UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The neuroanatomy of extrastriate area 19 : a modular mosaic Stewart, Tara H. 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-860639.pdf [ 19.62MB ]
Metadata
JSON: 831-1.0091380.json
JSON-LD: 831-1.0091380-ld.json
RDF/XML (Pretty): 831-1.0091380-rdf.xml
RDF/JSON: 831-1.0091380-rdf.json
Turtle: 831-1.0091380-turtle.txt
N-Triples: 831-1.0091380-rdf-ntriples.txt
Original Record: 831-1.0091380-source.json
Full Text
831-1.0091380-fulltext.txt
Citation
831-1.0091380.ris

Full Text

The Neuroanatomy of Extrastriate Area 19: A Modular Mosaic By Tara H . Stewart B.Sc , Dalhousie University, 1996 A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy In The Faculty of Graduate Studies (In the Department of Ophthalmology; the Graduate Neuroscience Program) We accept this/thesis as conforming to J(|e^5quired sj^dard The University of British Columbia UBC Rare Books and Special Collections - Thesis Authorisation Form http://www.library.ubc.ca/spcoll/thesauth.html In presenting t h i s thesis i n p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for fi n a n c i a l gain s h a l l not be allowed without my written permission. Department of ( yMmfftiafe The University of B r i t i s h Co. Vancouver, Canada Date 1 of 1 8/22/03 2:39 PM Abstract Studies of the primary visual cortex of the cat have been essential in establishing features of visual cortical organization that are fundamental to all mammalian species. One feature discovered in the cat is the presence of separate pathways that process different aspects of visual space in parallel. These parallel pathways can be defined by their organization with respect to the punctate pattern of cytochrome oxidase (CO) staining in the primary visual cortex (i.e. CO blob versus interblob). The extrastriate areas of the cat however, have not enjoyed such close scrutiny and the relationship between these areas and the parallel pathways is unknown. The aim of this study is to examine the modular organization of extrastriate area 19 (defined as a modular mosaic) in order to identify features which are shared with other mammalian species and which may be critical to overall extrastriate organization and function. Specifically, this study demonstrates that cells located in both the CO blobs and interblobs project to area 19, but they project to different regions or modules within area 19. Furthermore, it is found that intrinsic connections within area 19 form a patchy network. Two of the main recipients of projections from 19 are extrastriate areas LS and 21a. These two areas are distinguished physiologically: LS is responsive to motion, while 21a is sensitive to form (Dreher et al., 1996a). This study found that cells in area 19 that project to either 21a, or LS are organized in bands, which have spacing larger than the intrinsic patches. The bulk of the 21a and LS projection bands interdigitate, however regions that are less densely labeled tend to overlap. ii The characterization the contribution of parallel pathways to the modular mosaic in area 19 identifies several features of extrastriate cortical organization that may be shared by other mammals. First, the parallel pathways are not limited to the primary visual cortex but extend into extrastriate cortex. Second, these pathways are only partially segregated and with regions of convergence; this pattern is important, as it would allow for the efficiency of parallel processing while providing communication between the separate pathways. Table of Contents Abstract u Table of Contents iv List of Figures vii List of Tables vx Chapter 1: Introduction Pathways: Parallel and Hierarchical 1 Primary visual cortex 4 Afferents 4 Cytochrome oxidase architecture 8 Functional organization 9 Extrastriate visual cortex 11 Area 19 12 Areas Lateral Suprasylvian and 21a 17 Primate Comparison 21 Primary visual cortex: V I 21 V 2 .23 V3 26 MT/V4 27 Extrastriate Relay of Parallel Streams: Hypothesis 29 Chapter 2: Methods Animal and Surgical Procedures 34 Perfusion and Sectioning 35 Histology 36 Cytochrome Oxidase 36 Cholera Toxin 37 Horseradish Peroxidase 38 Data Analysis 38 Chapter 3: The Projection From C O Blobs and Interblobs to Area 19 Introduction 45 Methods 50 Results 52 Global Projections to Area 19 52 Patchy Projections to Area 19 53 Comparison of CO and Patches of Label 56 iv L G N Labeling 64 Discussion 65 Methodological Considerations , 65 Patchy Area 19 Efferent Cells in Primary Visual Cortex 68 Relationship of Efferent Patches to CO Staining 71 Primate Comparison 73 Conclusion 74 Chapter 4: Organization of Intrinsic Connections in Areal9 Introduction 75 Methods 80 Results 82 Description 82 Spacing 87 Cluster Analysis 87 Lateral Spread of Intrinsic Clusters 89 Discussion 91 Identification of Area 19 91 Size of Intrinsic Connections 92 Relationship to Other Modules 97 Conclusion 98 Chapter 5: Organization of Efferent Neurons in Area 19: Projecting to Area 21a Introduction 99 Methods 101 Results 103 Injection Placement 103 Tangential Organization 104 Spacing of Tangential Bands 107 Width and Mediolateral density of Tangential Bands 107 Tangential Organization of Area 18 110 Discussion 112 Tangential Organization 112 Comparison to Coronal Reports 114 Primate Comparison 114 Conclusion 115 Chapter 6: The Relationship Between 21a and LS Efferent Neurons in Area 19 Introduction 117 Methods 120 Results 122 LS Modules 122 Comparison of 21a and LS Bands 122 Clustering of LS and 21a Projection Bands 125 Spacing of LS versus 21a Projection Bands 127 Interdigitation with 21a Modules 127 Laminar Pattern 133 Discussion 134 Methodological Considerations 134 Laminar Distribution of Efferent Cells 136 Implications for Separate Parallel Pathways 137 Primate Comparison 139 Conclusion 140 Chapter 7: General Discussion Species Justification 142 Methodological Considerations I 4 4 Variability Within Area 19 1 4 7 Cortical Column Hierarchy ^ ® Parallel Pathways 1 5 2 Function of Area 19 1^4 Species Comparison 155 Future Exploration 1**2 Chapter 8: Summary Summary of findings 168 Conclusion 169 References 172 vi List of Figures Chapter 1 1.1. A schematic of the parallel pathways from the eye to area 17 6 1.2. Cytochrome oxidase staining of blobs and interblobs in area 17 of the cat 8 1.3. A schematic of the cat brain 12 1.4. The hypothesized organization of area 19 32 Chapter 2 2.1. An example of cell charts and transect measurements 40 Chapter 3 3.1 A schematic drawing of hard and soft patterned connections 47 3.2. The pattern of label in area 17 following a large injection in area 19 52 3.3. Variation in the amount of "patchiness" between different cases 54 3.4. A graph of the range of CI values 56 3.5. Patches of labeled cells are concentrated in the blobs or interblobs 58 3.6. Two dimensional spatial cross correlations 59 3.7. One dimensional spatial correlations 60 3.8. The distribution of correlation co-efficients 62 3.9. Auto-correlations of CO and labeled cell images 51 3.10 Labeled cells in the L G N following a small injection in area 19 64 3.11. A schematic of the label in the L G N 67 Chapter 4 4.1. A typical example of intrinsic labeling in area 19 83 4.2. An example of a large intrinsically labeled cluster 84 4.3. Intrinsically labeled cells in area 19 86 4.4. Isolated clusters of intrinsically labeled cells 88 4.5. A schematic of a flattened tangential section 91 4.6. A schematic of the measurements of intrinsic patches 95 Chapter 5 5.1 A tangential section of flattened cortex stained for CO 103 5.2 A typical example of 21a labeling in areal9 105 5.3 Measurement periodicity 108 5.4 Measurements of modular density and width 109 5.5 A chart of 21a projecting cells I l l vii Chapter 6 6.1. A typical pattern of labeled cells in area 19 following a LS injection 123 6.2. A chart of the LS projecting bands in area 19 124 6.3. Different spacing within one case 126 6.4 A chart of both LS and 21a projecting cells 129 6.5 A cell chart of the interdigitation and overlap of LS and 21a projections 130 6.6. The interdigitation between LS and 21a projecting cells 131 6.7. One-dimensional spatial correlations 132 6.8 A coronal slice of LS projecting cells in area 133 6.9. A schematic of a flattened tangential section 135 6.10 A schematic of the projections from V 2 to V4 140 Chapter 7 7.1 A schematic of the different proposed hierarchy 151 7.2 A schematic of parallel pathway in the primate 157 7.3 A schematic of parallel pathways in cat visual cortex 159 Chapter 8 8.1 A schematic of the parallel pathways traveling through area 19 169 8.2 A schematic of the hypothesized relationship between area 19 efferent bands and intrinsic connections 169 viii List of Tables Chapter 1 Table 1.1. Summary of the different properties of the Y , X and W pathways in the cat 4 Chapter 3 Table 3.1. Cluster Index and Periodicity Values for Each Case 55 Table 3.2. Correlation Analysis on CO arid Cell Labeling 61 Table 3.3. Spacing of the Different Anatomical Structures in Area 17 69 Chapter 4 Table 4.1. Measurements of Intrinsic Clusters for Individual Cases 89 Chapter 5 Table 5.1. Measurements from separate cases of area 19 efferent bands 106 Chapter 6 Table 6.1. A breakdown of Measurements for the different cases 125 Table 6.2. A comparison of 21a and LS measurements 134 Chapter 7 Table 7.1. A comparison between different area 19 modules 142 ix Chapter 1: General Introduction P a t h w a y s : P a r a l l e l a n d H i e r a r c h i c a l There are two divergent concepts of cortical organization, which dominate vision research: hierarchical organization and parallel pathways. A hierarchical organization requires that information is serially processed with each subsequent area doing a more complex analysis. This concept was promoted by the simple, complex and hypercomplex classification of cells developed by Hubel and Wiesel (1965), which was well suited for hierarchical interpretation. On the other hand, parallel processing requires that a single input arriving to the visual system be divided into different processing streams or pathways. The individual processing streams, each of which performs a specialized function, activate various parts of the visual system in parallel. This increases the computational power of the visual system and decreases the time required for processing. This model became popular after the discovery that functionally distinct X , Y and W cell types have different patterns of visual cortical connections (Stone et al., 1979). Current research supports a dualistic view of parallel processing that incorporates hierarchical organization with functional specialization. Thus, there is an overall hierarchy that encompasses different visual areas, but within that hierarchy there is some division of input into parallel pathways (DeYoe and Van Essen, 1988; Felleman and Van Essen, 1991). It is within this dualistic context that the hierarchical extension of parallel pathways from primary visual cortex into extrastriate cortex (specifically area 19) will be presented. The anatomical hallmark of a parallel processing stream is modular organization. A module is defined here as a discrete group of cells that is distinguished from neighbouring cells by virtue of a common element. Anatomical modules are typically distinguished on the basis of a characteristic connectivity pattern. These characteristic connections are often referred to as segregated since the afferent and efferent connections of an anatomical module are sequestered from surrounding cortex. The segregated connections to and from a module imply it is performing a specialized process or function distinct from neighboring cortex. Maps may have formed as part of the parallel pathway scheme to allow for shorter connections between neighboring neurons doing related analysis, as opposed to the more metabolically taxing long connections that would result if the neurons were not grouped (Kaas, 1997). One example of modular organization is the patchy input of the koniocelluar pathway to the CO blobs in primate V I (Casagrande, 1994). Patchy is defined here as a discontinuous labeling pattern that results when the majority of labeled cells belong to aggregates. Specificity for a single stimulus property is often manifest by a patchy pattern that varies across the surface of a visual area. While not all modules have such a well-defined relationship with known parallel pathways, optical imaging experiments have revealed modules that have functional specialization (Ghose and Ts'o, 1997; Ts'o et al., 2001). The term mosaic has been used to describe the arrangement of functionally defined modules in the primary visual cortex. This term refers to the fact that there is no basic module, which is homogenous in design but different in function. Rather the modules for each function appear to be different in structure and arranged in a complex semi-independent fashion (Swindale, 1998). For example, in area 17, singularities are 2 centered on ocular dominance bands while low spatial frequency domains tend to avoid the borders of ocular dominance columns (Hubener et al., 1997). The arrangement of heterogeneous modules is a modular mosaic. A modular mosaic includes information on both the structure of a module as well as the spatial relationship between different modules. This spatial relationship is important because it provides a basic framework or guide for future physiological exploration of the functional modular composition of an area. More importantly, the spatial relationships between modules help define the map of an area. It is believed a map is organized with specific geometric relations between different functional modules in order to optimize coverage (Hubener et al., 1997). Coverage is the degree to which all possible combinations of stimulus properties are distributed uniformly across a cortical area (Swindale et al., 2000). Segregated pathways, which are indicated by the presence of modular or patchy anatomical connections, mean that one cortical region will receive different information for processing than a neighboring cortical region. In V 2 of the primate for example, the thin, thick and pale stripes, which were first identified anatomically (Tootell et al., 1983; Livingstone and Hubel, 1984a), represent functionally differentiated modules that are preferentially tuned to colour, motion and orientation stimulus parameters respectively (DeYoe and Van Essen, 1985; Shipp and Zeki, 1985; Hubel and Livingstone, 1987). Thus, in this case the anatomical modules underlie a functional segregation that prevents total coverage (Hubener et al., 1997). Since a map of stimulus features for area 19 has yet to be determined, the anatomical modules identified in the present study, will be the first step towards defining a map of area 19. Ultimately, a necessary part of defining the map of 19 will be to work out the geometric relationships between the different modules that constitute its modular mosaic. Primary visual cortex: Afferents In the cat, visual inputs that travels from the eye to the lateral geniculate nucleus (LGN) and then to the cortex are divided into three different processing streams: the X , Y and W streams. Each of these streams or pathways is composed of X , Y or W cell types that are distinguished both morphologically and physiologically (see Table 1.1 for summary). Based upon the individual features of these streams it has been proposed that each serves a unique function; the most common supposition is that the X - pathway may be involved in spatial analysis and the Y pathway in temporal resolution. This is supported by the fact that X-cells have good spatial resolution, responses that are linear Table 1.1. Summary of the different properties of the Y , X and W pathways in the cat. Y-cells X-cells W-cells Morphology traits • Large soma • Large dendritic tree • Medium soma • Small dendritic tree • Small soma • Long sparse dendrites Physiological traits • Nonlinear summation • Fast conducting • Medium receptive fields • Receptive fields evenly distributed • Transient response • Medium spatial resolution • Linear summation • Medium conducting • Small receptive fields • Receptive fields concentrated in the area centralis • Sustained response • High spatial resolution • Slow conducting • Large receptive fields • Receptive fields evenly distributed • Sluggish response • Poor spatial resolution For review see Sherman and Spear (Sherman and Spear, 1982). 4 and sustained, and small receptive fields concentrated in the area centralis, all of which could contribute to spatial analysis. On the other hand, Y-cells prefer fast velocities and have transient responses with larger receptive fields spread fairly evenly across the retina, thus making them appropriate candidates for temporal resolution (Stone et al., 1979). It is difficult to speculate on the function of W-cells, as there is substantially less information about them. One reason for this is that their small size and sluggish response makes it difficult to record from W cells (Stone and Dreher, 1973; Rowe and Dreher, 1982). With its main retinal input traveling to the superior colliculi (Hoffmann, 1973; Fukuda and Stone, 1974), the W pathway has been linked with fixation/orientation and saccade responses (Waleszczyk et al., 1999). Although, functional significance has been placed on X , Y and W cells, it is important to keep in mind that not all retinal and geniculate cells fit into these classifications (Humphrey and Murthy, 1999) and there is still much debate as to the subtypes within the pathways (Rowe and Stone, 1977; Rowe and Stone, 1980; Friedlander et al., 1981; Kolb et al., 1981; Mastronarde, 1983; Stanford et al., 1983; Humphrey et al., 1985a; Chen et al., 1996). This uncertainty leaves open the possibility for other pathways additional to X , Y and W. Axons of the Y , X and W retinal ganglion cells travel from the eye to the L G N (Fukuda and Stone, 1974; Cleland et al., 1975; Saito, 1983). The cat L G N is divided into six layers with the top five alternate layers receiving input from the contralateral or the ipsilateral eye (Boycott and Wassle, 1974; Hickey and Guillery, 1974; Cleland et al., 1976) [See Figure 1.1 for illustration]. The bottom layer, C3 receives input from the superior colliculus (Hoffmann, 1973; Halting and Guillery, 1976; Graham, 1977; Kawamura et al., 1980; Itoh et al., 1981). Each layer contains a retinotopic map, which is 5 to contralateral hemisphere Area 17 Figure 1.1. A schematic of the parallel pathways from the eye to area 17. As the retinal ganglion axons leave the eye the nasal fibers travel to the contralateral hemisphere while the temporal fibers travel to the ipsilateral LGN. The shaded regions of the L G N represent the layers that receive ipsilateral input and unshaded regions represent layers that receive contralateral input. Note that layer C3 does not receive a retinal input like the other layers, but is instead contacted by the superior colliculus (SC). The partially segregated input of the L G N is then relayed to the primary visual cortex, including area 17. The darkened regions in area 17 represent the CO blobs, which receive the segregated W and Y input. 6 in register with the maps in other layers. Thus, a coronal electrode penetration would record from the same region in visual space as it descended through the layers (Sanderson, 1971). The top two laminae, A and A l , both receive X and Y input while the next laminae, layer C, receives W and Y input The remaining three laminae, C1-C3, receive W input exclusively (Leventhal et al., 1985). The partial segregation of the processing streams in the L G N further underlies their parallel nature. The X , Y and W cells of the L G N send axons to the primary visual cortex. In the cat, the primary visual cortex includes both areas 17 and 18, as these two areas both receive substantial thalamic input and share many of the same properties (Payne, 2002). The W cell projection from the C1-C3 layers terminate in a patchy pattern in layer 3 of the primary visual cortex (LeVay and Gilbert, 1976; Boyd and Matsubara, 1996; Kawano, 1998). Also terminating in a patchy pattern, are the Y thalamocortical projections from layer C. These Y terminals are found in layer 4a of the cortex and align with the W-cell terminals of layer 3 (Boyd and Matsubara, 1996). Since Layers 4a and 4b receive mixed X and Y terminals from layers A and A1 of the L G N , it is difficult to determine if there is any X / Y segregation occurring here. There is indication from intracellular studies that the Y fibers prefer to terminate in the upper portion of layer 4, while X fibers tend to terminate in the lower portion (Freund et al., 1985a; Humphrey et al., 1985a; Humphrey et al., 1985b). However, these studies examined only a small number of cells and also report finding considerable overlap between the X and Y input. Thus, like the L G N , the primary visual cortex likely also receives a partially segregated input. 