stimuli, intense sensory stimuli, primary rewards, and arbitrary stimuli classically conditioned by association with primary rewards evoke a stereotypic sensory response from DA neurones in many species (Schultz 1998). This response comprises a characteristic short latency (70-100 ms), short duration (<200 ms) burst of activity. However, it is the capacity of phasic DA responses to change when experimental conditions are altered that has provoked most interest.
- The novelty response of DA neurones habituates rapidly when a sensory stimulus is repeated in the absence of behaviourally rewarding consequences.
- A phasic DA response will emerge following the presentation of a neutral sensory stimulus that predicts a primary reward. Under these conditions the DA responses to the predicted reward gradually diminish.
- When a predicted reward is omitted, a reliable depression in the spontaneous activity of the DA neurones occurs 70-100 ms after the time of expected reward delivery.
It is largely on the basis of these data that the reward-prediction error hypothesis was originally formulated. More recently, additional supporting investigations have established that the phasic DA signal complies with the contiguity, contingency and prediction error tenets of contemporary learning theories (Schultz 2006). This body of evidence provides powerful support for the reward prediction error hypothesis which is now widely accepted by both biological and computational neuroscientists. Within this framework, the hypothesised errors in reward prediction signalled by phasic dopamine activity are presumed teaching signals for appetitive learning and ensure that actions maximising the future acquisition of reward are selected more often.
However, recent evidence from studies that have identified sources of short-latency sensory input to midbrain dopaminergic neurones suggests that, in real world conditions where unexpected stimuli are both temporally and spatially unpredictable, the identity of unexpected events (and hence their reward value) will be determined after, rather than before the time of phasic dopaminergic signalling (Redgrave and Gurney 2006).
- Although dopamine neurones have reliable responses to reward-related stimuli they also exhibit strong phasic responses to unexpected sensory events that have no obvious appetitive reinforcement consequences (Horvitz 2000).
- Despite reward-related stimuli coming in all sorts of shapes and sizes, the phasic dopamine signal is highly stereotyped (latency \(\sim\) 100 ms, duration \(\sim\) 100 ms) and largely independent of animal species, stimulus modality, and perceptual complexity of eliciting events (Schultz 1998).
- The 100 ms response latency of dopaminergic neurones is reliably shorter than the latency of the gaze-shift that brings the unexpected event onto the fovea for detailed analysis by cortical visual systems. Necessarily this means that dopamine responses are triggered as a consequence of limited pre-attentive, pre-saccadic sensory processing (Redgrave and Gurney 2006).
- Recent evidence indicates that the sensory inputs to dopaminergic neurones derive largely, if not exclusively as a consequence of early, subcortical sensory processing (Redgrave and Gurney 2006). In the case of vision, the midbrain superior colliculus is configured to indicate where an unexpected event is rather than what it is (Wurtz and Albano 1980). Perhaps it is no coincidence that, in almost all studies showing phasic dopamine signals can signal reward prediction errors (Schultz 2006), the economic values predicted by the conditioned stimuli are correlated with the spatial location of stimulus presentation. It therefore remains to be determined whether dopamine neurones can signal continuous values of reward prediction errors in real world conditions where unexpected events are both temporally and spatially unpredictable.
In the light of these considerations, it has been suggested that short-latency signalling by dopaminergic neurones may be suited more to reinforcing a form of learning with less stringent perceptual requirements(Regrave and Gurney 2006). Specifically, short-latency dopamine reinforcement signals could promote the discovery of agency (i.e. those initially unpredicted events that are caused by the agent) and subsequent identification of critical causative actions, irrespective of the outcome's economic value. This hypothesis is based on the sensory and motor signals likely to be present in target structures (principally the striatum) at the time of the precisely timed phasic dopamine response ( Figure 8). The role of dopamine in this scheme is to promote the reselection of components of behaviour and context that immediately precede unpredicted sensory events ( Figure 9). When the animal/agent is the cause of an event, repeated trials should enable the basal ganglia to converge on behavioural and contextual components that are critical for eliciting it, leading to the emergence of a novel action.
