Connectivity

Electronics Repair Manuals

Schematic Diagrams and Service Manuals

Get Instant Access

Our knowledge of the anatomical connectivity of human occipital cortex largely arises from inferences based on animal experiments. In animals, differences in connectivity within the occipital lobe provide an important key to identifying different functional subdivisions and their interrelationships. To the extent that connectivity information derived from animals accurately reflects connectivity in the human brain, it provides a valuable basis for interpreting the results of neuroimaging studies and other types of human data.

1. Visual Input to the Occipital Lobe

Figures 3A and 3B summarize the major thalamic projections to occipital visual cortex. Visual input from the retina is relayed through the lateral geniculate nucleus (LGN) of the thalamus to terminate in the primary visual cortex (striate cortex, area 17, V1). A second major pathway arises from retinal projections that bypass the LGN and project to the superior colliculus. The superior colliculus then projects to the thalamic pulvinar nucleus, which in turn distributes widely to the cortex of the occipital lobe.

The geniculocortical pathways are subdivided into three main components associated with different neuronal subclasses in the retina and LGN. One pathway originates from small P-cells in the retina and is relayed through the parvocelluar layers of the LGN to terminate in layer 4Cb of V1 and more sparsely in layers 4A and 6 (not illustrated in Fig. 3B). A second pathway to striate cortex begins with large M-type retinal ganglion cells and is relayed through the magnocellular layers of the LGN to terminate in layer 4Ca accompanied by a light projection to layer 6 (not shown in Fig. 3B). A third source of projections to V1 comes from a class of small cells within the koniocel-lular (K) or intercalated layers of the LGN. These neurons typically receive input from both the retina and the superior colliculus and terminate in the supragranular layers (above layer 4) of striate cortex. These afferents tend to cluster into "pufflike" regions of layers 2-3 that are also unique for their high levels of the metabolic enzyme cytochrome oxidase (co).

Figure 3 (A) Distribution of thalamic inputs to the occipital lobe and nearby portions of the parietal and temporal lobes. (From Zilles and Clarke, 1997.) (B) Concurrent input to V1 from K, M, and P retinogeniculate pathways with a schematic of intrinsic circuitry and cytochrome oxidase defined puffs (dashed ellipses). V2 receives segregated input from different sets of output neurons in V1 and distributes projections to different sets of extrastriate visual areas via distinct populations of output neurons in the different cytochrome oxidase defined compartments. Such circuitry forms the anatomical basis of multiple concurrent processing pathways within and among occipital visual areas. (Adapted from Casagrande and Kaas, 1994.) (C) Functional interpretation of V1 circuitry. (Adapted from Callaway, 1998.) Abbreviations: LGN, lateral geniculate nucleus; sc, superior colliculus; K, koniocellular; M, magnocellular; P, parvocellular.

Figure 3 (A) Distribution of thalamic inputs to the occipital lobe and nearby portions of the parietal and temporal lobes. (From Zilles and Clarke, 1997.) (B) Concurrent input to V1 from K, M, and P retinogeniculate pathways with a schematic of intrinsic circuitry and cytochrome oxidase defined puffs (dashed ellipses). V2 receives segregated input from different sets of output neurons in V1 and distributes projections to different sets of extrastriate visual areas via distinct populations of output neurons in the different cytochrome oxidase defined compartments. Such circuitry forms the anatomical basis of multiple concurrent processing pathways within and among occipital visual areas. (Adapted from Casagrande and Kaas, 1994.) (C) Functional interpretation of V1 circuitry. (Adapted from Callaway, 1998.) Abbreviations: LGN, lateral geniculate nucleus; sc, superior colliculus; K, koniocellular; M, magnocellular; P, parvocellular.

Afferents to striate cortex via the alternate tectopulvi-nar pathway terminate most heavily in laminae 1 and upper 2-3.

Neurons of the P, M, and K, pathways differ in their visual response properties. Neurons of the parvocel-lular path tend to be relatively more numerous in the central retina, have smaller receptive fields with sustained responses, and are often color opponent. This makes them well-suited for conveying information about fine spatial detail and color. Magnocellular neurons tend to have larger receptive fields, prefer low spatial frequencies, and have more transient responses. They do not discriminate wavelength differences well but tend to have the greatest sensitivity to luminance contrast, especially at low spatial frequencies and low luminance levels. They are optimized for processing rapid temporal changes such as flicker and movement. The function of K-pathway neurons has not been studied as extensively as that of the M and P systems, but they appear to be a major determinant of co-puff cell properties and may play a key role in the modulation of responses evoked by the M and P pathways.