7 Primary visual cortex: Cytochrome oxidase architecture One tool that has been very useful for examining parallel pathways is the histological stain for cytochrome oxidase (CO) (Wong-Riley, 1979). Cytochrome oxidase is located in the mitochondria and is the terminal enzyme in the electron transport chain (Wikstrom et al., 1981). As such, it is essential for the oxidative metabolism of many organs, including the brain. Staining for C O in the primary visual cortex results in a distinctive mosaic of dark patches (blobs) interspersed by lighter stained regions (interblobs) (Horton and Hubel, 1981; Horton, 1984; Murphy et al., 1995; Boyd and Matsubara, 1996). A n example of typical C O staining is shown in Figure 1.2. This differential staining suggests some underlying metabolic difference between cells in blobs versus cells in interblobs. Thalamic input is believed to be a possible contributor to the increased levels of cytochrome oxidase found in the blobs. One indication of this is that various visual deprivation paradigms, such as monocular and binocular deprivation, enucleation and impulse blockade, all result in decreased C O staining Figure 1.2. Cytochrome oxidase staining of blobs and interblobs in area 17 of the cat. in the primary visual cortex (Wong-Riley, 1979; Kageyama and Wong-Riley, 1986c; Wong-Riley et al., 1989a; Wong-Riley et al., 1989b; Murphy et al., 1995). This means 8 that maintaining normal levels of C O in the primary visual cortex depends on afferent thalamic input. Also, the W and Y cells in the C lamina of the L G N directly innervate the CO blobs, but not the CO interblobs (Boyd and Matsubara, 1996). The fact that the selective termination pattern of the C layers corresponds with the CO blobs suggests that the blobs may also represent a functional module. As the cells in the C O blobs receive a segregated input, it is possible that the visual processing in a C O blob is different from processing in an interblob. It is clear, as will be presented later, that the cells in the CO blobs have a different afferent/efferent pattern of connections than cells in the interblobs. It is this segregated input and output that has identified C O staining as a useful marker for examining parallel pathways. Primary visual cortex: Functional organization The organization of functional modules within primary visual cortex is similar, with area 18 demonstrating slightly larger periodicities. In layer 4 there are branching ocular dominance bands, in which alternating bands receive input from one eye and then the other. These bands have been demonstrated both anatomically (LeVay and Gilbert, 1976; Shatz et al., 1977; Shatz and Stryker, 1978; Anderson et al., 1988) and functionally (Tieman and Tumosa, 1983; Crair et al., 1997; Muller et al., 2000). In addition to the ocular dominance map, various stimulus features are also mapped onto the primary visual cortex. One of the first features discovered to have an orderly map was orientation selectivity. Cells that prefer similar orientation are clustered together (Hubel and Wiesel, 1962; Hubel and Wiesel, 1963) and these clusters are grouped together to form a pinwheel structure (Bonhoeffer and Grinvald, 1991). Clusters representing all orientation 9 preferences meet at the center of the pinwheel, which is called a singularity and has broad orientation tuning. There is a relationship between orientation maps and ocular dominance bands, as pinwheel structures tend to be centered over ocular dominance bands and iso-orientation lines tend to cross the bands at right angles (Crair et al., 1997; Hubener et al., 1997; Muller et al., 2000) forming a modular mosaic. Direction preference maps are inherently linked with the orientation maps (as one component of direction is orientation) and demonstrate similar patterns of organization but on a smaller scale (Shmuel and Grinvald, 1996). There is also a map for spatial frequency with bands of cells preferring low spatial frequency and high velocities interdigitating with bands of cells preferring high spatial frequency and low velocities. In fact, the low spatial frequency bands that prefer high velocities coincide with CO blobs, thus demonstrating a functional relationship in the cat between CO staining and spatial frequency tuning (Shoham et al., 1997). Another modular feature is the patchy intrinsic connections of the primary visual cortex (Gilbert and Wiesel, 1983; Luhmann et al., 1986; Boyd and Matsubara, 1991; Luhmann et al., 1991; Kisvarday and Eysel, 1992; Kisvarday et al., 1997). These patchy intrinsic connections have a relationship with the orientation map. The excitatory intrinsic connections that originate from an iso-orientation domain terminate mainly in regions of cortex that have similar orientation preference (Michalski et al., 1983; Nelson and Frost, 1985; Ts'o et al., 1986; Gilbert and Wiesel, 1989, Kisvarday et al., 1997). Not all studies are in agreement with this finding (Matsubara et al., 1985,1987), but this may have more to do with the different methods employed in the various studies. 10 Extrastriate visual cortex It is established that anatomy and function are both modularly organized in the primary visual cortex. Oh the other hand, extrastriate visual cortex has not been studied in such detail and its organization remains largely unknown. In order to determine the fundamental cortical structure of the cat visual system (parallelism versus hierarchy), it is important to rectify this gap in knowledge of extrastriate organization. The next logical area to examine is area 19, since it borders the primary visual cortex and is believed to be the next processing center in the visual hierarchy (Sherk, 1986b; Burke et al., 1998; Scannell et al., 1999). Ascertaining the connections of area 19 is important in revealing the functional subdivisions within the extrastriate visual cortex, since the function of an area is believed to be dependent upon the information which that area receives (Sherk, 1986b). And on a broader scale, tracing the pathways through extrastriate cortex and comparing cat cortical organization with primate organization will aid in determining and defining the common fundamental features of the visual system. Fifteen different visual areas have been identified in the cat and are illustrated in Figure 1.3. Most of these areas were initially defined by a series of exhaustive retinotopic studies (Palmer et al., 1978; Tusa et al., 1978; Tusa et al., 1979; Tusa and Palmer, 1980). Experiments using anterograde and retrograde tracers later verified the unique pattern of connections for each visual area (Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b). To date, most research on extrastriate cortex is restricted to three areas: area 19, area 21a and a portion of the lateral suprasylvian area (LS) called the posteromedial lateral suprasylvian area (PMLS). This thesis will focus on these three areas, particularly area 19. l l Figure 1.3. A schematic of the cat brain. The visible visual areas are labeled and area boundaries are marked by dotted lines. Areas found in the lateral sulcus (lat) or the suprasylvian sulcus (SS) are indicated by arrows. The suprasylvian genu is located just below area 21a where the suprasylvian sulcus makes a 90° turn. A M L S : Anterior Medial Lateral Suprasylvian area. A L L S : Anterior Lateral Lateral Suprasylvian area. PMLS: Posterior Medial Lateral Suprasylvian area. PLLS: Posterior Lateral Lateral Suprasylvian area. V L S : Ventral Lateral Suprasylvian area. DLS: Dorsal Lateral Suprasylvian area. The area locations are based on (Tusa et al., 1978; Palmer et al., 1978; Tusa et al., 1979; Tusa and Palmer, 1980). Extrastriate visual cortex: Area 19 Area 19 lies adjacent to the lateral border of area 18 and the retinotopy of the two areas closely mirror each other with an emphasis on the lower half of the visual field (Tusa et al., 1979). Although there is intra-animal variation in the border between areas 18 and 19 (Donaldson and Whitteridge, 1977; Tusa et al., 1979; Albus and Beckmann, 12 1980; Raczkowski and Rosenquist, 1983), it is visibly defined by a decrease in CO staining in all laminae of area 19 (Price, 1985a). Despite the retinotopic similarities between the two areas, area 19 is not considered part of the primary visual cortex (Payne, 2002) and any deviations from the pattern of organization established in the primary visual cortex may provide insight into the functional role of area 19. Behavioral experiments on lesioned animals have lead to multiple suggestions of the function of area 19. This area is implicated in form discrimination (Doty, 1971; Sprague et al., 1977; Hughes and Sprague, 1986; Kruger et al., 1988) and discrimination of partially obscured figures (Cornwell et al., 1980a; Cornwell et al., 1980b). However, the lesion site in these experiments included areas other than area 19 so it is difficult to attribute a specific perceptual loss to just area 19. One study with a lesion restricted to area 19 found deficits in the detection of slow and fast moving patterns but not in medium moving or stationary patterns (Dinse and Kruger, 1990). Based upon these findings area 19 may be involved in both form and motion processing. The physiology data also support the notion that area 19 is involved in form and motion analysis. However, electrophysiological study of area 19 is limited because of the difficulty in getting area 19 to respond under anaestsia (Feldon et al., 1978; Saito et al., 1988; Sherk, 1990). It is worth noting that the receptive field properties (e.g. orientation tuning, end-stopping, direction selectivity) obtained from area 19 has a greater range than those recorded from extrastriate areas 21a and PMLS, two higher-order visual areas (Mizobe et al., 1988). There is a consensus that the majority of cells in area 19 are binocular and disparity tuned (Leventhal and Hirsch, 1983; Tieman and Tumosa, 1983; Ptito et al., 1991; Guillemot et al., 1993a; Guillemot et al., 1993b; Wang and Dreher, 13 1996) with a preference for bars of a specific length (Hubel and Wiesel, 1965; Kimura et al., 1980; Duysens et al., 1982a; Duysens et al., 1982b; Tanaka et al., 1987; Mizobe et al., 1988; Saito et al., 1988; Toyama et al., 1994). This preference for bar length is called end-stopping. It has been suggested that areal9 is selective for stereoscopically divergent stimuli, while area 17 prefers stereoscopically convergent disparities (Pettigrew and Dreher, 1987). However, disparity specialization divided between two areas is unlikely since area 19 has broader disparity tuning than area 17 and probably plays a minor role in disparity processing (Guillemot et al., 1993a; Guillemot et al., 1993b). The orientation, direction, spatial frequency, and temporal tuning of cells in area 19 are largely disputed (Dreher et al., 1980; Duysens et al., 1982a; Duysens et al., 1982b; Tanaka et al., 1987; Mizobe et al., 1988; Guillemot et al., 1993a; Guillemot et al., 1993b; Toyama et al., 1994; Bergeron et al., 1998). Nonetheless, response properties of cells in area 19 are similar to those found in areas 17 and 18, suggesting that area 19 is performing a redundant or parallel analysis (Kimura et al., 1980; Duysens et al., 1982b; Bergeron et al., 1998). The discrepancy in the physiological data may arise from the fact that there are different populations of cells in area 19. Toyama et al. (1994) defined two separate response profiles in area 19: 1) cells that display strong end-stopping and moderately selective orientation and direction tuning. 2) cells which are weakly end-stopped with highly selective orientation tuning. They concluded that there are two populations of cells acting in area 19, one that codes for position and motion of discontinuous features and another that codes continuous elements. Saito et al. (1988) also reported two different populations of cells in area 19, one that responds to spots of light (dot 14 responsive) and a second population mat prefers long lines (elongation-requiring). Dot responsive and elongation-requiring cells are located in columnar patches in area 19, implying a functional segregation. Similar to Toyama et al.'s classification, one cell type (the dot responsive) is well suited to processing terminations in contours and the other cell type (the elongation-requiring) is suited for processing orientation of continuous elements. The patchy organization of these functional cell types raises the question of whether the anatomy in area 19 is also organized in patches. The predominant projection of fibers from the primary visual cortex is to area 19 (Bullier et al., 1984b; Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b; Mulligan and Sherk, 1993). Although this projection is topographic (Heath and Jones, 1970; Montero, 1981), studies using small deposits of anterograde and retrograde tracers determined that discrete patches of cortex in area 17 project to patches of cortex in area 19 (Gilbert and Kelly, 1975; Bullier et al., 1984b; Symonds and Rosenquist, 1984a; Price and Blakemore, 1985a; Price and Blakemore, 1985b; Mulligan and Sherk, 1993). Even with large deposits of tracer, area 17 efferents to area 19 are reported to be clustered in patches (Ferrer et al., 1992). However, the spacing of these deposits was 1 mm apart and may have missed modules in area 19 that occur on a smaller scale between the 1 mm deposits. Area 19 also demonstrates patchy callosal organization such that the callosal projecting cells and callosal terminals overlap in patches along the area 19/21a border (Segraves and Rosenquist, 1982a). The majority of area 19 projections are to areas 7,17, 18,20a, 21a, and P M L S (Heath and Jones, 1970; Squatrito et al., 1981; Symonds and Rosenquist, 1984b; Price, 1985b). The results from experiments looking at small injections (Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b) as well as 15 large injections (Boyd and Matsubara, 1999) report clustering of efferent cells in area 19. Thus, it would seem that the organization of afferent, callosal and efferent connections in area 19 is modular. Area 19 also receives direct thalamic input. The majority of this input originates from the C layers of the L G N (layers A and A l only project to the primary visual cortex) (Maciewicz, 1975; Hollander and Vanegas, 1977; Raczkowski and Rosenquist, 1980; Raczkowski and Rosenquist, 1983) and terminates in patches in area 19 (Kawano, 1998). Patches of label in area 19 are also visible following an ocular injection of anterograde tracer (Anderson et al., 1988). As area 19 does not have ocular dominance columns (Tieman and Tumosa, 1983), this patchy labeling is likely the result of patchy input from the L G N . Area 19 also receives projections from subcortical structures, such as the lateral posterior-pulvinar complex, superior colliculus and the medial interlaminar nucleus (MIN) (Graybiel, 1972; Updyke, 1977; Niimi et al., 1981; Symonds et al., 1981; Raczkowski and Rosenquist, 1983; Bullier et al., 1984b; Harting et al., 1992). The input from these structures is not included in Figure 1.1, as there is not enough information on the nature of their connections to place them in the parallel pathway scheme. There is evidence that W and Y cells relay through these subcortical structures (Cleland et al., 1975; Hoffmann and Schoppmann, 1975; Wilson and Stone, 1975; Cleland et al., 1976; Wilson et al., 1976; Dreher and Sefton, 1979; Rowe and Dreher, 1982; Crabtree et al., 1986; Berson, 1987), providing W and to a lesser extent Y input to area 19 (Dreher et al., 1980). 16 Extrastriate visual cortex: Areas LS and 21a The cortex surrounding the suprasylvian sulcus is first described by Marshall et al. (1943) and Clare and Bishop (1954) as a visually responsive cortical region. It has since been divided up into multiple visual areas (see Figure 1.3) based upon exhaustive retinotopic studies (Palmer et al., 1978; Tusa and Palmer, 1980). The region of cortex immediately bordering the suprasylvian sulcus is divided into four groups, P M L S , PLCS, A M L S and A L L S . This is the classic division of the lateral suprasylvian area (LS) and is the classification system used by the majority of researchers to date. However, the retinotopic organization in this region is complex with substantial local and between animal variations (Grant and Shipp, 1991; Mulligan and Sherk, 1993). At the time of the retinotopic studies Palmer et al. stipulated "...meaningful division of the cortex based on retinotopy alone must remain speculative until they [sic] can be verified by other means." This classification of LS was later challenged based upon new retinotopic studies and the finding that a large part of the suprasylvian region shared similar connections. A new area was officially defined as LS and included portions of 21a, PLLS, A M L S and all of PMLS (Sherk, 1986b; Grant and Shipp, 1991). More recently, on the basis of parametric cluster analysis of connectional data, it is suggested that perhaps LS ought to once again be divided up into the different areas 21a, PMLS, PLLS, A M L S (Hilgetag and Grant, 2000). As there is still no collective agreement as to the border or definition of these areas, this study will use the designation of LS as defined by the more detailed experiments that included both physiological and anatomical data (Sherk, 1986b; Grant and Shipp, 1991). Although comparing studies using different area classifications can get 17 confusing, PMLS lies within in LS and therefore findings from studies of PMLS are applicable to the LS designation. Much of the physiological data supports the notion that LS is a major contributor to the "high-order motion stream'' (Dreher, 1986; Spear, 1991). Over all, cells in LS are end-stopped with large receptive fields, contain a contralateral bias, and have good direction selectivity but poor orientation tuning (Hubel and Wiesel, 1969; Spear and Baumann, 1975; Smith and Spear, 1979; Y i n and Greenwood, 1992; Toyama et al., 1994; Dreher et al., 1996c). A preference for visual noise and motion are also pronounced in LS cells (Yin and Greenwood, 1992; Toyama et al., 1994; Dreher et al., 1996c; Mulligan et al., 1997; Akase et al., 1998; Merabet et al., 2000; L i et al., 2001). Removal of LS does not affect the ability to discriminate stationary objects but does affect velocity and direction discriminations, as well as structure-from-motion and integration of local motion cues (Sprague et al., 1977; Spear et al., 1983; Pasternak et al., 1989; Rudolph and Pasternak, 1996). Cooling of LS results in deficits in motion, spatial and attention tasks (Lomber, 2001). More recently, studies have proposed that LS plays a significant role in optic flow processing (Sherk et al., 1995; Kim et al., 1997; Mulligan et al., 1997; L i et al., 2000; Brosseau-Lachaine et al., 2001; L i et al., 2001). Various functional properties appear to be segregated within LS. Microstimulation of spatially distinct regions within LS result in either lens accommodation (Bando et al., 1981; Maekawa and Ohtsuka, 1993; Bando et al., 1996), pupillary responses (Bando et al., 1988) or convergent eye movements (Takagi et al., 1992; Toda et al., 2001). Also, there are reports of an anterior/posterior dichotomy in optic flow responses in LS. Sherk et al. (1995) found that the posterior portion of LS 18 responds to centrifugal (optic flow) directions while the anterior portion of LS responds to directions that are orthogonal to the optic flow. On the other hand, Akase et al. (1998) found that the posterior portion of P M L S is sensitive to receding or frontoparallel motion while the anterior portion of PMLS prefers approaching stimuli. Another group also found anterior/posterior differences in optic flow responses in P M L S but attributed this to the fact that the area centralis is located in the posterior portion of this area while the peripheral representation is located more anteriorly (Brosseau-Lachaine et al., 2001). Another pattern of functional organization in LS is proposed for orientation selectivity. Cells with similar orientation are grouped together (Hubel