If action selection and reinforcement learning are normal functions of the basal ganglia, it should be possible to interpret many of the human basal ganglia-related disorders in terms of selection malfunctions. For example, the akinesia of Parkinson's disease may be seen as a failure to inhibit tonic inhibitory output signals on any of the sensorimotor channels. Aspects of schizophrenia, attention deficit disorder and Tourette's syndrome could reflect different forms of failure to maintain sufficient inhibitory output activity in non-selected channels. Conseqently, insufficiently inhibited signals in non-selected target structures could interfere with the output of selected targets (expressed as motor/verbal tics) and/or make the selection system vulnerable to interruption from distracting stimuli (schizophrenia, attention deficit disorder). The opposite situation would be where the selection of one functional channel is abnormally dominant thereby making it difficult for competing events to interrupt or cause a behavioural or attentional switch. Such circumstances could underlie addictive compulsions or obsessive compulsive disorder. Finally, the new hypothesis of phasic dopamine function (Redgrave and Gurney 2006) could provide further insights into behavioural stereotypies and disturbances of the sense of agency.
Albin, R. L., Young, A. B. and Penney, J. B. The functional anatomy of basal ganglia disorders. Trends Neurosci. 12, 366-375 (1989).
Alexander, G. E., DeLong, M. R. and Strick, P. L. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 9, 357-381 (1986).
Bevan, M. D., Booth, P. A. C., Eaton, S. A. and Bolam, J. P. Selective innervation of neostriatal interneurons by a subclass of neuron in the globus pallidus of the rat. J Neurosci 18, 9438-9452 (1998).
Cebrian, C., Parent, A. and Prensa, L. Patterns of axonal branching of neurons of the substantia nigra pars reticulata and pars lateralis in the rat. J Comp Neurol 492, 349-369 (2005).
Chan, C. S., Surmeier, D. J. and Yung, W. H. Striatal information signaling and integration in globus pallidus: Timing matters. Neurosignals 14, 281-289 (2005).
Chevalier, G. and Deniau, J. M. Disinhibition as a basic process in the expression of striatal functions. Trends Neurosci. 13, 277-281 (1990).
Comoli, E. et al. A direct projection from superior colliculus to substantia nigra for detecting salient visual events. Nat Neurosci 6, 974-80 (2003).
Frankle, W. G., Laruelle, M. and Haber, S. N. Prefrontal cortical projections to the midbrain in primates: evidence for a sparse connection. Neuropsychopharmacol 31, 1627-36 (2006).
Fudge, J. L. and Haber, S. N. The central nucleus of the amygdala projection to dopamine subpopulations in primates. Neuroscience 97, 479-494 (2000).
Gerfen, C. R. and Wilson, C. J. in Handbook of chemical neuroanatomy, Vol 12: Integrated systems of the CNS, Part III. (eds. Swanson, L. W., Bjorklund, A. and Hokfelt, T.) 371-468 (Elsevier, Amsterdam, 1996).
Gonzales, C. and Chesselet, M.-F. Amygdalonigral pathway: An anterograde study in the rat with Phaseolus vulgaris leucoagglutinin (PHA-L). J Comp Neurol 297, 182-200 (1990).
Haber, S. N., Fudge, J. L. and McFarland, N. R. Striatonigrostriatal pathways in primates form an ascending spiral from the shell to the dorsolateral striatum. J Neurosci 20, 2369-2382 (2000).
Horvitz, J. C. Mesolimbocortical and nigrostriatal dopamine responses to salient non-reward events. Neuroscience 96, 651-656 (2000).
Joel, D. and Weiner, I. The organization of the basal ganglia-thalamocortical circuits: open interconnected rather than closed segregated. Neuroscience 63, 363-379 (1994).
Kimura, M., Minamimoto, T., Matsumoto, N. and Hori, Y. Monitoring and switching of cortico-basal ganglia loop functions by the thalamo-striatal system. Neurosci Res 48, 355-360 (2004).
