The primate basal ganglia: parallel and integrative networks

Abstract

The basal ganglia and frontal cortex operate together to execute goal directed behaviors. This requires not only the execution of motor plans, but also the behaviors that lead to this execution, including emotions and motivation that drive behaviors, cognition that organizes and plans the general strategy, motor planning, and finally, the execution of that plan. The components of the frontal cortex that mediate these behaviors, are reflected in the organization, physiology, and connections between areas of frontal cortex and in their projections through basal ganglia circuits. This comprises a series of parallel pathways. However, this model does not address how information flows between circuits thereby developing new learned behaviors (or actions) from a combination of inputs from emotional, cognitive, and motor cortical areas. Recent anatomical evidence from primates demonstrates that the neuro-networks within basal ganglia pathways are in a position to move information across functional circuits. Two networks are: the striato-nigral-striatal network and the thalamo-cortical-thalamic network. Within each of these sets of connected structures, there are both reciprocal connections linking up regions associated with similar functions and non-reciprocal connections linking up regions that are associated with different cortical basal ganglia circuits. Each component of information (from limbic to motor outcome) sends both feedback connection, and also a feedforward connection, allowing the transfer of information. Information is channeled from limbic, to cognitive, to motor circuits. Action decision-making processes are thus influenced by motivation and cognitive inputs, allowing the animal to respond appropriate to environmental cues.

Introduction

The basal ganglia (BG) work in concert with cortex to orchestrate and execute planned, motivated behaviors requiring motor, cognitive, and limbic circuits. While best known for their motor functions, the BG are involved in several aspects of goal-directed behaviors, including not only its expression through the control of movement, but also the processes that lead to movement, including the elements that drive actions, such as emotions, motivation, and cognition. Indeed, regions within each of the BG nuclei are anatomically and physiologically associated with each of these functional circuits. Ventral regions of the basal ganglia play a key role in reward and reinforcement and are important in the development of addictive behaviors and habit formation (Schultz, 1997, Wise, 1998, Koob, 1999, Rolls, 2000, Everitt et al., 2001). More central basal ganglia areas are involved in cognitive functions such as procedural learning and working memory tasks (Mishkin et al., 1984, Phillips and Carr, 1987, Jueptner et al., 1997, Levy et al., 1997, Jog et al., 1999). Finally, the dorsolateral portion of the striatum, caudal to the anterior commissure is associated with the control of movement. Consistent with this topography, diseases affecting mental health, including schizophrenia, drug addiction, and obsessive compulsive disorder, are all linked to pathology in the basal ganglia, as are diseases affecting motor control (Stevens, 1973, Kalivas et al., 1993, McGuire et al., 1994, Breiter et al., 1996, Koob and Nestler, 1997, Pantelis et al., 1997, Kegeles et al., 2000, Menon et al., 2001, Rauch et al., 2001). The association of the basal ganglia with frontal cortical function along with its relationship to multiple neurological and psychiatric diseases emphasizes the importance of understanding the basal ganglia with respect to cortical function. Differentiation of frontal cortex and basal ganglia structures as they relate to human function and disease are best modeled from a combination of physiological, anatomical, and imaging studies in primates (both human and non-human). Thus, the present review describes the organization of the primate basal ganglia from the perspective of cortical function. In some situations, when necessary and indicated, data are presented from rodent work.

The BG includes the caudate n., putamen, and the globus pallidus and three closely related structures, the substantia nigra (SN), the ventral tegmental area (VTA), and the subthalamic nucleus (STh). Based on connectivity, histology, and functional considerations, the concept of the ventral striatum was introduced as the ventral extension of the striatum that includes the N. accumbens, the medial and ventral portions of the caudate n. and putamen, and the striatal cells of the olfactory tubercle (Heimer, 1978, Heimer et al., 1994). The ventral striatum contains a subregion, the shell. This region, which was first demonstrated in rodents (Zaborszky et al., 1985), is best distinguished by its lack of calbindin-positive staining (Martin et al., 1991, Meredith et al., 1996, Haber and McFarland, 1999). While the ventral and medial borders of the ventral striatum are relatively are clear, the dorsal and lateral boundaries of the ventral striatum merge imperceptibly with the dorsal striatum (Fig. 1) (Haber and McFarland, 1999). The striatum is the main input structure to the basal ganglia. Its afferent projections are derived from three major sources: (1) cerebral cortex; (2) thalamus; and (3) brainstem. The striatum projects to the pallidal complex and to the substantia nigra, pars reticulata (SNr). The pallidal complex includes the external (GPe) and internal. segments (GPi) of the globus pallidus and the ventral pallidum (VP), the pallidal segment connected to the ventral striatum. The substantia nigra (and VTA) contains the dopaminergic cells of the pars compacta (Snc/VTA), and the pars reticulata (SNr) (Fig. 2A). The outputs from the GPi and SNr are to the thalamus, which then projects back to the cortex, completing what is referred to as the ‘direct’ cortico-basal ganglia pathway. The GPe is reciprocally connected to the STh, which in turn projects to the GPi. This is referred to as the ‘indirect’ cortico-basal ganglia pathway. For a comprehensive review of basal ganglia pathways (see Percheron et al., 1994, Graybiel, 1995, Parent and Hazrati, 1995, Parent et al., 2000, Middleton and Strick, 2002, Haber, 2003).