and Wiesel, 1969), while 2-deoxyglucose (2DG) experiments show only discrete portions of LS responding to orientation tasks (Vanduffel et al., 1997). Also, when the LS cortex is silenced with kinaic acid the input from other areas can be recorded in LS. Moving the recording electrode through LS in an orderly manner reveals a systematic shift of preferred orientation and direction for LS afferents travelling from area 17. This means that there is a slight shift in preferred orientation for afferents inputting into two neighbouring regions (Sherk, 1990). Thus, there is some evidence of modular organization of the functional aspects of LS. The majority of input into LS originates from areas 17,18 and 19 (Kawamura et al„ 1974; Sherk, 1986a; Lowenstein and Somogyi, 1991; Maekawa and Ohtsuka, 1993; Norita et al., 19%). The anatomical inputs to LS from the primary visual cortex originate from cells located in the CO blobs (Boyd and Matsubara, 1999) and terminate in patches in LS (Sherk, 1986a; Grant and Shipp, 1991; Shipp and Grant, 1991). Similar to area 19, LS also has patchy callosal innervation, in which callosal terminals overlap with 19 contralateral projecting cells (Segraves and Rosenquist, 1982), and patchy L G N input (Sherk, 1986a; Sherk, 1986b). The posterior-pulvinar complex (Raczkowski and Rosenquist, 1980; Tong et al., 1982; Raczkowski and Rosenquist, 1983; Kato et al., 1984; Sherk, 1986a) and superior colliculus (Segal and Beckstead, 1984) also project to LS. The functional specificity of this area combined with the selective modular afferent input makes LS a candidate to part of a single segregated pathway. Despite their juxtaposition, there exists both physiological and anatomical data supporting the idea that 21a and LS are two functionally separate areas. Area 21a contains a representation of only the upper central 20 degrees of visual field (Tusa and Palmer, 1980). Unlike LS, area 21a cells have an ipsilateral bias and are highly orientation selective but not direction selective (Dreher et al., 1993; Toyama et al., 1994; Dreher et al., 1996c). Despite its selectivity for orientation, 2DG and lesion studies do not suggest that area 21a is involved in discrimination of orientation (Sprague, 1996; Vanduffel et al., 1997). On the other hand, it is form recognition and not motion tasks that are affected when area 21a is inactivated through cooling (Lomber, 2001). Although areas 21a and LS receive input from the same thalamic nuclei, as well as the same cortical layers in area 17, this input originates from two non-overlapping populations of cells (Dreher et al., 1996a). For example, in area 17 both 21a and LS projecting cells are found in the CO blobs. However, there are very few cells in the CO blobs that project to both 21a and LS (Conway et al., 2000). Thus, the source of 21a inputs are separate from the source of L S inputs and is involved in a different function than LS , which is likely related to form analysis 20 Primate Comparison The primate is also a frontal-eyed mammal whose visual system has been extensively studied. The primate line diverged from the carnivore line (which includes cats) more than 65 million years ago and each developed different ecological niches (Young, 1981). The macaque Old World monkey is generally the preferred primate for visual studies but there have also been studies that have looked at the visual system of New World monkeys, tarsiers and prosimians (e.g. galagos). Despite the divergence of the carnivore and primate lines, these two orders share many visual characteristics. Primary Visual Cortex: VI Like the cat, cells in the L G N of the primate are also classed into three distinct groups: the magnocellular (M), parvocellular (P) and koniocellular (K) cells. Different retinal ganglion cell types terminate in different layers of the L G N , which in the primate consists of six layers. The ventral two layers receive an M input while the four dorsal layers receive a P input (Leventhal et al., 1981; Perry et al., 1984). The K input originates from cells found in the intercalated layers, in between the M and P layers. The dorsal two K layers relays low-acuity information, while the middle two K layers relay short-wavelength input and the ventral two K layers are related to the superior colliculus function (Hendry and Reid, 2000). In terms of function and morphology, the primate and cat pathways can be compared such that M pathway is Y-like, while the P pathway is X -like and the K pathway is similar to the W pathway (Norton and Casagrande, 1982; Irvin et al., 1986; Norton et al., 1988; Casagrande, 1994; Hendry and Reid, 2000). However, this comparison remains controversial as there are functional differences between the 21 primate and cat parallel pathways (Dreher et al., 1976; Sherman et al., 1976; Benardete et al., 1992). Like areas 17 and 18 of the cat, the first visual area (VI) in primates also demonstrates CO blob and interblob staining. While it has been suggested that CO blobs are specialized for color processing (Livingstone and Hubel, 1984a) this is contradicted by the fact that even nocturnal species with a single cone have CO blobs (Casagrande, 1994; Ding and Casagrande, 1998; Shostak et al., 2002). In addition to this, a subsequent study found no link between CO staining and the degree of orientation, direction or color selectivity in V I (Leventhal et al., 1995). Like the cat, K (W-like) cells input into C O blobs in layer 3, while the P and M pathways terminate in layer 4C8 and 4Ca respectively (Hubel and Wiesel, 1972; Wiesel et al., 1974; Hendrickson et al., 1978; Livingstone and Hubel, 1982; Blasdel and Lund, 1983; Freund et al., 1989; Lachica et al., 1992; Ding and Casagrande, 1997). Despite this segregation of L G N input there is still some convergence of the pathways as exemplified by the dendritic overlap of M and P receiving neurons inthe mid part of layer 4 and input into layer 3 blobs and interblobs from all layers but layer 1 (Lund and Yoshioka, 1991; Peters and Sethares, 1991; Lachica et al., 1992; Lachica et al., 1993; Levitt et al., 1996; Boyd et al., 2000). Layer 4C is further divided into ocular dominance bands with input from the left and right eyes innervating alternate bands (Levay et al., 1975; Florence and Kaas, 1992). There are also clusters of cells in layer 2/3 that prefer similar orientation and are arranged in a pinwheel fashion. Unlike the cat, both orientation pinwheels and CO blobs are centered over ocular dominance columns. However the centers of orientation pinwheels do not coincide with the center of C O blobs and the two modules are spatially independent of 22 each other (Bartfeld and Grinvald, 1992). Thus, while primate V I demonstrates many similarities to the cat primary visual cortex, both anatomically and functionally the modular mosaic of these two areas differ. V2 V I is the only visual area in the primate to receive a substantial L G N input. Thus, area V 2 is considered the first extrastriate area in the primate visual system. However, it is worthwhile noting that various extrastriate areas, including V 2 have been shown to receive a minor geniculate input, which may originate from the K cells (Benevento and Yoshida, 1981; Fries, 1981; Yukie and Iwai, 1981; Bullier and Kennedy, 1983; Kennedy and Bullier, 1985; Lysakowski et al., 1988; Tanaka et al., 1990; Hendry and Yoshioka, 1994; Hernandez-Gonzalez et al., 1994). Nevertheless, inactivating V I has strong depressive effects on the functioning of V 2 (Schiller and Malpeli, 1977; Girard and Bullier, 1989). This suggests that unlike area 19, which sustains its performance even after primary visual cortical lesion (Dinse and Kruger, 1990), primate V 2 cannot function on thalamic input alone and relies heavily on V I . Extrastriate V 2 has a very distinct pattern of CO staining with thick and thin stripes of dense CO staining interspersed by pale stripes of CO staining (Tootell et al., 1983; Livingstone and Hubel, 1984a). These C O stripes have traditionally been viewed as separate functional compartments and at close inspection appear as a band of fused CO patches (Carroll and Wong-Riley, 1984; Cusick and Kaas, 1988; Tootell and Hamilton, 1989; Krubitzer and Kaas, 1990). The patchy intrinsic connections in V 2 also support the notion of a patchy substructure within the bands (Malach et al., 1994). Analysis of 23 intrinsic connections reveal cytochrome oxidase rich stripes (thin and thick) preferentially connect with other C O rich bands while the connections emanating from the pale stripes are not preferentially concentrated in either the pale, thin or thick stripes (Levitt et al., 1994a). These CO stripes also mark the location of thalamic connections, as the majority of cells in V 2 that project to the superior colliculus are located in the thick bands (Abel et al., 1997) and input from the pulvinar terminates in thick and thin stripes (Levitt et al., 1995). This pattern of CO staining in V 2 is not consistent for all primates. Prosimians are a suborder of primates that are separate from the human containing simian suborder. Although prosimian primates have CO bands that are either poorly developed or are not apparent (Condo and Casagrande, 1990; Krubitzer and Kaas, 1990; Preuss et al., 1993; Rosa et al., 1997) there still exist connectionally distinct bands of cells in V2 , as in simian primates (Collins et al., 2001). The projection from V I to the V 2 CO stripes is also fairly specific. The thick stripes receive input from layer 4B and are associated with the M stream while the thin stripes receive input from cells located in CO blobs and thus are believed to receive information from the K system. The pale stripes are innervated from cells in the interblobs and are believed to constitute part of the P pathway (Livingstone and Hubel, 1984; Livingstone and Hubel, 1987). Although historically, much research has centered on this segregated input from V I into V2 , this scheme has recently been challenged by Sincich and Horton (2002a). They report that the different stripes receive projections from the same layers of V I . Specifically, the thick and pale stripes both receive input from C O interblob columns extending from layer 3 to layer 4b with about one third of cell projecting to both the thick and pale stripes. The thin stripes receive input from the 24 blob columns and do overlap with the interblob column projection (for illustration see Figure 7.1). If these new findings by Sincich and Horton are indeed accurate, then the model of parallel pathways in the primate will have to be rethought with an emphasis on convergence into two pathways instead of the segregation of three pathways. It is important to point out that there exist numerous projections between the different CO stripes and hence between the different pathways (Levitt et al., 1994a). A convergence of pathways is further suggested by single cell recordings that show individual cells respond to a variety of stimuli associated with the different pathways. Thus, cells in a particular stripe may be selective for both velocity (M-associated) and chromatic (P-associated) stimuli. However, pathway convergence is less visible on a larger scale such that within a CO stripe there is an overall bias towards a single functional type of stimuli (e.g. color, form, motion) (Levitt et al., 1994b). This segregation of function according to different C O stripes is even more apparent in optical imaging data. Patches of V 2 that respond to color and luminance are found in the thin stripes while the thick stripes have clusters of cells that respond to retinal disparity (Roe and Ts'o, 1999). Clusters of broadly tuned cells are found in thin stripes and clusters of highly tuned orientation cells are found in thick stripes (Ts'o et al., 1990; Malach et al., 1994; Roe and Ts'o, 1995). The clusters seen in the optical imaging may represent the patchy substructure suggested by the C O staining. The optical imaging data supports the idea of tripartite organization of function in V2, with the thick stripes processing motion and disparity, the thin stripes processing color and the pale stripes processing orientation (DeYoe and Van Essen, 1985; Shipp and Zeki, 1985; Hubel and Livingstone, 1987). So the C O stripes in V 2 represent both a functional and anatomical modular mosaic. 25 However, ocular dominance is one functional segregation that is absent in V 2 (Friedman et al., 1989; Tootell and Hamilton, 1989; Ts'o et al., 1990). Absence of ocular dominance bands is characteristic of extrastriate areas in both the primate and cat. V3 Initially V3 of the primate was called area 19 and was studied with the aim of drawing similarities between primate and cat area 19 (Cragg, 1969; Zeki, 1969). However, it soon became apparent that the area that was originally viewed as V3 or area 19 in the primate, may in fact be parceled into several different visual areas (i.e. V3d, V3v, V P and DM). The division of V 3 remains highly contentions and if in fact is accurate the subsequent areas would be a substantial departure from cat area 19 (for review see Kaas and Lyon, 2001). It is difficult to draw conclusions as to the function of such an area(s) when its existence is in dispute. Consequently not much research has focused on this area. The physiological studies that have looked at this region have concluded that V3 has strong binocular interactions (Zeki, 1978a; Zeki, 1978b) and a preponderance of orientation and disparity tuned cells that are organized in columns (Zeki, 1978b; Adams and Zeki, 2001). Cells in V 3 are also direction selective, color selective and end-stopped. This suggests that both M and P input is processed in this area (Felleman et al., 1987; Gegenfurtner et al., 1997). It would be worthwhile to determine if V3 received input from all of V 2 or if only a specific CO stripe(s) type projected to it. It is known that V3 projects to both V4 and M T (Zeki and Shipp, 1988; Felleman et al., 1997). Based on the above findings it is thought that V3 may be processing dynamic form (Zeki and Shipp, 1988; Gegenfurtner et al., 1997). Despite the dispute over the 26 division of V3 and the limited research on this area, V3 appears to shows some physiological similarities with cat area 19. MT/V4 Primate areas M T and V 4 are two extrastriate areas that share many similarities to cat areas LS and 21a respectively. These similarities have led to M T and LS, as well as V 4 and 21a, being labeled as homologous areas (Payne, 1993). However, the issue of homology will only be answered with a cladistic study of visual systems that encompasses more species than just primates and cats (Kaas and Lyon, 2001). Hence, for propriety sake these areas will be treated as analogous, but this does not preclude the possibility of a true homology. The medial temporal areas (MT — also known as V5) is strongly implicated in the analysis of motion. Optical imaging reveals distinct maps for both direction and orientation selectivity in M T (Malonek et al., 1994). Similar to LS, direction selectivity in M T varies for bars versus textures (Albright, 1992; Mulligan et al., 1997). Also, both LS and M T neurons are sensitive to pattern motion (Rodman and Albright, 1989; Stoner and Albright, 1992; Albright and Stoner, 1995; L i et al., 2001). Damage to M T impairs direction discrimination and as well as the ability to detect and pursue moving stimuli (Newsome et al., 1985; Newsome and Pare, 1988). This too is similar to the impaired performance on detection and discrimination motion tasks following LS lesioning or cooling. Based on its physiological features it has been put forward that M T might be involved in pursuit eye movements and analysis of optic flow (Lennie, 1998), two functions which are also proposed for LS in the cat. 27 The main anatomical input to M T is well characterized. Cells in V I that project to M T are preferentially located below CO blobs (Boyd and Casagrande, 1999). In the cat, LS projecting cells are found in the CO blobs. Projections from V I terminate in patches within M T (Montero, 1980). The termination pattern of area 17 projections to LS is also clustered in the cat. And M T receives a major input from the thick stripes in V2. The primary source of input to M T is believed to originate in the M pathway (DeYoe and Van Essen, 1985; Shipp and Zeki, 1985). In turn, M T provides the major input into the medial superior temporal area (MST) (Maunsell and van Essen, 1983; Boussaoud et al., 1990), which is an area also strongly implicated in optic flow analysis (Saito et al., 1986; Sakata et al., 1986; Tanaka et al., 1989; Tanaka and Saito, 1989; Duffy and Wurtz, 1991; Orban et al., 1992; Graziano et al., 1994; Lagae et al., 1994). The majority of input that leaves M T travels to other parietal areas that comprise the dorsal pathway. This pathway is believed to be responsible for identifying where a particular object is and in what directions it is moving (Ungerleider and Mishkin, 1982). Area V4 , on the other hand, is an area that has been implicated in form processing. Initial findings on V 4 stressed the chromatic responses of cells (Zeki, 1973). But subsequent studies show that the chromatic responses found in V 4 are similar to V I (Kruger and Gouras, 1980; Schein et al., 1982) and lesions of this area only slightly impair color discrimination (Gross et al., 1971; Hey wood and Cowey, 1987; Heywood and Cowey, 1992; Kulikowski et al., 1994). Instead, V 4 appears to be involved in form discrimination, like cat area 21a. Recently, experimenters have preferentially focused on the response of V 4 to contour features and complex shapes (Gallant et al., 1993; Kobatake and Tanaka, 1994; Gallant et al., 19%; Pasupathy and Connor, 1999). Within 28 V 4 there are reports of clustering of chromatic, luminance (Tanaka et al., 1986; Ghose et al., 1994a; Ghose et al., 1994b; Ghose et al., 1995), spatial frequency (DeYoe et al., 1992) and Cartesian versus non-Cartesian responses (Gallant et al., 1993). Also, optical imaging has revealed patches cortex containing cells with similar orientation selectivity and patches that are selective for small stimuli (Ghose and Ts'o, 1997). Hence, the physiology in V 4 is also clustered in a modular mosaic. Cells within the thin and pale stripes in V 2 project to V 4 (DeYoe and Van Essen, 1988; Shipp and Zeki, 1989; Felleman et al., 1997). This projection from V 2 to V4 shows both segregation and convergence of inputs from the thin and inters stripes (Xiao et al., 1999). Efferent projections from V 4 to PITv are also organized in a patchy fashion (Felleman et al., 1997). Despite the functional specialization between V 4 and M T it is important to note that these areas are not acting in isolation and there are connections between them. However, the majority of V 4 efferents project to the areas in the inferotemporal cortex (Van Essen et al., 1990; Felleman and Van Essen, 1991) that constitutes the ventral pathway. This pathway is commonly believed to be involved in the identification of objects (Ungerleider and Mishkin, 1982). E x t r a s t r i a t e R e l a y o f P a r a l l e l S t r e a m s : A n H y p o t h e s i s Visual cortical research has traditionally been fueled by the desire to understand human cortical organization. This is one reason that the primate is a popular choice as an animal model. However, the amount of information that can be gather from this approach is limited. Only by examining multiple species is it possible to make between species comparisons. In V I , cross-species studies have determined which characteristics 29 have been conserved between species, have lead to interpretations of V I function and have resulted in predictions for V I organization in less studied mammals (Rosa and Krubitzer, 1999; Preuss, 2000). For example, the function of C O blobs was initially believed to be related to colour analysis (Livingstone and Hubel, 1988). However, the discovery of CO blobs in multiple species lacking acute colour vision (Horton, 1984; Casagrande, 1994; Murphy et al., 1995; Preuss and Kaas, 1996; Ding and Casagrande, 1998; Shostak et al., 2002) has seriously challenged this belief and initiated a new search to identify the function of CO blobs. The next step in this process is to amass information on extrastriate organization from multiple species. This information will make it possible to eventually: 1) determine which areas may have existed in a common ancestor, 2) identify clusters of similarly structured areas that may have evolved from a single ancestral area, 3) distinguish distinctive specialized areas that evolved independently in different species, and 4) reconstruct the visual cortex of ancestors. Therefore, studying species from mammalian orders other than the primates (e.g. the cat), provides more information and ultimately allows for a broader approach that is not limited to a comparison to the human. The purpose of this thesis is to determine if the anatomical organization underlying parallel pathways exists in area 19 of the cat (the first extrastriate area of the visual hierarchy) in the form of a modular mosaic. Features of the modular mosaic in area 19 that are shared with extrastriate areas in other species may represent fundamental building blocks of visual extrastriate cortex. In order to examine this possibility, three specific hypotheses were generated to test the organization of pathways traveling to area 30 19, within area 19, and from area 19 to other specialized areas. The hypotheses are as follows: 1. Afferents from CO blob and interblob pathways in area 17 remain segregated within area 19 (illustrated in Figure 1.4). 