Kitai, S. T., Shepard, P. D., Callaway, J. C. and Scroggs, R. Afferent modulation of dopamine neuron firing patterns. Curr Opin Neurobiol 9, 690-697 (1999).
Lindvall, O. and Bjorklund, A. The organization of the ascending catcholamine neuron systems in the rat brain as revealed by the glyoxylic acid fluoresence method. Acta Physiol Scand Suppl. 412, 1-48 (1974).
Mailly, P., Charpier, S., Mentrey, A. and Deniau, J. M. Three-dimensional organization of the recurrent axon collateral network of the Substantia Nigra Pars Reticulata neurons in the rat. J Neurosci 23, 5247-5257 (2003).
Mana, S. and Chevalier, G. The fine organization of nigro-collicular channels with additional observations of their relationships with acetylcholinesterase in the rat. Neuroscience 106, 357-374 (2001).
McHaffie, J. G., Stanford, T. R., Stein, B. E., Coizet, W. and Redgrave, P. Subcortical loops through the basal ganglia. Trends Neurosci 28, 401-407 (2005).
Medina, L. and Reiner, A. Neurotransmitter organization and connectivity of the basal ganglia in vertebrates: Implications for evolution of basal ganglia. Brain Behav Evol 46, 235-258 (1995).
Mena-Segovia, J., Bolam, J. P. and Magill, P. J. Pedunculopontine nucleus and basal ganglia: distant relatives or part of the same family? Trends Neurosci 27, 585-588 (2004).
Mengual, E., de las Heras, S., Erro, E., Lanciego, J. L. and Gimenez-Amaya, J. M. Thalamic interaction between the input and the output systems of the basal ganglia. J Chem Neuroanat 16, 187-200 (1999).
Middleton, F. A. and Strick, P. L. Basal ganglia and cerebellar loops: motor and cognitive circuits. Brain Res Rev 31, 236-250 (2000).
Mink, J. W. The basal ganglia: Focused selection and inhibition of competing motor programs. Prog Neurobiol 50, 381-425 (1996).
Mouroux, M. and Feger, J. Evidence that the parafascicular projection to the subthalamic nucleus is glutamatergic. Neuroreport 4, 613-5 (1993).
Nambu, A., Tokuno, H. and Takada, M. Functional significance of the cortico-subthalamo-pallidal 'hyperdirect' pathway. Neurosci Res 43, 111-117 (2002).
Ni, Z. G., Gao, D. M., BoualiBenazzouz, R., Benabid, A. L. and Benazzouz, A. Effect of microiontophoretic application of dopamine on subthalamic nucleus neuronal activity in normal rats and in rats with unilateral lesion of the nigrostriatal pathway. Eur J Neurosci 14, 373-381 (2001).
Nicola, S. M., Surmeier, D. T. and Malenka, R. C. Dopaminergic modulation of neuronal excitability in the striatum and nucleus accumbens. Annu Rev Neurosci 23, 185-215 (2000).
Nicola, S. M., Woodward Hopf, F. and Hjelmstad, G. O. Contrast enhancement: a physiological effect of striatal dopamine? Cell Tissue Res 318, 93-106 (2004).
Omelchenko, N. V. and Sesack, S. R. in SFN Abstr 722.12. (Atlanta, GA, 2006).
Oorschot, D. E. Total number of neurons in the neostriatal, pallidal, subthalamic, and substantia nigral nuclei of the rat basal ganglia: a stereological study using the cavalieri and optical disector methods. J Comp Neurol 366, 580-599 (1996).
Parent, A. et al. Organization of the basal ganglia: the importance of axonal collateralization. Trends Neurosci 23, S20-S27 (2000).
Parent, M., Levesque, M. and Parent, A. The pallidofugal projection system in primates: evidence for neurons branching ipsilaterally and contralaterally to the thalamus and brainstem. J Chem Neuroanat 16, 153-165 (1999).
Plenz, D. When inhibition goes incognito: feedback interaction between spiny projection neurons in striatal function. Trends Neurosci 26, 436-443 (2003).