Although these three pathways are distinct, their functional capabilities can overlap significantly. It is primarily at the extremes of spatial and temporal frequencies that the M- and P-cell capabilities are most different. Both M and P pathways may contribute to processing intermediate spatial and temporal frequencies, but P-cells will tend to dominate for high spatial frequencies that are slowly varying. Conversely, mag-nocellular neurons will respond better than P-cells at low spatial frequencies that are rapidly varying. Chromatically, P-cell color opponency (activation by some wavelengths and suppression by others) appears to be critical for hue discrimination, but M-cells can respond more strongly to some wavelengths than others, though they are not opponent and their tuning is typically broader than that of P-cells. The significance of this functional overlap is that, under natural viewing conditions, all three inputs to the visual cortex are likely to be concurrently active. Under extreme visual conditions, the specialized capabilities of one or another of the pathways may take over and thereby extend the range of useful sight.

2. Intrinsic Circuitry

The retinal information relayed through the LGN is processed further by the intrinsic circuitry of V1 and then distributed to other cortical areas via output neurons in layers 2-3 and 4B (see Fig. 3B). Subcortical projections to the superior colliculus and feedback projections to the LGN originate from layers 5 and 6, respectively. It is through this selective distribution of visual information combined with the characteristics of local processing that different visual areas achieve their functional uniqueness. Although a detailed review of the local circuitry of the occipital cortex is outside the focus of this article, Figs. 3B and 3C attempt to summarize the basic plan for V1. The intrinsic processing of V1 can be divided into two major levels. Input from the LGN terminates at the first level in the different subdivisions of layer 4C. Information is then distributed primarily to level 2 output neurons within layers 2-3 and 4B. The latter output signals are modified by neurons in layers 5 and 6 that combine information about the inputs and outputs of each processing level and feed it back onto neurons at the same level.

The laminar organization of V1 circuitry is complemented by horizontal connections that tend to connect functionally similar zones distributed within specific laminae. In layers 2-3, long horizontal projections preferentially interconnect pufflike zones containing cells that have similar visual response properties and that are rich in the metabolic enzyme cytochrome oxidase. Likewise, interpuff zones tend to be interconnected most strongly with other interpuff zones. In V2, a similar pattern of specific horizontal connections link co-defined compartments termed the thick-, thin-, and interstripe zones.

Together, the laminar specificity of vertical intrinsic connections and the functional specificity of horizontal connections provide an anatomical substrate for creating different subsets of output neurons whose visual response properties reflect different combinations of the P, M, and K afferent pathways, as well as modulatory effects of cortical feedback pathways and other subcortical inputs. Output neurons in layer 4B of V1 are dominated by M-pathway characteristics, as are the responses of the extrastriate visual areas, such as V2 thick stripes and area MT, to which the 4B cells project. In contrast, output neurons in the interpuff regions of V1 tend to be strongly biased by P-cell characteristics, which are subsequently imparted to downstream visual areas such as V2 (thin- and interstripe compartments), VP, and V4. Output neurons in co-puff subdivisions tend to exhibit characteristics that appear to reflect a mixture of influences. As a result, later processing stages, such as V4, also can be shown to exhibit a mixture of influences. Overall, then, the intrinsic circuitry of V1 acts to process, combine, and redistribute the visual information available in the different afferent pathways, thereby creating a new set of output signals that reflects the afferent inputs but that also encodes more complex visual properties or features. Intrinsic circuitry within subsequent cortical processing stages, though less well-studied, presumably functions in a similar manner.

3. Corticocortical Connectivity

The output neurons of layers 2-3 and 4B of V1 selectively distribute different types of visual information to subsequent processing stages in extrastriate visual cortex. Each extrastriate visual area itself has unique feedforward and feedback connections within and, often, outside the occipital lobe. These patterns of selective connectivity are a primary determinant of the functional specificity of each visual area. Finally, interhemispheric projections through the corpus cal-losum connect the left and right halves of each visual area into a functionally integrated whole.

In monkeys, connections among cortical visual areas tend to follow a consistent pattern of laminar distribution that differentiates forward, backward, and lateral types of connectivity (Fig. 4). Generally, forward connections arise from neurons in layers 2-3 of the lower area and distribute to layer 4, tapering off into the lower reaches of layer 3 in the higher visual area. In some visual areas, however, forward projections can also originate in subgranular layers (below layer 4), though the terminal distribution remains targeted at layer 4 of the recipient area. In contrast, backward-type projections tend to originate in layer 6 of the higher area and distribute outside layer 4 of the lower area, though some feedback projections can have a bilaminar origin. Finally, some visual areas have interconnections whose terminal fields engage nearly all laminae. This distribution pattern is thought to characterize lateral connections between visual areas at approximately the same processing level. On the basis of these patterns of connectivity, it is possible to arrange the various occipital visual areas into a processing hierarchy. This is illustrated in Fig. 5. The extent to which this plan applies to human visual cortex is not known, though the overall scheme is likely to be similar.2

2New techniques in neuroimaging, including the analysis of functional connectivity (correlation strengths among visual areas) and diffusion-tensor imaging, hold promise for directly charting connectivity in the human brain.