2. The intrinsic patches in area 19 are larger and have a greater spacing than those found in areas 17 and 18. The spacing of the intrinsic correspond to the 21a and LS efferent modules. 3. The efferent output to the functionally distinct areas 21a and LS are segregated in an interdigitated band fashion (illustrated in Figure 1.4). Hypothesis one, which examines the relationship between the CO blob and interblob pathways and area 19 is detailed in Chapter 3; hypothesis two, which deals with characterizing the intrinsic connections is covered in Chapter 4; Chapter 5 and 6 deal with hypothesis three and the spatial relationship between the 21a and LS projecting cells within area 19. The results from tests of these three hypotheses show that area 19 is organized in a well-defined modular mosaic that corresponds to different parallel pathways. The first demonstrated that area 19 receives input from both the blob and interblob pathways but this input remains segregated. Segregation means that the input from one pathway (the C O blobs) is spatial separate or isolated from the input arriving from another pathway (the C O interblobs). Second, the intrinsic connections of area 19 are organized in a variably patchy fashion, that is the majority of intrinsically labeled cells are found in multiple clusters which are separated from each other by regions of cortex with relatively few labeled cells. The spacing of the intrinsic patches does not 31 CO rich area CO poor area Figure 1.4. The hypothesized organization of area 19. Within area 19 there are blob-recipient modules and interblob-recipient modules. These modules preferentially connect with each other in a patchy fashion. Area 21a receives input from one type of area 19 efferent module while LS receives input from the other type of area 19 efferent module. Thus in this model the input, intrinsic and efferent organization of area 19 is segregated. Correspond to other known modular units in area 19. This means that interconnected intrinsic patches can sample from a variety of different modules. Third, the efferent cells projecting to 21a and LS are organized in discrete bands that projecting bands a classic characteristic of modular mosaic but may also be a partially overlap. This partial overlap between 21a and LS projection bands may be a fundamental feature of extrastriate 32 organization, as it has been reported in extrastriate areas of the primate. From this I conclude that area 19 is a recipient of two parallel pathways (CO blob and interblob) and a propagator of parallel pathways (partially segregated input to 21a and LS). However, the relationship, if any, between the afferent and efferent pathways still needs to be established. Also, the patterns of segregation with regions of overlap found in this area are not specific to primates, but represents a general principle of mammalian visual cortical organizations. 33 Chapter 2. Methods The experiments that are presented in this thesis share similar procedures. For the sake of brevity, the methods that are common to all the experiments will be expounded in this chapter. Details that vary from experiment to experiment (such as the tracer used and injection placement), or methods that are specific to a single experiment will be reported in the appropriate subsequent chapter. Animals and Surgical Procedure Normal adult cats of both sexes were used. Animals received injections of three retrograde tracers: 1) cholera toxin subunit B conjugated to 7 nm colloidal gold (CTX-A u ; List Biological; 1 % in 0.9% sterile saline), 2) injections of wheat germ agglutinin-horseradish peroxidase (WGA-HRP; Sigma; 1% solution in 0.9% saline) and 3) cholera toxin subunit B (CTB). The animals were preanaesthetized with a subcutaneous injection of glycopyrilate (0.05 mg/kg, Associated Veterinary Purchasing, CO. Ltd.), followed by an intramuscular injection of ketamine (20 mg/kg, Associated Veterinary Purchasing, Co. Ltd.) and diazepam (2 mg/kg, Sabex Inc.). Animals were also given an intramuscular injection of dexamethasone (0.5 mg/kg, Austin Laboratories) to prevent brain edema. During surgery, anesthesia was maintained by inhalation of trifluourethane (halothane, M.T.C. pharmaceuticals). Under sterile surgical conditions an incision was made along the midline of the head and a long lasting local anesthetic, 0.25 % bupivacaine hydrochloride (Marcaine, Winthrop Laboratories), was injected into the incision. A 34 craniotomy and duratomy were performed at stereotaxic co-ordinates above the appropriate cortical areas. Long glass micropipettes, with tapered ends that had inner diameters of 16-20/«n, were used to make the injections. Three different tracers were used at different times in this study to verify that the pattern of label was independent of tracer used and to also allow for more than one set of connections to be traced in a single case (e.g. LS projecting cells versus 21a projecting cells). Tracer was injected by a brief application of positive air pressure, which allowed for control of the volume injected. The total volume of tracer injected was measured by calculating the drop in the meniscus of the solution in the pipette. A l l injections were made at a cortical depth of 500 pirn. In cases where multiple injections were made in a single area the spacing of the injections was approximately 500 urn to 1 mm. These multiple injections created an aggregate injection that saturated a large portion of the area. Gelfoam was then put in the craniotomies and the fascia and skin along the incision site sutured. Following the surgery, daily injections of dexamethasone (0.25 mg/kg, Austin Laboratories) were given to the animals to reduce brain edema. Perfusion and Sectioning The animals were given an overdose of barbiturate anesthetic three days after surgery. They were then perfused transcardially with 700 ml of phosphate buffer (0.1 M , pH 7.2) with 0.5% sodium nitrite. A solution of 2% paraformaldehyde in phosphate buffer was delivered with a perfusion pump at a rate of 60 ml/min for four minutes. The visual cortex was then unfolded and flattened tangentially as described by Olavarria & 35 Van Sluyters (1985). The visual hemispheres were left between two glass slides submersed in 4% paraformaldehyde and 20% sucrose in phosphate buffer. After five hours, the glass slides were removed and the sections were allowed to float free for eight hours in a solution consisting of 4% paraformaldehyde and 20% sucrose in phosphate buffer. The tissue was then frozen and cut tangentially at 50 ]im on a microtome. In order to prevent degradation in signal, sections were reacted for the appropriate tracer immediately following slicing. For both hemispheres, the posterior portion of the thalamus that contained the L G N was blocked. The L G N tissue block was then postfixed and cryoprotected in 4% paraformaldehyde and 20% sucrose and phosphate buffer for 24 hours. Next, coronal sections of the L G N block were cut at 50 fim on a freezing microtome. The sections were then processed for visualization of the appropriate tracer. These L G N sections were used to aid verification of the injection placement Histology Cytochrome Oxidase Cytochrome oxidase staining was used to help locate area borders and to reveal the CO blob and interblob pattern in areas 17 and 18. Alternate tangential sections from the superficial half of the cortex were stained for CO using a cobalt and nickel enhancement method (Crockett et al., 1993; Dyck and Cynader, 1993; Liu et al., 1993). The CO staining solution contained 20 mg diamine-benzidine, 30 mg cytochrome C, 15 mg catalase, 2 g sucrose, and 50 ml of 0.01 M phosphate buffer (pH 7.2) to which 5 ml of 1% nickel ammonium sulphate was added. One percent cobalt chloride was then added 36 dropwise until the solution became cloudy. The tissue sections were incubated in the solution for 4-6 hours then rinsed in phosphate buffer and mounted on slides. Once dried the sections were then dehydrated and coverslipped. Cholera Toxin Both the retrograde tracers cholera toxin subunit B (CTB) and cholera toxin subunit B conjugated to 7nm colloidal gold (CTX-Au) were used in the following experiments. The tracer CTB was visualized with immunocytochemistry. Tissue sections were incubated for two days at 4 degrees Celsius on a shaker table with the primary antibody, goat anti-CTB (List Biological; 1:10 000). Following this the sections were placed in the secondary antibody, biotinylated rabbit anti-goat IgG (Vector Laboratories; 1:250), for two hours. Both of the solutions containing the primary and secondary antibodies also contained 1% Triton-X-100 and 1 % normal rabbit serum. The tissue sections were then placed for one hour in avidin-biotin-HRP ("Elite" A B C kit; Vector Laboratories; 1:150). Finally, a glucose oxidase driven D A B reaction was used to visualize the label, which appeared as a brown precipitate (Itoh et al., 1979). Incubation of the sections in each of the solutions was separated by three 15 minute washes in buffer. The C T X - A u was visualized by silver intensification and appeared as a black precipitate. This was done by incubating the sections for 1-2 hours in a silver enhancement solution (Jannsen IntenSEM). Sections reacted for C T X - A u were counterstained for Nissl substance with neutral red to improve contrast and help visualize labeled cells. Sections were then dehydrated and coverslipped. 37 Horseradish Peroxidase The retrograde tracer wheatgerm agglutinin-conjugated horseradish peroxidase (WGA-HRP) was also used. The tissue sections were reacted for WGA-HRP using the standard tetramethylbenzidine (TMB) method (Mesulam, 1978). To prevent any loss of the reaction product, sections were mounted immediately on glass slides and air-dried. The next tissue slides were counterstained with Neutral Red and sections were dehydrated and coverslipped. Data Analysis Cell charting typically began several days to several months after the processing the sections. IGOR Professional 3.1 software (Wavemetrics, Inc.), a compound microscope (A Nikon Optiphof), an x-y coordinate stage encoder and a drawing tube were used to determine the spatial layout of labeled cells and landmarks in each tissue section. To superimpose the computer records onto the tissue slide, the drawing tube, which was attached to the microscope, was focused on a computer monitor. The computer mouse cursor was positioned so that it aligned with the object being recorded or charted (e.g. labeled cell). Once the two were aligned, the x, y coordinates were recorded in IGOR Pro. To prevent charting the same object twice a symbol appeared on the computer screen that was superimposed upon the recorded object with the aid of the drawing tube. Moreover IGOR Pro used the stage microscope stage encode so that the image on the screen moved in register with the stage. This allowed for large regions of the section to be scanned without missing labeled cells or recording the same cell twice. The type of labeled cell (CTB, C T X - A u or WGA-HRP) being recorded was also entered into IGOR Pro for comparisons between cells labeled by different tracers. The cross 38 sections of blood vessels were used as landmarks to align multiple sections and collapse them into a single plane, creating a cell chart. Examples of such a cell chart are illustrated in Figure 2.1 A and B. Density plots were made from the cell charts. The density plots had bins of 0.1 x 0.1 mm so that the value of each pixel corresponded to the number of cells found within the 0.01 mm 2 region. Also, a smoothing of the cell charts was done in which the position of each charted cell was overlaid with a 2D Gaussian low-pass filter. This was made by positioning a convolution kernel, with a radius of 0.25 mm (o = 1.25 mm), over each data point. Examples of the resulting smoothed image are illustrated in Figure 2.1C and D. From the smoothed image, transects were drawn, by eye, across two labeled modules perpendicular to the axis of elongation (solid line in Figure 2.1C). A module was defined in this study as being cluster of labeled cells that was greater than 0.5mm in width and was bordered by regions of cortex with relatively few or no labeled cells. Each transect produced a profile plot (examples in Figure 2.IE and F) that measured the density of pixels the transect passed through. Thus, a transect through a module resulted in a profile plot in which there was a rise in the density of label from zero, or close to zero, a peak in density label and then a decline back to zero, or close to zero values. The length of the transect corresponded to the distance from the outer edge of one module to the far edge of a second module. A peak in the profile plot corresponded to a single module, so a single transect would have two peaks. Within a module there were sometimes variations in labeling density resulted in miniature peaks. Figure 2.IF is an example of a profile plot in which there are miniature fluctuations in the labeling density within two modules. 39 These miniature peaks were not included in spacing estimates i f the peaks did not meet the criteria of a module as defined above. Due to the irregular shape of the modules, E F Figure 2.1. A ) and B) are cell charts. Each dot represents a single retrogradely filled cell. A ) represent C T X - A U labeled cells while B) represent C T B labeled cells. C) The cell chart in A after a 2D Gaussian low-pass filter was used to smooth the chart. The solid line is an example of a transect, made perpendicular to the axis of elongation, to measure spacing between two modules. D) The cell chart in B filtered with a 2D Gaussian . The dashed lines represent the progression of transects from medial to lateral. The measurements from each of these transects is added up and then averaged to give a mean spacing for those two modules. E) and F) are profile plots generated from the transect (solid line) in C) and D) respectively. The profile plots represent the density of label along the length of the transect. Note in F) there are multiple miniature peaks within the two modules. These miniature peaks were not replicable and were not included in spacing measurements. Scale bar in A ) = 1 mm while the scale bar in B) = 2mm. 40 multiple transects progressing through the module from medial to lateral were taken (dotted and solid lines through Figure 2.1 D) and then averaged for each pair of modules. The spacing of the modules was determined by dividing the distance between peaks by the number of modules (two). The spacing measurements were then averaged for an overall mean spacing per case. Measurements were not adjusted to take into account shrinkage of tissue from both fixation and processing and are thus minimal estimates of width and periodicity of the modules. A quantification of clustering was also performed for each chart. The Cluster Index (CI) (Ruthazer and Stryker, 1996) which was based on the Hopkins' statistic (Hopkins, 1954; Ripely, 1981) required a measurement of nearest-neighbor distance (w) and the distance between a randomly chosen point and the closest data point (JC). The CI value was Log [(^/(w2)] with random distributions resulting in values near zero and clustered distributions resulting in higher CI values. The CI index does not measure the number of cells within a given cluster but rather is a measurement of the total distribution of cells. Thus, the greater the number of cells in between cell clusters the lower the CI value. In IGOR Pro, 25000 cells were chosen, along with an equal number of randomly chosen points, and were used to calculate CI. Ten different samples were preformed and the results averaged to give a CI value for each case. The location of retrogradely labeled cells with respect to the CO blobs was also determined. Images of the CO sections were obtained using a digital camera (Coolpix 990, Nikon) at high resolution. A correlation between the two was preformed. First, the CO image and cell chart were aligned in IGOR Pro using blood vessels. A 2D Gaussian low-pass filter was applied to the cells to create a smoothed image of the labeling density. 41 Both the GO image and smoothed cell chart had scale axes attached. The two images were imported into PhotoShop and their axes aligned using the opacity option. Once aligned the overlaid images were cropped tightly around the labeling pattern. Typically the patchy pattern of cells was strong in the central area and faded towards the edges, probably reflecting the topographical connectivity between the injection site and the labeled cells, To focus the calculations on the area of interest a third image was created with a strong Gaussian blur with a radius of 50 pixels that reduced the cell pattern to a Gaussian centered on the area of interest This third image thus showed the regions of the cell image that had labeling and the regions that did not and was used in the correlation to weigh the data so the regions without cell labeling had less of an impact on the analysis. Lastly a fourth image was created from the CO image that masked blood vessels, tears and tissue edges from the correlation analysis. A l l images had a resolution of 300 pixels/inch. A correlation was then preformed to give a correlation co-efficient for the CO and the cell image. A hypothetical case of cells concentrated in the interblobs would result in a negative correlation while a hypothetical case of cells concentrated in the blobs would result in a positive correlation. Next, this correlation was repeated but with the image of the cells shifted from -0.6mm to 0.6 mm at 0.1mm intervals. The range of -0.6 to 0.6 was chosen as the CO blobs are about 0.7mm in size and a shift of that magnitude ought to include at least one cycle through a CO blob. It was theorized that if the two patchy systems (CO versus patches of cells) were related then the pattern of correlations over the -0.6 to 0.6 mm shift would show a U shaped function (cells related to the interblobs) or a inverted U shaped function (cells related to the CO blobs). Thus, the correlation co-efficient for the -0.6 to 0.6 mm shift was graphed. 42 This simple analysis was sometimes backed with a full two dimensional cross correlation between CO density and labeled cells (Boyd and Casagrande, 1999). To do this the image was first low-pass filtered in PhotoShop. The levels were then adjusted to improve contrast. The whitest parts of the CO image, which corresponded to blood vessels, tears in the tissue and the glass slide beyond the edge of a slice, were then selected, with the colour select option and filled in with black to mask these regions from the analysis. The image was then transferred into IGOR Pro and the charted cells aligned with the CO image using blood vessels . For a hypothetical case in which labeled cells were concentrated in CO blobs, CO staining density would be higher than average close to the labeled cells. This would create a central area of high correlation (indicted by darkness in the resultant image), with the spot having the same size, orientation and spacing as an average CO blob. In a hypothetical case in which the cells are located in interblobs the dark CO staining from the blobs would occur some distance from cells and thus the peak positive correlations would be offset from the origin of the plot. A one-dimensional summary of the spatial correlation plot was also created that quantified the correlation and hence the density of CO staining at different distances from labeled cells. The darkest staining was represented by the value of zero while lighter CO staining corresponded to higher values. For example, in a case in which the labeled cells overlap with CO blobs, the CO values increased as the distance from labeled cells increased. This same procedure was performed to give a "segregation" estimate between modules labeled by separate tracers. Thus, if the cells were segregated or interdigitated the one-dimensional spatial correlation plot showed that as the distance from one type of labeled cell increased so did the density from a second type of labeled cell. Also, a cell chart 43 including one type of labeled cell (e.g. CTX-Au), which was illustrated by either red or blue dots, was superimposed upon the density plot generated from the cell chart from a second type of labeled cell (e.g. CTB). Since the density plot was illustrated in grey scale, the overlaid red and blue cells were easily visible. This allowed for a relative comparison by eye of the label density between labeling patterns generated by the two types or tracers. 