Prescott, T. J., Gonzalez, F. M. M., Gurney, K., Humphries, M. D. and Redgrave, P. A robot model of the basal ganglia: Behavior and intrinsic processing. Neural Networks 19, 31-61 (2006).
Redgrave, P. and Gurney, K. The short-latency dopamine signal: a role in discovering novel actions? Nature Rev Neurosci 7, 967-975 (2006).
Redgrave, P., Prescott, T. and Gurney, K. N. The basal ganglia: A vertebrate solution to the selection problem? Neuroscience 89, 1009-1023 (1999).
Schultz, W. and Dickinson, A. Neuronal coding of prediction errors. Annu Rev Neurosci 23, 473-500 (2000).
Schultz, W. Behavioral theories and the neurophysiology of reward. Annu Rev Psychol 57, 87-115 (2006).
Schultz, W. Predictive reward signal of dopamine neurons. J Neurophysiol 80, 1-27 (1998).
Smith, Y., Bevan, M. D., Shink, E. and Bolam, J. P. Microcircuitry of the direct and indirect pathways of the basal ganglia. Neuroscience 86, 353-387 (1998).
Snaith, S. and Holland, O. in From Animals to Animats: Proceedings of the First International Conference on the Simulation of Adaptive Behaviour (eds. Meyer, J.-A. and Wilson, S.) 255-262 (MIT Press, Cambridge, MA, 1990).
Soubrie, P., Reisine, T. D. and Glowinski, J. Functional aspects of serotonin transmission in the basal ganglia: a review and an in vivo approach using the push-pull cannula technique. Neuroscience 13, 605-25 (1984).
Sulzer, D. The complex regulation of dopamine output: A review of current themes. Clin Neurosci Res 5, 117-121 (2005).
Tepper, J. M. and Bolam, J. P. Functional diversity and specificity of neostriatal interneurons. Curr Opin Neurobiol 14, 685-92 (2004).
Voorn, P., Vanderschuren, L. J., Groenewegen, H. J., Robbins, T. W. and Pennartz, C. M. Putting a spin on the dorsal-ventral divide of the striatum. Trends Neurosci 27, 468-74 (2004).
White, F. J. Synaptic regulation of mesocorticolimbic dopamine neurons. Annu Rev Neurosci 19, 405-36 (1996).
Wurtz, R. H. and Albano, J. E. Visual-motor function of the primate superior colliculus. Annu Rev Neurosci 3, 189-226 (1980).
Xiang, Z. X., Wang, L. and Kitai, S. T. Modulation of spontaneous firing in rat subthalamic neurons by 5-HT receptor subtypes. J Neurophysiol 93, 1145-1157 (2005).
- Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.
- Eugene M. Izhikevich (2006) Bursting. Scholarpedia, 1(3):1300.
- Olaf Sporns (2007) Complexity. Scholarpedia, 2(10):1623.
- Eugene Roberts (2007) Gamma-aminobutyric acid. Scholarpedia, 2(10):3356.
- Okihide Hikosaka (2007) Habenula. Scholarpedia, 2(6):2703.
- Jim Houk (2007) Models of basal ganglia. Scholarpedia, 2(10):1633.
- Peter Jonas and Gyorgy Buzsaki (2007) Neural inhibition. Scholarpedia, 2(9):3286.
- John Dowling (2007) Retina. Scholarpedia, 2(12):3487.
- Wolfram Schultz (2007) Reward. Scholarpedia, 2(3):1652.
- Wolfram Schultz (2007) Reward signals. Scholarpedia, 2(6):2184.
- Don H. Johnson (2006) Signal-to-noise ratio. Scholarpedia, 1(12):2088.
- S. Murray Sherman (2006) Thalamus. Scholarpedia, 1(9):1583.
Amygdala, Classical Conditioning, Cortex, Dopamine, Hippocampus, Models of Basal Ganglia,Motivation, Operant Conditioning, Reinforcement, Reinforcement Learning, Reward, Reward Signals, Thalamus, Up and Down States