| |
Structural and visuotopic organization
The anatomical limits of the pulvinar and the nuclei comprising it are not readily defined. Work on the cat and monkey, using various experimental approaches, has shown this extrageniculate structure to be parceled into several nuclei. As discussed below, these subregions' boundaries vary across studies (especially in primates), and additional work is necessary to establish a clear structural organization that relates to function adequately.
In cats, the pulvinar is referred to as the lateral posterior-pulvinar (LP-pulvinar) complex. Internal subdivisions of the LP-pulvinar have been defined by connectivity patterns (Berson and Graybiel, 1978; Updyke, 1977, 1981), cyto- and chemoarchitectonics (Berson and Graybiel, 1980, 1983; Updyke, 1983), and functional properties (Chalupa and Abramson, 1988, 1989; Chalupa et al., 1983). Of note is the histochemical work by Graybiel and Berson (1980) and Berson and Graybiel (1983), who parceled the LP-pulvinar by revealing thalamic acetylcholinesterase (AChE) activity. Several laboratories currently use this method to reveal the three main LP-pulvinar subregions that are generally agreed upon: the lateral and medial parts of the LP nucleus (LPl and LPm) and the pulvinar. Each division contains a representation of the contralateral visual field that encompasses slightly the vertical meridian, and overrepresents the binocular part of the visual field. The overall visuotopic organization of the LP-pulvinar is rather complex because of multiple visual field representations (Raczkowski and Rosenquist, 1981; Updyke, 1983); up to five have been proposed by Hutchins and Updyke (1989). These multiple representations and the large extent of receptive fields contribute to an LP-pulvinar visuotopic organization that is less rigorous than that of the LGN.
The pulvinar of primates is relatively larger than that of cats; the structural boundaries of its subdivisions are more easily differentiated, though firm delimitations using Nissl and myelin stains remain difficult. Four nuclei have been identified on classical cytoarchitectonic grounds (Olszewski, 1952; Walker, 1938): the inferior, lateral, medial, and oral (or anterior) pulvinar. An LP nucleus also exists, but it is not well developed and does not seem to receive any prominent visual input (Jones, 1985). As work progressed, it became evident that there was no clear correspondence between the cytoarchitectonically defined structural organization and the functional and connectional subdivisions of the primate pulvinar (Robinson and Cowie, 1997). In recent years, pulvinar organization has been challenged by studies using a host of chemoarchitectonic techniques. In particular, Cusik and collaborators, with the use of calbindin, AChE, cytochrome oxydase (CO), and parvalbumin stains, have divided the inferior pulvinar into four to five subregions that encompass the entire ventral part of the original lateral pulvinar (Gray et al., 1999; Gutierrez et al., 1995; in humans, Cola et al., 1999). Comparable subdivisions have been described by Stepniewska and Kaas (1997) and Adams et al. (2000) using Nissl, calbindin, myelin, CO, AChE, and Cat-301 stains. According to their proposals, however, the inferior pulvinar does not extend strongly over the lateral division. Clearly, the internal organization of the primate pulvinar remains vaguely defined. A solution to this anatomical puzzle is necessary to better understand the pulvinar's anatomo-functional organization and overall function. Nevertheless, the inferior and lateral subdivisions of the pulvinar are generally considered as the main visual nuclei. Both contain a complete topographic representation of the contralateral visual field where receptive field size increases with eccentricity (Bender, 1982; Petersen et al., 1985) and where the central visual field is overrepresented (Bender, 1981). A third area, located in the dorsomedial part of the lateral pulvinar (Pdm), has been described: it does not show a structured visuotopic organization, and the neurons comprising it have large receptive fields (Benevento and Miller, 1981; Petersen et al., 1985).
| |