44 3. The Projection from CO Blobs and Interblobs to Area 19 Introduction The periodic modular mosaic of cell networks is a notable feature of the cat primary visual cortex. The different anatomical structures of modules reported in visual cortex include patches/blobs (Boyd and Casagrande, 1999), stripes (Olavarria and Van Essen, 1997) or irregularly shaped clusters (Xiao et al., 1999; Tanaka, 2003). This modular organization is visible in tract tracing studies that examine geniculate (LeVay and Gilbert, 1976; Boyd and Matsubara, 1996), mtrinsic (Ferrer et al., 1988; Gilbert and Wiesel, 1989), efferent (Gilbert and Kelly, 1975; Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b; Boyd and Matsubara, 1999; Conway et al., 2000) and callosal connections (Segraves and Rosenquist, 1982; Berman and Payne, 1983; Voigt et al., 1988; Payne and Siwek, 1991; Boyd and Matsubara, 1994a; Olavarria, 2001). The patchy distribution of connections to and from the primary visual cortex is spatially related to the pattern of CO staining (Boyd and Matsubara, 1994a; Boyd and Matsubara, 1996; Boyd and Matsubara, 1999; Conway et al., 2000). In the primate, this modular organization is an anatomical hallmark of the parallel processing streams associated with the M , P and K classes of retinal ganglion cells. Functionally, these three pathways differ in such variables as contrast, spatial sensitivities, receptive fields and temporal resolution (for review see Casagrande and Norton, 1991). Cytochrome oxidase blobs in the primary visual cortex of the primate receive a direct thalamic input from the K cells (Lachica and Casagrande, 1992). Also, in V2, CO thick and pale stripes receive projections from the 45 CO interblobs, whereas the thin stripes of V2 receive input from the CO blobs (Sincich and Horton, 2002a). It may be in the cat that the modular populations of cells that align with CO blobs represent a distinct processing stream. The purpose of this chapter is to examine primary visual cortex efferents to extrastriate area 19 in order to determine i f this population of efferent cells has any spatial relationship with CO staining. Previous studies (Gilbert and Kelly, 1975; Bullier et al., 1984a; Symonds and Rosenquist, 1984b; Price and Blakemore, 1985a; Price and Blakemore, 1985b; Kato et al., 1993) have shown that small injections of a retrograde tracer in area 19 yield patches of labeled cells in area 17. As patchy efferents to other extrastriate areas coincide with CO blobs (Boyd and Matsubara, 1999; Conway et al., 2000), it may be that patches of area 19 efferent cells are also localized in the CO blobs. However, because of the small sizes of the injections used, these studies do not address the issue of whether the connections between the primary visual cortex and area 19 are hard or soft patterned, as defined by Shipp and Grant (1991). Hard patterned connections result from a discontinuous organization of either efferent cells (Figure 3.1 A) or terminals (Figure 3. IB). Modular organization is an inherent feature of hard patterning. Alternatively, soft patterned connections occur when cells projecting to another area are organized continuously and involve the entire efferent area (Figure 3.1C). Soft patterning can also demonstrate modular organization on a local scale (Shipp and Grant, 1991). For example, in Figure 3.1C i f a small injection that labeled only a shaded region was made, it would result in modular label even though all of area X projects to area Y . Therefore, to determine i f a connection is hard or soft patterned the size of the injection must be 4 6 Figure 3.1. A schematic drawing of hard and soft patterned connections between hypothetical areas X and Y . A) An example of hard patterning in which the efferent cells are clustered in discrete modules. B) An example of hard patterning of afferents. The axon terminals in this case are restricted to a single module. C) This is an example of soft patterning, such that on a global scale the connections are continuous. A l l regions of area X project to area Y . However, there are still discontinuities or modules within the areas so that if an injection was restricted to a single module in either area X or Y then a modular pattern of labeling would result in the other area. 47 large enough to cover existing modules. Although most studies looking at the projections from area 17 to area 19 only look at small injections, one study by Ferrer et al. (1992) reports that clusters of cells in area 17 that project to area 19 are independent of the amount of retrograde tracer injected. Ferrer et al. concludes that only specific patchy regions within area 17 send input to area 19. While this result suggests that organization of this efferent population may be hard patterned, the Ferrer et al. study does not specify the size or amount of the tracer injected and therefore it is difficult to say if the injection was sufficiently large to determine hard versus soft patterning. In the primate, input to V2, which is the next area in the visual hierarchy following VI (Felleman and Van Essen, 1991; Scannell and Young, 1993), is soft patterned, that is the connections between VI and V2 are globally continuous and only small injections reveal a modular pattern (Van Essen et al., 1986). As area 19 in the cat is the next area in the visual hierarchy following the primary visual cortex (Scannell et al., 1999) and receives the bulk of its input from area 17 (Bullier et al., 1984a; Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b; Mulligan and Sherk, 1993) it is a likely candidate for soft patterning, like V2. In fact, it has been found that when sufficiently large injections of retrograde tracer (diameter of approximately 5 mm) are made in area 19 the label in area 17 is continuous. That is the projection from area 17 to area 19 is soft patterning such that there are patches of retrogradely labeled cells in both CO blobs and interblobs (Stewart et al., 2003b). In the present experiment, the relationship between CO blobs and the efferent cells in area 17 that project to area 19 is addressed. It was hypothesized that ce^ ls located in CO blobs would project to modules that were separate from the modules receiving 48 input from the CO interblobs. This is important in identifying which parallel pathway/s are projecting to area 19 and establishing the pattern of the afferent contribution (i.e. soft versus hard patterning) to the modular mosaic within area 19. A schematic of this is visible in Figure 3.1C where the blobs and interblobs are projecting to different compartments. 49 Method Animal and Surgical Procedure Twelve hemispheres from nine normal adult cats were used. A craniotomy and duratomy were performed at stereotaxic co-ordinates A P -3 to +6 and M L +7 to +15. The craniotomies were made this size to allow for a small injection to be made in area 19, in addition to the injection of other tracers in areas 21a and LS for the purpose of experiments detailed in Chapters 4-6. In one case, a bulk injection CTB was made in area 19. The small injection was made in the posterior portion of area 19, on the crown of the suprasylvian gyrus, medial to the genu. The pipette was oriented perpendicular to the cortical surface and the injections were spaced 1 mm apart in a 2 by 2 grid. A total of 3-4[xl of tracer was injected. In the remaining eleven hemispheres a single small injection of a retrograde tracer was made in area 19 on the far medial side of the suprasylvian gyrus above the genu. Amounts ranging from 0.04-1.3 jxl of WGA-HRP or CTB were made in area 19. Multiple injections of CTX-Au , which were placed in area 21a for another experiment, resulted in label in area 17 that was used as an additional marker for CO blobs (Conway et al., 2000). Histology The animals survived 1-3 days following surgery. The visual cortex was unfolded, flattened, and sliced tangentially. Alternate tangential sections from the superficial half of the cortex were stained for CO using a cobalt and nickel enhancement method (Silverman and Tootell, 1987). The remaining tangential slices, as well as the 50 L G N sections, were reacted for their respective retrograde tracer. The CTX-Au was visualized by silver intensification. The WGA-HRP was visualized using the standard tetramethylbenzidine (TMB) method (Mesulam, 1978). The CTB label was reacted for by immunocytochemistry and visualized by the glucose oxidase-driven diaminobenzidine method (Itoh et al., 1979). Data Analysis The labeling in area 17 was charted and aligned with photographs of the CO staining using blood vessels as landmarks. Two-dimensional histograms measured the density of labeled cells. Two-dimensional Gaussian histograms and transects measuring the spacing of labeled cells in area 17 were also made. A cluster analysis was conducted and each case received a CI index to demonstrate the degree of "patchiness". To show that the distribution of labeled cells in relation to the CO staining was greater than chance, a chi square analysis was performed. Finally, 2D cross correlations were used to further demonstrate the relationship between the pattern of labeled cells and CO staining. 51 Results Global Projections to Area 19 A portion of area 17 was labeled (as illustrated by the composite cell chart in Figure 3.2A) following a bulk injection (3-4ul) of retrograde tracer that covered over 1 mm 2 of area 19. The resulting pattern of label in area 17 was continuous; there were no i 1 1 1 r~ -23.5 -23.0 -22.5 -22.0 -21.5 mm Figure 3.2. The pattern of label in area 17 following a large injection in area 19. A) a composite cell chart of labeled cells in area 17. Note that there were no discrete patches of label. B) A 2D histogram in which each pixel corresponds to a 0.01 mm 2 region of cortex and the darkness of the pixel represents the number of cells found within it. C) a 2D Gaussian low-pass filter image demonstrates fluctuations in the density of labeled cells. The dotted line is a transect drawn through the image to measure density. D) the resultant profile plot from the transect drawn in C. Note the variations in labeling density. The mean spacing of these fluctuations was 0.78mm. Scale bar = 1mm. Tracer used: CTB. 52 discrete clusters or modules separated by regions of unlabeled cortex, except a little around the edges of the labeled zone where the density of label fell off. This is illustrated in figure 3.2B in which a 2D histogram shows the distribution and density of labeled cells throughout the labeled patch. The CI value for the large injection was low at 0.28 (S.D. = 0.04). This large zone of continuous label encompassed both CO blobs and interblobs. Thus, the projection from area 17 to extrastriate area 19 is not just restricted to one type of CO module, as is the case for LS and 21a projections (Boyd and Matsubara, 1999; Conway et al., 2000). There was some structure within the labeled region as fluctuations in the density of labeled cells varied in a periodic manner. Transects measuring labeling density, as demonstrated in Figure 3.2C, revealed regions that contained a higher density of labeled cells (Figure 3.2D). These most densely labeled regions had a mean spacing of roughly 0.78 mm, similar to spacing of CO blobs, which is approximately 0.75 mm (Boyd and Matsubara, 1996). However, when compared with an alternate section that had been stained for CO there was no obvious relationship between the pattern of CO staining and labeled cells. Patchy Projections to Area 19 A small injection ranging from 0.04-1.3 ul of retrograde tracer in area 19 resulted in patchy labeling in area 17. The amount and degree of clustering varied for different cases and the CI ranged from 0.32 to 1.42. The cluster values from each case are presented in Table 3.1. Cell charts in Figure 3.3A, B and C illustrate this variability. The 2D histograms in Figure 3.3D, E and F show the density of the patches. Unlike the continuous labeling found after a large injection, the patchy label resulting from the small 53 Figure 3.3. Variation in the amount of "patchiness" between different cases. A) B) and C) Cell charts of patchy labeling in area 17 following a small injection in area 19. Their CI values were 0.72. 0.57 and 0.32 respectively. The tracer used for each case is WGA-HRP, WGA-HRP and CTB respectively. D) E) and F) a 2D histogram illustrating the density of label for each pixel. Scale bar = 1mm. injection has regions of cortex between the clusters of label that contain few labeled cells. The overall mean CI value for the patches resulting from a small injection was 0.62 54 (SD=0.337). This CI values was significantly greater (t-test, p < 0.0001) than those for a random distribution. The average spacing of labeled patches was approximately 0.63 mm. Table 3.1: Cluster Index and Spacing Values for Each Case. Case No. Cluster Index (CI) Spacing (mm) SI* 0.28 0.78 S3 0.34 0.68 S4 0.36 0.67 S5 0.72 0.79 S6-LH 0.64 0.60 S6-RH 0.72 0.68 S7 0.57 0.57 S9 0.91 0.70 S10-LH 0.42 0.66 S10-RH 0.32 0.66 S l l - L H 0.90 0.72 S l l - R H 1.45 0.66 *Case in which a large injection was mac e in area 19. The clustering data from 21a and LS projecting cells in area 17 was obtained from previous experiments (Boyd and Matsubara, 1999; Conway et al., 2000) and compared to the clustering of area 19 projecting cells. The biggest difference was that following a large injection in either area 21a or L S, labeled cells were concentrated in the CO blobs. In the case of area 19 a large injection of retrograde tracer labeled cells in both the CO blobs and interblobs. Also, after a small injection in area 19 labeled cells clearly had a larger range of CI values (0.28-1.44) in area 17 compared to either the 21a (0.44-1.16) or LS (0.28-0.309) projecting cells. This is illustrated in Figure 3.4. The large range for area 19 projecting cells is not unexpected since cells in both the CO blobs and interblobs projected to area 19. Thus, injections in area 19 created a large number of possible • labeling outcomes and thereby increased the range of clustering. 55 19 21a LS F i g u r e 3.4. A comparison of the range of CI values obtained from cells in area 17 that project to area 19,21a and LS. Note that the area 19 projecting cells produce the greatest range of CI values. There are cases with very high clustering of areal9 projecting cells and cases with relatively low clustering of area 19 projecting cells. Comparison of CO and Patches of Label Since the pattern of labeled cells were clustered a direct comparison between labeled cells and CO blobs and interblobs was done. This comparison was preformed (n=7) by staining alternate tissue sections for either CO or retrograde label and then aligning the sections using radial blood vessels. When the labeled cells were overlaid on the CO sections they appeared to be either concentrated in the blob compartments or, in other cases, concentrated in the interblob compartments. This impression was tested 56 quantitatively by generating two dimensional spatial cross correlation that determined the average CO staining intensity at different distances from the labeled cells. The relationship between CO staining and labeled cells was also examined by correlating the pattern of cell label with the CO image. The cells were then shifted 0.6mm to - 0.6m in 0.1mm increments. After each shift, the correlation between the CO staining and labeling pattern was determined. In overlaid sections, the labeled cells appear to be preferentially located in some cases in the CO blobs or in other cases in the interblobs. Figure 3.5B and D are examples of labeled cells (red dots) overlaid on corresponding CO sections. In Figure 3.5B and D the majority of labeled cells are encountered in the blobs (the more darkly stained regions), while in Figure 3.5D the labeled cells have a tendency to fall in the interblobs (the less densely stained regions). The patchy distribution of the labeled cells illustrated in Figure 3.5, is evident in the smoothed images of the labeling displayed in A and C; a Gaussian filter was applied to more easily identify the cell clusters. The qualitative observations were supported by the results from the 2D cross correlation analysis. This analysis demonstrated a tendency for the cells to be located in the CO blobs or interblobs. Figure 3.6 show the 2D cross correlations for six cases. The origin of the plot (white circle) is the reference point. Figure 3.6 A, B, C and F are examples in which the majority of labeled cells are found in the interblobs. Figure 3.6D and E are examples in which the majority of labeled cells are found in the blobs. The interblob cases tended to have light pixels around the origin, at the center of the plot. In these cases the pixels began to darken around 0.4mm away from the origin. This indicates that the CO staining intensity was greater away from the cells. In comparison 57 3.0 3.5 4.0 4.5 5.0 5.5 6.0 mm Figure 3.5. A comparison of C O staining and retrogradely labeled cells. A) To more easily view the patches of labeled cells, a Gaussian filtered was applied to the cell chart. B) The C O stained section is overlaid with the cell chart that was smoothed in A . Each labeled cell is indicated by a red dot. Note that the majority of labeled cells co-localize with the C O blobs (darker stained regions) while there are some cells also scattered in the interblobs (lighter stained regions). C ) A cell chart that was smoothed with a Gaussian filter. The patchy nature of the labeled cells is visible. D) The C O stained section is overlaid with the cell chart that was smoothed in C. Each retrogradely labeled cells is indicated by a red dot. Note in this case that the majority of labeled cells co-localize with the C O interblobs (lighter stained regions) while there are some cells also scattered in the blobs (darker stained regions). The tracers used in these cases were C T B and W G A - H R P respectively. the blob cases had dark staining at the origin of the plot. This indicated that the C O staining was darker close to the blobs. The 2D cross correlations also showed many 58 I I I I I -0.4 -0.2 0.0 0.2 0.4 • i i i i -0.4 -0.2 0.0 0.2 0.4 "1 1 1 r -0.4 -0.2 0.0 0.2 0.4 i 1 1 1 r -0.4 -0.2 0.0 0.2 0.4 Figure 3.6. Two dimensional spatial cross correlations that look at the average CO staining intensity at different distances from the labeled cells. Dark pixels correspond to darker CO staining. The white circle represents the origin of the plot. Light regions surrounding the origin illustrate that the CO staining near the cells is lighter (i.e. interblobs). Dark regions surrounding the origin of the plot illustrate that the CO staining near the cells was darker (i.e. blobs). Thus, A , B, C, and F are examples of cells being concentrated in the interblobs, while D and E are examples of cells that are concentrated in the blobs. The axes are in mm. 59 dark pixels near the center of the plot, but also along the left edge of the plot. These variations are likely due to fluctuations in the CO pattern of staining. This is the case for Figure 3.6D in which the auto-correlation of the CO (Figure 3.9 - S9) did not show a consistent pattern. Plots of one-dimensional spatial cross correlation are shown in Figure 3.7. Dark 1 > 1 1 1 > 1 1 i 1 1 1 1 r 1 0.0 0.2 0.4 0.6 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.2 0.4 0.6 0.0 0.2 0.4 0.6 Distance (mm) Distance (mm) Figure 3.7 One dimensional spatial correlations between the CO staining intensity and different distances from the labeled cells. Examples A - F correspond to those shown in Figure 3.6 respectively. Higher values represent lighter CO intensity. Note that in cases A , B , C and F the CO staining shifts from higher levels (lighter CO intensity) to lower levels (darker CO intensity) over a 0.6mm distance. Thus at further distances from the cells the CO intensity is increasing. The opposite happens in D and E. In these cases the CO staining is shifting from low levels (darker CO intensity) to higher values (lighter CO intensity). Thus, at further distance from the cells the CO intensity is decreasing. This further suggests that the cells in examples A , B, C and F are found in the interblobs and cells in examples D and E are located in the blobs. 