Figure 4 Schematic diagram of the laminar patterns of corticocortical connections among visual areas in macaque monkeys used to define ascending, descending, and lateral types of connectivity. The cell bodies of projection cells are found in one of two common patterns, unilaminar (left) or bilaminar (B) (right). Unilaminar projections can originate from supragranular layers (S) or infragranular layers (I). Ascending terminations tend to be focal in layer 4 (F). Lateral terminations tend to be columnar (C). Descending terminations tend to be multilayered (M). (From Felleman and Van Essen, 1991.)

Figure 4 Schematic diagram of the laminar patterns of corticocortical connections among visual areas in macaque monkeys used to define ascending, descending, and lateral types of connectivity. The cell bodies of projection cells are found in one of two common patterns, unilaminar (left) or bilaminar (B) (right). Unilaminar projections can originate from supragranular layers (S) or infragranular layers (I). Ascending terminations tend to be focal in layer 4 (F). Lateral terminations tend to be columnar (C). Descending terminations tend to be multilayered (M). (From Felleman and Van Essen, 1991.)

This connectional hierarchy incorporates several important organizational features. First, it is generally consistent with the concept that successively more complex visual response properties are represented at successive processing stages, though detailed comparisons of the response properties of efferent neurons at one stage with the properties of recipient neurons at the next stage are generally lacking. Another key aspect of this corticocortical network is that nearly every forward connection is matched by a corresponding backward connection, thereby establishing reciprocity between visual areas. Alternate connectivity schemes based on the strength of response correlation among visual areas rather than anatomical connectiv ity have stressed this reciprocity and suggest that a less hierarchical, more dynamic, processing network may also provide a useful model of occipital connectivity. A third important principle illustrated in Fig. 5 is that multiple, parallel pathways exist within, and between, each processing level. There are no hierarchical levels that are interconnected by only a single pathway. [In Fig. 5, the multiple connections between entorhinal cortex (ER) and hippocampal cortex (HC) are represented only schematically.] The arrangement of multiple parallel pathways within the cortex thereby provides a basis for concurrent processing of different aspects of the incoming visual information. Thus, the M, P, and K pathways that provide concurrent

Figure 5 Connectivity and hierarchy of vision-related areas in the macaque cerebral cortex. Visual areas at a given hierarchical (vertical) position typically receive forward projections from areas below and backward projections from areas above. Most connections between areas are reciprocal (one forward, one backward). Areas at the same hierarchical level tend to have lateral connections. Forward, backward, and lateral types of connectivity are defined in Fig. 4. Visual areas V1, V2, V3, VP, V4, and V3A are completely or partially within the macaque occipital lobe. (From Felleman and Van Essen, 1991.)

Figure 5 Connectivity and hierarchy of vision-related areas in the macaque cerebral cortex. Visual areas at a given hierarchical (vertical) position typically receive forward projections from areas below and backward projections from areas above. Most connections between areas are reciprocal (one forward, one backward). Areas at the same hierarchical level tend to have lateral connections. Forward, backward, and lateral types of connectivity are defined in Fig. 4. Visual areas V1, V2, V3, VP, V4, and V3A are completely or partially within the macaque occipital lobe. (From Felleman and Van Essen, 1991.)

processing within the retina and LGN are further modified in V1 to yield a new set of signals. These are then passed on to subsequent stages of processing via distinct sets of corticocortical output neurons. This same process is then reiterated at each successive processing stage.

Altogether, the pattern of afferent connections, the intrinsic circuitry, and the selective distribution of multiple output signals at each stage are ultimately responsible for the unique functional properties of neurons in each cortical visual area. We turn now to a more detailed account of the topography of these areas and their functions.

Was this article helpful?

0 0
All About Alzheimers

All About Alzheimers

The comprehensive new ebook All About Alzheimers puts everything into perspective. Youll gain insight and awareness into the disease. Learn how to maintain the patients emotional health. Discover tactics you can use to deal with constant life changes. Find out how counselors can help, and when they should intervene. Learn safety precautions that can protect you, your family and your loved one. All About Alzheimers will truly empower you.

Get My Free Ebook


Post a comment