60 CO staining is represented by lower values in these graphs. The A-F placement in this figure corresponds to the A-F placement respectively in Figure 3.6, so 3.7A is derived from the same case as 3.6A, and so on. For the interblob cases (A, B , C, F), the ID plots show an increase in CO staining (represented by a decrease in the value of the left axis) as you move approximately 0.6mm away from the cells. In the blob cases (D, E) the ID plots show a decrease in CO staining (represented by an increase in the value of the right axis as you move approximately 0.6mm away from the cells. Images of the CO staining and the labeling pattern were also correlated. Although the correlation coefficients were relatively low (listed in Table 3.2), they showed a Table 3.2 Correlation Analysis on CO and Cell Labeling. Case C O Analysis r Figures S5 Interblobs -0.047 3.8D S6-LH Interblobs -.0152 3.6C/3.7C/3.8C S6-RH Interblobs -0.119 3.6B/3.7B/3.9 S7 Interblobs -0.04 3.6F/3.7F/3.8A S9 Blobs 0.264 3.6D/3.7D/3.9 S10-RH Interblobs -.054 3.6A/3.7/3.8E S l l - L H Blobs 0.062 3.6E/3.7E/3.8B distinct pattern. The interblob cases resulted in negative correlations while the blob cases resulted in positive correlations. One of the aligned images was then shifted in 0.1 mm increments to the right and the left and the correlation after each shift was calculated. Graphing the correlation distribution resulted in a rough U shaped pattern, shown in Figure 3.8. In a blob case (Figure 3.8B), an inverted U shape become apparent with the aligned correlation being positive. As you shift to right and left to -0.6 mm and 0.6 mm the correlation coefficient becomes negative. In the interblob cases (3.8A, C, D and E), 61 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Distance (mm) Figure 3.8 The distribution of correlation co-efficients as the image of the labeled cells is shifted with respect the CO image. While the correlation co-efficients are relatively low, they change in a systematic fashion. Zero represents the point at which the two images are aligned. In A , C and D the peak correlation is offset from the aligned position by 0.2-0.3mm. A n exception is E , which is a case in which the cross correlations suggest the cells are located in the interblobs (Figure 3.6A and 3.7A). In E the correlation co-efficients cycle through two peaks in relatively small distance. This is not seen in the other cases. as you shift away from the aligned position the correlation decreases and shifts into positive correlations. In three cases (S5, S6-RH and S7) the peak correlation was offset from the aligned position by 0.2 -0.3mm. Case S10-RH was an exception, in that it 62 cycled through a rise and fall of the correlation coefficient in about half the distance of the other cases. There were two cases that when shifted did not show this U shaped pattern in correlation coefficients. A s the cells were shifted from -0.6 to 0.6mm the correlations increased in a linear fashion. One of these cases (S6-RH) had a very large edge, which likely affected the calculations. For the other case (S9) the C O oxidase staining was irregular, which is demonstrated in the auto-correlation in Figure 3.9, and thus shifting the cells did not show the typical inverted U shaped pattern. CO Staining Labeled Cells S9 RHS6 Figure 3.9 Auto-correlations of the C O and labeled cell images for the two cases that did not show a U shaped pattern of correlation coefficients. For S9, the auto-correlation of the C O staining shows a non-typical pattern of C O staining, which would effect the results of a shifting analysis. The auto correlations for C O staining and labeled cells for S 6 - R H are also shown. For this case, there was a large edge effect that resulted in an linear increase in the correlation coefficient as the labeled cells were shifted away onto the C O stained tissue and away from the edge. 63 LGN Labeling In cases that received a small injection of retrograde tracer in area 19, coronal sections of the L G N were also reacted for the appropriate tracer. These sections consistently revealed labeling in the C layers of the L G N and medial interlaminar nucleus (MIN). A n example of typical L G N label following an injection in area 19 is shown in Figure 3.10. The arrows are pointing to labeled cells in the C layers and M I N . There were an absence of labeled cells in layers A and A 1 . Figure 3.10 Labeled cells in the L G N following a small injection in area 19. The labeled cells appear as dark spots located in the C layers of the L G N and M I N . Arrows indicate several labeled cells. The background has been stained for Nissl to help visualize the geniculate layers. 64 Discussion A bulk injection in area 19 that covered a large region (about 2.5 mm2) resulted in continuous label in area 17. Within this continuous label there were periodic regions that contained denser label. On the other hand, small focal injections of retrograde tracer in area 19 resulted in a patchy distribution of labeled cells. This means that the afferent connections in area 19 that originate in area 17 are soft patterned. Within area 19 there is a modular organization enmeshed in a global projection. This scenario is illustrated in Figure 3.1C. To differing degrees, the majority of labeled cells were concentrated in either the blobs or the interblobs. This suggests that within area 19 there are specialized compartments that received input from cells primarily located either in the blobs or the interblobs. Therefore, when more than one type of compartment is injected (i.e. large bulk injection) cells in both the blobs and interblobs become labeled. Although the specialized 17 recipient compartments remain to be defined in area 19, there presence likely contribute to the area 19 modular mosaic. Methodological Considerations While a large injection of tracer in area 19 yielded continuous labeling, the density of labeling fluctuated in periodic clusters. It could be argued that fluctuations in labeling density arose from uneven absorption and transport of tracer. The size, structure, and spacing of the dense patches of labeling were consistent with other similar experiments that involved a bulk injection in area 19 (Stewart et al., 2003b). There is some evidence that area 19 is modular, with large efferent modules projecting to area 21a - the details are found in Chapter 5 (Stewart et al., 2000). If this is the case, the periodic 65 fluctuations in density may have resulted from injecting more tracer into one type of module over another. This may have occurred with the large injection so that the denser patches of label may represent cells projecting to a single type of module within area 19, which just happened to receive more tracer. With the small injections, the amount of tracer used was sufficiently small to minimize spread from one module to another. Injection sites for area 19 were detennined based upon stereotaxic co-ordinates and sulcul patterns. Comparing alternate sections that were stained for CO allowed for visual confirmation that the retrograde tracers were correctly placed in area 19. Cytochrome oxidase staining remarkably drops off at the area 18/19 border (Price, 1985a). This border is clearly visible in tangential sections (Boyd and Matsubara, 1996). Thus, alignment of CO stained sections with sections processed for labeling allowed identification of the injection site with reference to the area 18/19 border. The placement of the small focal injection was further verified by examining L G N slices for labeling. Figure 3.11 is a schematic of typical labeling in the L G N , and the neighboring subcortical nucleus MIN, after an injection in either area 18,19 or 21a. It was ruled out that the focal injections were placed in any of the three areas, which border area 19 around suprasylvian and lateral genu (areas 18, 21a and 7). First, the L G N layers A and A l did not contain any label confirming that the injection placements did not encroach on area 18, as area 18 and not area 19 projects to A and A l (Stone and Dreher, 1973; Maciewicz, 1975; LeVay and Gilbert, 1976; Hollander and Vanegas, 1977; Harvey, 1980; Geisert, 1985; Humphery et al., 1985a; Kawano, 1998). Not surprisingly, layers C1-C3 of the L G N , which is the main source of geniculate input to area 19 (Maciewicz, 1975; LeVay and Gilbert, 1976; Hollander and Vanegas, 1977; Dreher et al., 66 1980; Raczkowski and Rosenquist, 1980; Raczkowski and Rosenquist, 1983; Kawano, 1998), had labeling present. Area 7, which borders area 19 along its anterior lateral Figure 3.11. A schematic of the label in the L G N following an injection into areas 18,19 and 21a. Open circles represent area 18 projecting cells, open diamonds are area 19 projecting cells and the closed circles represent 21a projecting geniculate cells. The location of area 18 projecting geniculate cells is based upon previous research (Stone and Dreher, 1973; Maciewicz, 1975; LeVay and Gilbert, 1976; Hollander and Vanegas, 1977; Harvey, 1980; Geisert, 1985; Humphery et al., 1985; Kawano, 1998). However, the position of area 19 and 21a projecting cells is based upon data from this experiment and previous studies in our lab (Conway et al., 2000). Area 18 geniculate efferents originate in layers A , A l , and C (but not C1-C3) as well as MIN. Area 19 injections on the other hand resulted in label in the C1-C3 layers of the L G N as well as MIN, while 21a injections only labelled the C1-C3 layers and not MIN. extent, does not receive projections from the L G N (Olson and Lawler, 1987), so the presence of label in the C1-C3 layers ruled out the possibility of the injection accidentally being placed in area 7, but this does not rule out the possibility of spillage of the tracer into area 7. Last, MIN also contained labeled cells. As area 19, and not area 21a, Calculating the correlation of the CO and cell images as they were shifted relative to each other resulted in a consistent pattern, despite the low correlation coefficients. The receives 67 low correlation coefficients are likely related to the noise in the CO image. This noise originated from portions of the image that had no tissue or labeling present, or where the where the slice went through a layer without CO blob/interblob, tears in the tissue, blood vessels and variation in the intensity of CO staining within a section. However, notwithstanding this noise, for the cases that the cross correlation revealed a correspondence between the cells and the interblobs, the correlation for the two aligned images was negative. As the image of the cells was shifted either to the right or left of the aligned position the correlations value decreased created a U shaped pattern (Figure 3.8A, C, D, E). For a case that the cross correlation revealed a correspondence between cells and blobs the correlation between these images was positive. As the image of the cell was shifted to the right or left the correlation systematically decreased (Figure 3.8B). Interestingly, for three of these cases there were better correlations when the images were shifted in one direction 0.2-0.3mm. This difference between the aligned images and the peak correlation is greater than minor misalignments that come from differential shrinkage and stretching of the two tissue section. It is possible that while there is a relationship between the CO staining the patches of labeled cells, they are slightly offset from each other by about 0.2 mm. However, there were also two cases where the peak correlations corresponded to the aligned images, suggesting that this is not the case Patchy Area 19 Efferent Cells in Primary Visual Cortex The large zone (approximately 2.5 mm2) of labeling in area 17 following a bulk injection of retrograde tracer in area 19 suggests that the organization of these efferent cells was soft patterned. That is, the projections from area 17 to area 19 were not 68 restricted to a single type of module or CO blob; area 17 projections to area 19 were continuous. Having said that, there were periodic fluctuations in the density of labeled cells within the labeled region in area 17. This fluctuation has also been noted for other cases that have received a large injection in area 19 (Stewart et al., 2003b). Thus, while the projections from area 17 to area 19 were continuous there is a suggestion that the projections is uneven, with some regions having a greater number of cells projecting to area 19. It was recently reported in the primate that the pale stripes in V2, which receive input from the CO interblobs, receive a stronger projection than the other stripe compartments. Thus when bulk injections of an anterograde tracer are placed in V I the pale stripes show the densest number labeled terminals (Sincich and Horton, 2002b). By logical extension, it would be expected that bulk injections of a retrograde tracer in V 2 would then result in heavier label of the CO interblobs. This may be the case with the bulk injections in area 19. In this current study it was discovered that labeled cells could form a discontinuous patchy pattern, but only when a small subset of area 17 efferent cells are labeled. This was accomplished by making small focal injections in area 19. The spacing of the area 19 projecting patches fall within the range of other anatomical modules reported for a area 17 (see Table 3.3), but are closest in value to the CO blobs. Table 3.3 Spacing of the Different Anatomical Structures in Area 17 Afferent W input Afferent Y input Ocular Dominance Columns Blobs Intrinsic Clusters Efferent 19 Projection Efferent 21a projections Efferent LS projections Area 17 <1.0 <1.0 1.1 0.75 0.92 0.63 0.73 <1.0 (Boyd and Matsubara, 1996; Kisvarday et al., 1997; Boyd and Matsubara, 1999; Conway et al., 2000; Rathjen et al., 2002) 69 Other studies also reported a patchy organization of area 17 cells projecting to extrastriate area 19 (Gilbert and Kelly, 1975; Bullier et al., 1984a; Symonds and Rosenquist, 1984b; Price and Blakemore, 1985a; Price and Blakemore, 1985b; Ferrer et al., 1992). Most of these studies used small injections of tracer. However, Ferrer et al., also made large injections of retrograde tracer in area 19 and found discrete patches of labeled cells that were independent of the injection size. This differed from the present findings that bulk injections in area 19 resulted in density fluctuations of labeled cells but did not create labeled patches that were discreetly separated from one another. One reason for this difference could be that different planes of viewing were used in the two studies. For analysis, Ferrer et al. reconstructed sections from a one in three series of coronal sections, while the present experiment was conducted on tangential sections. Patterns viewed across a large area of cortex are often easier to detect in the tangential plane. This is exemplified by the fact that CO blobs were only discovered in cat primary visual cortex once the cortex was examined tangentially (Murphy et al., 1995). It may be that the regions of area 17 which were more sparsely labeled were harder to detect in the coronal plane. Also, Ferrer did not state the size of the injection and may not have made an injection site substantial enough to cover more than one type of afferent module in area 19. The degree of clustering was quantified and represented by the CI value (see Chapter 2 for details). Random distributions result in CI values close to zero while greater numbers represent greater clustering of a distribution. The CI values for labeling in area 17 following a small injection in area 19 was significantly greater than random, but the CI values varied considerably. In comparison, the range of CI values for 21a and LS projecting cells in area 17 [which was determined from previous experiments (Boyd and Matsubara, 1999; Conway et al., 2000)] was less. One reason for the variability of area 19 projecting patches compared to the 21a and LS projections is that area 19 projections arose from both CO blobs and interblobs, while 21a and LS projecting cells were clustered only around the CO blobs. Thus, the clustering of efferents to 21a and LS in area 17 were restricted to one module while efferents to area 19 had no such restrictions and showed greater variability in the degree of clustering. Also, i f the CO blobs and interblobs projected to spatially distinct compartments in area 19 (see section below for details) injections that intruded into a neighboring compartment would produce different degrees of clustering. For example, i f an injection was placed in a CO blob recipient compartment but extended into an interblob recipient compartment the resulting label would be primarily concentrated in the CO blobs but some label would also be present in the interblobs. Relationship of Efferent Patches to CO Staining Since the projection between area 17 and 19 is soft patterned (i.e. there are discrete patches after a small injection but not after a large injection) this means that the afferents to area 19 are organized in a modular fashion (See Figure 3.1C for details). The results from this study, which show that after a small injection the majority of labeled cells that were concentrated in either the CO blobs or interblobs. This suggests that the afferent modules in area 19 are organized according to a preference for projections that originate in specific type of CO module (blob versus interblob). Area 19 does contain bands of 21a projecting cells (see Chapter 5). This band organization may not be 71 restricted to efferent cells but may be part of a general organizational structure within area 19. As the injection sites in these cases were small, it can be hypothesized that each injection hit a specific type of compartment, one that received the majority of its input from the blobs or a compartment that received input primarily from the interblobs. This hypothesis can also account for the wide range of the CI values and the varying correlations between labeled cells and CO staining. In some cases, the injection in area 19 may have spilled from one type of afferent compartments (e.g. blob recipient compartment) into a second type of afferent compartment (e.g. interblob recipient compartment). If this were so, then labeled cells should be found in varying ratios in both the blobs and interblobs, which was the case. Also, the spatial segregation of afferents in area 19 according to their relationship to the CO blob/interblob structure is further supported by the results from experiments with paired injections (Stewart et al., 2003b). In these experiments label in area 17 and 18 from the two different tracers either co-localized or interdigitated. In the instance in which the labeled cells from the two injections overlapped it can be hypothesized that the injections either hit the same CO recipient compartment. Or alternatively, the injections may have hit two similar but spatially distinct CO recipient compartments. In the cases that had interdigitation of the two tracers, it was likely each injection was placed in a different CO recipient compartment: one that receives the bulk of its input from CO blobs and the other from CO interblobs. CO blobs and interblobs receive different types of input from the parallel processing streams. Cytochrome oxidase blobs receive a direct input from the W cells from the L G N , while the interblobs do not (Boyd and Matsubara, 1996). Cells projecting 72 to 21a and LS are also localized to the CO blobs. Thus, CO blobs are a histological marker for parallel processing streams and anatomical divisions in the cat primary visual cortex. As shown by the large injection in area 19, this is an area that receives input from both CO blobs and interblobs. Also, the efferent output from area 19 to area 21a is segregated into bands. It may be that the function of area 19 is different from other extrastriate areas, such as LS and 21a, in that it is not dedicated to specific function or processing stream but that it acts as a sorter for the different parallel pathways coming in from the CO blobs and interblobs. Primate Comparison Primate V I efferents are directed to different types of modules (stripes) within V2, depending on if the input originated in a CO blob or interblob compartment (Livingstone and Hubel, 1983; Sincich and Horton, 2002a). One of the functions of V 2 is as a sorter and distributor of input from the different CO defined pathways. This may also be the case in cat area 19, which receives input from CO blobs and interblobs and sends input to specialized efferent populations. V 2 also contains specific efferent populations that connect to different extrastriate areas (Felleman et al., 1997; Xiao et al., 1999). For example, MT receives afferents that originate primarily from the thick stripes (DeYoe and Van Essen, 1985; Shipp and Zeki, 1985). While CO has proven to be a good marker for anatomical segregation in V I and V2, there are examples of modular organization in extrastriate areas in the absence of any distinguishing CO staining (Felleman et al., 1997). For example, the prosimian Galigo, does not have CO stripes in V 2 and but the projections to D M are still organized in bands (Beck and Kaas, 1998). 73 Therefore, the lack of CO markers in area 19 of the cat may not preclude the absence of modules. Conclusion As predicted in hypothesis one (pg.29), the results from this experiment demonstrate that area 19 receives input from both the CO blobs and interblob pathways of area 17. Furthermore, the majority of the afferents appear to be spatially localized to different regions within area 19, based upon whether they originated from cell within CO blobs or interblobs. Thus, area 19 has CO predominantly blob recipient compartments and predominantly CO interblob recipient compartments. These findings are important because they distinguish area 19 from the functionally specialized extrastriate areas 21a and LS in that these areas receive a restricted input from the CO blobs. The fact that area 19 is receiving input from multiple pathways indicates that this area may be involved in a more general sort of visual analysis. These findings are also important because they represent the first building blocks of the modular mosaic in area 19. As this afferent segregation according to CO pathway is similar to the V W 2 projection in the primate, it is possible that this feature of the modular mosaic is a representative feature of the first extrastriate visual area in the mammalian species. 74 4. Organization of Intrinsic Connections in Area 19 Introduction The cortex is composed of vertically aligned modules of neurons called cortical columns. These columns extend through layer 1 to layer 6 and share various physiological properties. Cortical columns are comprised of mini-columns, which in primate V I , consist of approximately 80-100 neurons. These mini-columns are bound together into a cortical column by local connections. The fundamental diameter of a cortical column varies from 300 to 1000 /mi, even between species with brains differing 103 in size (for review see Mountcastle, 1997). Long-range horizontal connections send axonal collaterals well beyond the originating cortical column since connections can span 3-4 mm, as in the cat primary visual cortex (Gilbert and Wiesel, 1983; Gilbert 1983; Gilbert and Wiesel, 1085; Gilbert and Wiesel, 1989; Kisvarday et al., 1997). These long-range horizontal connections target other cortical columns resulting in a patchy network of intrinsic connections. While the intrinsic connections of area 19 have never been examined before, the intrinsic connections of areal7 have been extensively studied. Area 17 intrinsic patches extend from the injection site approximately 3-4 mm and are 200-500 pim in diameter. Reports of the center-to-center spacing of these patches range from 1.1 mm to 0.4 mm (Gilbert, 1983; Luhmann et al., 1986; Luhmann et al., 1991; Kisvarday and Eysel, 1992; Kisvarday et al., 1997). These intrinsic patches are found with both retro- and anterograde tracers and when both types of tracers are used in combination the patches 75 tend to overlap, indicating that intrinsic patches form reciprocal connections (Luhmann et al., 1986; Luhmann et al., 1991). However, it was noted that not all the intrinsic patches form reciprocal connections since following injections of antero- and retrograde tracers together some patches contained only one tracer. The patches in area 18 resemble those in area 17 only they have a larger diameter of 500-750 jim (Boyd and Matsubara, 1991) and a spacing of approximately 1.2 mm (Kisvarday et al., 1997). Intrinsic patches in the primary visual cortex tend to be elongated along an axis (Gilbert and Wiesel, 1983; Luhmann et al., 1986). This is likely related to the fact that up to two thirds of patchy pyramidal neurons in layer 2/3 have elongated or anisotropic dendritic trees, while the remaining cells have circular dendritic fields (Thejomayen and Matsubara, 1993). Kisvarday and Eysel (1992) note that the elongation of some of the pyramidal neurons exceeds variations in magnification factor and therefore, anisotropy of intrinsic connections is not solely dependent on magnification factor. There is much speculation about the functional role of intrinsic connections. One possibility, is that intrinsic connections are responsible for synchronized oscillatory responses in cells with non-overlapping receptive fields (Gray et al., 1989; Lbwel and Singer, 1992). Another suggestions is that intrinsic connections are responsible for parleying information between cells of different receptive fields (Albus, 1975; Tusa et al., 1979; Ts'o et al., 1986). A third possibility is that patchy intrinsic connectivity contributes to orientation tuning (Eysel et al., 1987; Matsubara et al., 1987; Eysel et al., 1990). Excitatory intrinsic connections preferentially link orientation domains that share similar orientation preference originate from an iso-orientation domain terminate mainly in regions of cortex that have similar orientation preference (Michalski et al., 1983; 76 Nelson and Frost, 1985; Ts'o et al., 1986; Gilbert and Wiesel, 1989, Kisvarday et al., 1997). Not all studies are in agreement with this finding (Matsubara et al., 1985, 1987), but this may have more to do with the different methods employed in the various studies. While there has been much research and debate on these possibilities, the function of horizontal intrinsic connections remains to be conclusively defined. Intrinsic connections in the primary visual cortex of the primate have been studied in relation to cytochrome oxidase (CO) staining. It has been found that there is a bias with cells from CO blobs preferentially sending collaterals to cells in other CO blobs and vice versa for cells in interblobs (Livingstone and Hubel, 1984a; Lund et al., 1993; Malach et al., 1993; Yoshioka et al., 1996). This was not found to be the case in the tree shrew in which large injections resulted in a pattern of intrinsic label resembling a lattice containing enclosed holes with no label. The lattice did not consistently correspond with the CO staining (Rockland and Lund, 1982) and it was suggested that perhaps the intrinsic patches were related to orientation preference (Mitchison and Crick, 1982). It has since been demonstrated in that, in this species, the intrinsic patches preferentially connect domains with similar orientation preference in which the receptive fields occur along the axis of orientation. This also results in elongated or anisiotropic intrinsic connections (Chisum et al., 2003). In the primate extrastriate cortex the spacing and the spread of the patches increases but the size of the patches remains within the cortical column parameters, varying from 230//m -310/«n (Weller et al., 1984; Amir et al., 1993; Malach et al., 1997). In V2 , intrinsic patches originating in CO rich bands (thick and thin) prefer to send inputs to other CO rich bands. This is not the case for pale stripes, which show no 77 such bias (Levitt et al., 1994a; Malach et al., 1994). These intrinsic patches are smaller than the CO stripes and may constitute a sub-compartment organization to the CO stripes. In the cat extrastriate cortex, there seems to be more variability. An injection in area 7, which is believed to be a multimodal visual area (Thompson et al., 1963), result in patches that were reported to be irregular in shape with obvious organization (Callahan and Haberly, 1987). This is different from the primate V2 , for example in which intrinsic patch shape and diameters are more constant (Lund et al., 1993; Levit et al., 1994a; Malach et al., 1994). Also, Norita et al. (Norita et al., 1996) reports patches of intrinsic label in LS but does not specify the structure of these patches. The exact nature of intrinsic connections beyond the primary visual cortex in the cat remains to be explored. The link between intrinsic connections and preferred orientation in the primary visual cortex demonstrates that these patches have a functional correlate and are a probable source for local signal integration (Angelucci et al., 2002). Understanding the organization of these patches is an essential first step toward eventually determining their correlation with specific functional domains (e.g. iso-orientation domains). Also, defining the components of the modular mosaic in area 19 may provide important clues for establishing the fundamental features of mammalian extrastriate organization. Relatively little is known about either the function or organization of area 19. The purpose of this study is to define the intrinsic connections of area 19 in order to work out the specific anatomical architecture of an extrastriate area. It was hypothesized that intrinsic patches in area 19 would be larger in size and spacing than the intrinsic patches 78 in area 17. It was further hypothesized that these patches would be spatially related to other efferent modules (which are discussed in detail in Chapters 5 and 6). To address this hypothesis, small injections of retrograde tracer were made in area 19 and the intrinsically labeled cells were examined in the tangential plane. 79 Method The procedure for making a small injection in area 19 is the same as detailed in the Methods section of the last chapter. Please refer to it and Chapter 2 for further details. Animal and Surgical Procedure Eight hemispheres from seven normal adult cats were used. Five of these hemispheres were also used for the experiment in Chapter 3. A craniotomy and duratomy were performed at stereotaxic co-ordinates A P -3 to +6 and M L +7 to +15. A single small injection of a retrograde tracer was made in area 19 on the far medial side of the suprasylvian gyrus above the genu. The pipettes were placed perpendicular to the surface of the brain and injections were made at a cortical depth of 500 urn. Amounts ranging from 0.04-1.3 ul of WGA-HRP or CTB were injected. Two different tracers were used to verify that the resulting pattern of labeled cells was not tracer dependent. Histology The animals survived 1-3 days following surgery. The visual cortex was unfolded, flattened, and sliced tangentially. Alternate tangential sections from the superficial half of the cortex were stained for CO using a cobalt and nickel enhancement method (Silverman and Tootell, 1987). The remaining tangential slices, as well as the L G N sections, were reacted for their respective retrograde tracer. The WGA-HRP was visualized using the standard tetramethylbenzidine (TMB) method (Mesulam, 1978). 80 The CTB label was reacted for by immunocytochemistry and visualized by the glucose oxidase-driven diaminobenzidine method (Itoh et al., 1979). Data Analysis The labeled cells in each case were charted with Igor Pro and these charts were used for analysis. The density of labeled cells in patches was revealed using 2D histograms. Density histograms and transects measuring the spacing and spread of labeled clusters in area 19 were also made. The centre-to-centre spacing was determined by measuring the distance from densest pixels in the center of one module to the densest pixels of a second neighboring module. The occasional labeled cell located beyond the furthest cluster was not included in any distance measurements, as it did not belong to defined cluster. Transect measurements that from the most medial patch to the most lateral patch were made to give an estimate of mediolateral spread. Transects that measured the distance between the most anterior and posterior patches were made to give an estimate of anteroposterior spread. Details of transect measurements are outlined in Chapter 2. A cluster analysis was conducted and each case received a CI index to demonstrate the degree of "patchiness". 81 Results Description: A small injection of retrograde tracer in area 19 resulted in a ring dense labeling immediately surrounding the injection site. Further away from the injection site this ring of labeled cells gave way to small densely labeled clusters of cells. Figure 4.1 A is a typical example of intrinsic labeling in area 19. In Figure 4. IB a Gaussian low pass filter was applied to improve visualization of the clusters of labeled cells. The density of the labeled clusters is given in Figure 4.1C. Many clusters of labeled cells were circular in shape (long arrows) but other labeled clusters were larger and more irregularly in structure (short arrow). In all cases, the small circular clusters were concentrated closer to the injection site, while the irregular clusters were more often found at the extremities of the label, though in some cases large clusters were also found adjacent to the injection site. This is the case in Figure 4.2A in which there are large mtrinsic clusters within 2-3 mm of the injection. In this instance, the tracer used was WGA-HRP, which sometimes creates a halo of non-specific label around the injection site, obscuring labeled cells (for a discussion of this halo see Chapter 7, Methodological Considerations). So while the injection site in Figure 4.2A is indicated by the circle, in this case, there is an absence of the typical ring of labeled cells immediately surrounding the injection site. Frequently, large labeled clusters appeared to be an amalgamation of the smaller labeled clusters. Figure 4.3 A , which is an example of such, has a box surrounding a large cluster. Figure 4.3B, a 2D Gaussian low pass filter of the large cluster, reveals a periodic increase in labeling intensity within the cluster. The is demonstrates by the profile plot in Figure 4.3C that results when a transect (dashed line) is drawn through the two dimension 82 Figure 4.1. A typical example of intrinsic labeling in area 19. A ) A cell chart showing the patches of label. The C l value for this case was 1.54. Long arrows indicate a smaller circular patches while a short arrow indicates a large irregular patch. B) A 2D Gaussian low pass filter is applied to the chart in A to help visualize the patches of labeled cells. C) A 2D histogram reveals the density of labeling within the clusters. Notice in this case that the large irregular cluster also has a low density of label. It may be that some of the larger clusters are a collection of smaller clusters but that the low density of label obscures subtle patterns. L = Lateral, A=Anterior. Scale bar = 2mm. The tracer used in this case was C T B . 83 Figure 4.2. A n example of a large intrinsically labeled cluster that are within 2-3 mm of the injection site. A ) A cell chart of cells in area 19 that are intrinsically labeled. The circle at the center denotes the injection site. The outer circle surrounding it represents the halo that accompanies W G A - H R P injections and which obscured any labeled cells immediately surrounding the injection. Once again the long arrow points out the smaller circular cluster of labeled cells while the short arrow point out the larger clusters. Note that the far right arrow points to a cluster that is widely spaced from other clusters. B) A 2D Gaussian low pass filter that shows the relative labeling density of the labeled patches included in A . L = Lateral, A=Anterior. Scale bar = 2mm. The tracer used in this case was W G A - H R P histogram (Fig. 4.3B). One thing to note is that the density of label decreased as distance from the injection site increased, making it more difficult to detect subtle patterns of labeling within the larger more irregular clusters distant from the injection site. Although there were clear clusters of labeled cells, labeling was also found in the inter-patch regions. However, the relative amount of labeled cells in these inner patch regions varied per case. Cases that had the least inner patch labeling also had the fewest labeled cells. A n example of this is shown in Figure 4.4. Note in Figure 4.4B that the clusters contain a low density of labeled cells. There is almost no labeled cells between the injection site and the patches of cells. It is possible that the sparsely labeled interpatch cells were not detected, as the tracer WGA-HRP (which was used in this particular case) sometimes results in weak soma labeling (Kisvarday et al., 1997). Thus, this difference in the amount of inner patch label and the labeling density is likely a reflection of the small differences in the amounts of tracer being injected and sensitivity of different tracers being used. This seems unlikely since Figure 4.3 is a case that also used WGA-HRP and yet in this case, there are a number of labeled cells found between the clusters. 85 B C 1 1 1 1 1 r 0.0 0.5 1.0 1.5 2.0 2.5 mm Figure 4.3 Intr insical ly labeled cel ls i n area 19. A ) In this ce l l chart there is a large patch o f label shown by the short arrow. T h i s patch is irregular i n size and structure. The C I value for this case was 0.96. B ) A 2 D Gauss ian histogram representing the region o f labeled cel ls i n the box in F igure A . In this case the 2 D histogram shows variations w i th in the large labeled patch, suggesting that this large patch may consist o f smaller clusters o f cel ls . A dotted l ine represents a transect that is d rawn through the patch o f label to measure density variations. C ) T h e profile plot that results f rom the transect d rawn i n Figure B . In this case there is per iodic f luctuation i n labe l ing intensity that does suggest that the large labeled patch may be composed o f smaller patches. L=La te ra l , A = A n t e r i o r . Scale bar = 2 m m . T h e tracer used i n this case was W G A - H R P 86 Spacing: To obtain measurements of centre-to-centre spacing the cells plots were converted into a density plot. Transects were made which began at the densest labeled pixel at the centre of one patch and ended at the densest labeled pixel in the centre of a neighbouring patch. The average centre-to-centre spacing of the intrinsically labeled clusters of cells in area 19 was 2.5 mm (the details are listed in Table 4.1). This average took into account both the small circular clusters, which were densely packed around the injection site, as well as the larger clusters located further away. Typically surrounding an injection site and the halo of non-specific cell labeling, there was a zone of dense label, in which most of the clusters were found. Outside of this zone there were one to two isolated patches, located at a further distance from the injection site. These peripheral patches tended to increase the value of the mean centre-to-centre-spacing. One such case is illustrated in Figure 4.2A. The far right arrow points to the patch of label, which had a greater centre-to-centre spacing (5.3mm) than the patches closer to the injection site (avg. 2.3 mm). Cluster Analysis: To quantify the distribution of labeled cells a cluster analysis was performed that took into account nearest neighbour distances (see Chapter 2 for details). The cluster index (CI) value that was obtained for every charted case excluded the injection region and are listed in Table 4.1. The cell charts consisted of up to ten sections collapsed into a single plane thus data from multiple cortical layers was used for this quantification. A cluster analysis revealed well-defined clusters with values ranging from 0.95 to 2.51; a random distribution results in values close to zero. The two cases that resulted in the 87 L Figure 4.4. Isolated clusters of intrinsically labeled cells. A ) A cell chart of labeled cells. Note immediately surrounding the injections site is a dense ring of labeled cells and further away are three isolated clusters of cells. There are virtually no labeled cells in between the halo and the clusters. The CI value for this case is 2.51. B) A 2D histogram showing the density of labeled cells. The clusters have a relatively low density of labeled cells. L=Lateral, A=Anterior. Scale bar = 2mm. The tracer used in this case was W G A - H R P . 88 highest and lowest CI value both used the tracer WGA-HRP, so this difference was not a result of having employed two different types of tracers. Rather, differences in CI value reflected the intensity of inter-patch label, with the highest CI value belonging to a case, which has almost no labeled cells between widely spaced clusters of cells. This case is illustrated in Figure 4.4. The CI value was 1.55, which was significantly greater than random (t-test, p < 0.0001). Lateral Spread of Intrinsic Clusters: The spread of labeled cells was greater in along the anteroposterior cortical axis than along the mediolateral axis. This can be seen in Figures 4.1 - 4.4. The mean spread of patches along this axis was 9.75 mm. The mean spread of labeled patches along the mediolateral cortical axis was 5.79 mm. The anteroposterior spread of the patches tended to correspond with the size of the injection site (see Table 4.1 for details). Table 4.1. Measurements of Intrinsic Clusters for Individual Cases. Case Tracer Inj. Diameter (mm) Center-to-Centre Spacing(mm) Medio-lateral Spread (mm) Anterio-posterior Spread (mm) Cluster Index LHS1 CTB 1.5 2.04 5.2 12.4 0.978 LHS9 CTB 1.3 1.58 4.1 9.8 1.494 RHS7 W G A 0.9 1.7 5.0 6.1 0.957 RHS8* CTB — 2.05 5.3 6.3 1.789 RHS11 CTB 0.7 2.32 7.1 7.8 1.702 RHS5 W G A 1.4 3.25 5.0 12.6 2.518 LHS6 W G A 0.7 3.06 8.4 12.6 2.05 RHS9 CTB 1.3 4.00 6.2 10.4 1.542 * The tissue from case RHS8 was damaged around the injections site so accurate measurements of the diameter of the injection site was not possible. The density of labeled cells decreased further away from the injection site so that the more distant patches had fewer cells. There are several possible reasons for this 89 decrease. The transport of the label itself may play a role in decreasing the density of labeled cells in the long-range patches. The more distant a cell body is from the injection site the more time it takes to actively transport vesicles of tracer to the soma (Kobbert et al., 2000). This means that cells located closer to the injection site will tend to accumulate more tracer in a shorter period of time and thus be more visible. Also, the tracer W G A - H R P is susceptible to fading over time, possibly making the more lightly labeled cells in the distant patches challenging to detect (Matsubara et al., 1987). However, this was unlikely in the present study as cases that used CTB (which is more stable than WGA-HRP) also showed this decrease in density in the long-range patches. Another possibility is that, based on the transport time of the tracer, it would take a longer period of time for the more distant cells to transport the label from the injection site to the cell body (Kobbert et al., 2000). This means that within the three days given for the tracer to transport the cells closer to the injection site would accumulate more tracer than those further away. This decrease in density of labeled cells in distantly located patches was reported previously in both the primate (LeVay, 1988; Levitt et al., 1994a; Malach et al., 1994) and cat (Matsubara et al., 1987; Boyd and Matsubara, 1991). 90 Discussion Identification of Area 19 Area 19 injection placements were verified using several techniques. A schematic flattened tangential view of area 19 and its neighbouring areas are illustrated in Figure 4.5. Sulcal patterns were used to establish the border between areas 19 and 21a (Heath Figure 4.5 A schematic of a flattened tangential view of area 19 and neighbouring areas. Dotted black lines mark area boundaries. Shaded regions represent sulci and the grey line represents the fundus of the sulcus. L S ; lateral suprasylvian area. Lat; lateral sulcus. SS; suprasylvian sulcus. The area locations are based on (Tusa et al., 1978; Palmer et al., 1978; Tusa et al., 1979; Tusa and Palmer, 1980; Sherk, 1986a,b; Grant and Shipp, 1991; Shipp and Grant, 1991) and Jones, 1970; Tusa et al., 1979; Tusa and Palmer, 1980). The likelihood of the injection being placed in area 18 was assessed using alternate sections that were stained for cytochrome oxidase (CO). These C O sections were aligned, using blood vessels, with sections visualized for the retrograde tracer. The border between area 18 and area 19 was delineated by the marked drop in C O staining that occurs in area 19 (Price, 1985a; 91 Boyd and Matsubara, 1996). Lastly, L G N slices were examined to ensure there was labeling in both C1-C3 and in MIN, both of which project to area 19 (Maciewicz, 1975; LeVay and Gilbert, 1976; Hollander and Vanegas, 1977; Dreher et al., 1980; Raczkowski and Rosenquist, 1980; Raczkowski and Rosenquist, 1983; Kawano, 1998). This ruled out that any of the focal injections were placed in any of the three areas which border area 19 around suprasylvian and lateral genu (area 18, 21a and 7). Refer to the discussion in Chapter 3 for details. Size of Intrinsic Connections The spacing of the patches in area 19 was larger than that reported for area 17. In the striate cortex a single injection of a retrograde tracer leads to relatively constantly spaced patches (Luhmann et al., 1986) with a spacing of about 1 mm (Gilbert and Wiesel, 1979; Gilbert and Wiesel, 1983; Martin and Whitteridge, 1984). Area 18 has a slightly larger spacing of 1.2 mm (Kisvarday et al., 1997). This was not the case in area 19. The clusters of labeled cells were not always periodically spaced and the mean centre-to-centre spacing was greater, at about 2.5 mm. This increase was not unexpected as the spacing for intrinsic connections in the primate tends to gets larger in extrastriate areas as the visual hierarchy ascends. In fact, in the primate the periodicity of intrinsic patches increase from 0.61 mm in V I to 1.56 mm in area 7a (Amir et al., 1993). Thus, the increase in area 19 in intrinsic spacing from 1 mm in area 17 to 2.5 mm was outside the range of the different spacing reported for the primate visual cortex. Also, across different cortical areas, increases in the spacing of intrinsic clusters within an area have been correlated with increase in the mean size of the cluster (Lund et al., 1993). 92 A feature of extrastriate intrinsic connections is the range of centre-to-centre spacing found within a single extrastriate area across different cases. This was the case in area 19, in which the spacing ranged from approximately just under 1.58-3.25 mm. This variability in spacing has also been reported for the primate. The range of spacing for primate V 4 anterograde patches is quite large from 450 pirn tol300/<m (Yoshioka et al., 1992). In the tree shrew, extrastriate area V 2 has spacing that range from 0.5-1.0 mm (Lyon et al., 1998). This inconsistency in spacing suggests that perhaps the organization of intrinsic connections in extrastriate areas is not as simple as the regular clusters seen in the striate cortex. The size and structure of the larger intrinsically labeled clusters varied to such an extent that determining the diameter of the clusters was difficult and not particularly useful. Nevertheless, the diameter measurements that were made in area 19 exceeded the diameters found in the primary visual cortex. Within areal7, reports of the intrinsic patch diameter varies from 200/*m- 500/«m (Gilbert, 1983; Luhmann et al., 1986; Luhmann et al., 1991; Kisvarday and Eysel, 1992; Kisvarday et al., 1997) and area 18 has an intrinsic patch width of 500-750 pirn (Boyd and Matsubara, 1991). The irregular intrinsic patches in area 19 bore more resemblance those reported in area 7 (Callahan and Haberly, 1987). In contrast, the relative invariance of intrinsic patch size between different cortical areas has been remarked upon in the primate (Lund et al., 1993). This is different from the large variability found for spacing between different areas in the cat. Different visual areas in the macaque have intrinsic patches that vary from about 200-400 ptm in size (Amir et al., 1993; Lund et al., 1993). 93 s The size of some of the patches in area 19, though, exceeded the size of the cortical column, (300-1000 pirn) (Mountcastle, 1997). One reason for these larger intrinsic patches may be the size of the injection. In macaque V2 , injection diameters that increased from 300/*m to 400 pim resulted in a merging of intrinsically labeled clusters to form elongated patches or irregular stripes (Lund et al., 1993). This may be the case in the present experiment in which the average diameter of the injection sites (excluding the halo of non-specific label) was 1.3 mm. The size of these injections increased the possibility that the injection straddled two cluster systems. However, even in cases with a large injection that saturated a large area 19 there were patches of label similar to those seen in the small injections (Stewart et al., 2003a). Both the spread and the spacing of intrinsic patches resulting from the large injection were within the range of those from cases which small focal injections were made. A similar situation has been reported in the primate, in which larger injections increase the number of patches but with little changes in spread and patch diameter. This suggests that the intrinsic patches are discontinuous so that if two neighboring modules are injected with separate tracers the resulting labeled patches would not abut each other but rather patches from one injection site would be spatially separated from patches originating from the other injection site (Rockland, 1985b; Yoshioka et al., 1992; Amir et al., 1993; Lund et al., 1993). In the striate cortex of the cat, projections can extend up to 6 mm(Gilbert and Wiesel, 1979; Gilbert and Wiesel, 1983; Martin and Whitteridge, 1984). In the present study the mean anteroposterior spread was 9.75 mm and the mediolateral spread was 5.79mm. Figure 4.6 is a schematic of how the different measurements on the patches 94 L A 0 O e — Diameter of Injection Site ^ ™ Centre-to-Centre Spacing Mediolateral Spread Anteroposterior Spread Figure 4.6. A schematic of the measurements of intrinsic patches. The white circle represents the injection site. The grey halo surrounding the injection site represents the non-specific label surrounding the injection site. The elliptical grey patches represent intrinsically labeled patches. The different colour represents the different measurement made. Inj; injection site. A ; anterior. L ; lateral. were made. In the primate, the spread of intrinsic patches is much smaller [V2 = 3.-5 mm (Levitt et al., 1994a; Malach et al., 1994); V4 = 4.0 mm (Yoshioka et al., 1992); MT = 1.8 mm (Malach et al., 1997)]. However, there are reports of intrinsic connections spreading more than 10 mm (Kennedy and Bullier, 1985; Levitt et al., 1994; Stepniewska and Kaas, 1996). This increase in the spread of intrinsic connections in area 19 exceeds the modest increase in receptive field size seen in area 19. The receptive fields extend up to 4 degrees within 20 degrees of central field and regions representing greater than 20 degrees have up to 6.4 degrees receptive field (Duysens et al., 1982a). Yet, far reaching intrinsic patches spread was greater than the equivalent receptive field coverage of area 19 (Tusa et al., 1979; Mulligan and Sherk, 1993). One possibility is that these long range 95 projections provide a mechanism that contributes to surround or contextual effects (Gilbert et al., 1990; Amir et al., 1993). The injections in this study were made in the posterior portion of area 19 just above area 21a. This region contains a representation of the central 10 degrees of visual field (Tusa et al., 1979). The anterior portion of area 19 contains the peripheral visual field representation (Tusa et al., 1979) and received small focal injections in a similar experiment (Stewart et al., 2003a). The pattern of intrinsic label did not vary for the two injection locations, despite the very different retinotopy. The pattern of label as well as the spacing was similar in both cases. However, the spread was slightly further in the anterior injections, which might be explained by this region having larger receptive fields. Nevertheless, this increased distance was not statistically significant. Therefore, retinotopy does not appear to be a prime determinant of the intrinsic labeling pattern. One factor it may influence though is anisotropic labeling. Area 19 forms a thin narrow belt around area 18 and retinotopically is a mirror image of that area (Tusa et al., 1979). Therefore, like area 18 (Matsubara et al., 1987), the intrinsic patches in area 19 spread out in the anteroposterior axis because within area 19 the mediolateral axis is shorter than the anteroposterior axis. This spread of labeled cells only along one axis is referred to as anisotropy. It has been shown that dendritic fields of pyramidal neurons also show this anisotropy (elongation along a single axis). In macaques and tree shrews, as the dendritic spread increases, as you ascend the visual hierarchy from one extrastriate area to the next, so does the patch size (Lund et al., 1993). It would be interesting to determine if this holds true for area 19 as well. 96 Relationship to other modules The spacing and structure of the intrinsic patches was different from the efferent 21a and LS projecting bands found in area 19 (Stewart et al., 2000; see Chapters 5 and 6 for details). The 21a projecting efferent bands had a spacing of 2.5 mm but within these efferent modules there was a fluctuation in labeling intensity every 0.9 mm. These fluctuations were patchy, similar to the small circular intrinsic clusters and the centre-to-centre spacing between the intrinsic modules matches that of the area 21a projecting bands. Also, the mean spacing of the LS projection bands (2.6 mm) was similar to the intrinsic clusters. Thus, the spacing of the intrinsic patches appears to be on the same scale as the 21a and LS efferent bands and may be related to them. This is not the case in primate extrastriate area V4, in which the intrinsic patches have a different size and spacing than the afferent and efferent modules described for this area. One advantage to this independent system is that the intrinsic clusters can sample from a variety of different modules (Yoshioka et al., 1992). However, the irregular centre-to-centre spacing witnessed in area 19 could also hypothetically allow for sampling of more than one efferent band type. In V I and V 2 the intrinsic connections are related to the CO staining so that CO rich compartments preferentially connect to other CO rich compartments (Livingstone and Hubel, 1984a; Lund et al., 1993; Malach et al., 1993; Levitt et al., 1994a; Yoshioka et al., 1996). It can be speculated that an injection into a single band type(e.g. 21a projection band) might result in preferential labeling of other bands of that same type, with the odd patch found in the another band type (e.g. LS projection band). Further experiments that include labeling at least on efferent band type and the intrinsic 97 connections in area 19 are necessary to determine if there is indeed a relationship between the two. Conclusion Intrinsic connections within extrastriate area 19 show a patchy pattern. The characteristics of this patchy pattern further complicate the modular mosaic within area 19. These patches demonstrate an impressive variability in their size, structure and spacing. This variability is not found in the primate and may be specific to the cat. As expected, the spacing and spread of the intrinsic patches is larger than found in the primary visual cortex. Still, the centre-to-centre spacing of the intrinsic patches is very close to that of the 21a and LS efferent modules, suggesting that the two patchy systems may be related. It is important that the contribution of the intrinsic patches to the area 19 modular mosaic illustrate not only a pattern of features that are shared between two species (cat and primate), but also highlight differences between these two species in the degree of variability reported. 98 5. Organization of Efferent Neurons in Area 19: The Projection to Extrastriate Area 21a Introduction Patchy networks are a common feature of cat primary visual cortex (Gilbert and Wiesel, 1983; Luhmann et al., 1986; Voigt et al., 1988; Payne and Siwek, 1991; Dyck and Cynader, 1993; Boyd and Matsubara, 1994a; Galuske and Singer, 1996; Morley et al., 1997; Boyd and Matsubara, 1999; Conway et al., 2000) and some are related to the patchy system of CO blobs (Murphy et al., 1995; Boyd and Matsubara, 1996). Cortical areas beyond primary visual cortex also demonstrate patchy architecture. For instance, area 19 of the cat forms reciprocal and often patchy connections with as many as thirteen other visual areas (Symonds and Rosenquist, 1984b). Area 19 also receives patchy thalamocortical innervation (Kawano, 1998) and the callosal innervation is patchy with elongated and irregularly spaced stripes running in the mediolateral direction (Boyd and Matsubara, 1994a). Projections from area 19 to extrastriate area 21a have also been reported to be patchy in the coronal plane (Symonds and Rosenquist, 1984b; Sherk, 1986a; Dreher et al., 1996c). However, the exact nature of these patches, and their spacing, has not yet been examined in the tangential plane, which is the preferred plane of section for studying variations in labeling density across an entire area. Of interest in this study is whether the pathway from area 19 to area 21a demonstrates a patterned organization. This type of organization is a common feature of parallel processing streams in the cat and may bear a relationship with the functional 99 specialization of area 21a (Payne, 1993). Segregation of inputs to area 21a has already been found in the primary visual cortex where afferents to area 21a originate in the patchy CO blob columns (Conway et al., 2000). Area 21a also demonstrates several physiological features that suggest it is functionally specialized. For instance, most neurons in area 21a are selective for spatial frequency and orientation, but not are not selective for velocity (Toyama et al., 1994; Dreher et al., 1996a; Morley and Vickery, 1997). Based on the numerous functional distinctions of area 21a, it is suggested that this area is a crucial component of a single temporal pathway involved in object recognition (Lomber, 2001). This study examines area 21a inputs arising from area 19. Area 19 itself, is a pivotal visual area with connections to over 15 extrastriate visual areas (Symonds and Rosenquist, 1984b; Ferrer et al., 1992; Boyd and Matsubara, 1994b), indicating it likely plays a role in more than one processing stream in the cat visual system, including the 21a parietal pathway. Identifying the different pathways that are connected to area 19 and the pattern of these connection is vital to defining the modular mosaic within area 19. Defining the characteristics of this modular mosaic may provide important clues for establishing the fundamental features of mammalian extrastriate organization. In this experiment large bulk injections of a retrograde tracer were made in area 21a (as identified stereotaxically) to saturate most of the area. The organization of the efferent projection to area 21a was examined in the tangential plane. Retrograde labeling in area 19 revealed a complex pattern of bands, elongated in the mediolateral direction. Within these bands, clusters of densely labeled cells were found, which suggests the bands are an amalgamation of these clusters. 100 Methods Surgical Procedure Animals received injections of CTX-Au. Following the sterile surgical procedures detailed in Chapter 2, a craniotomy and duratomy were performed at stereotaxic co-ordinates A P -7 to +2 and M L +7 to +15 mm to reveal area 21a. Tracer injections were made just medial to the genu along the posterior portion of the suprasylvian gyrus. Long micropipettes were placed perpendicular to the surface of area 21a and injections were made at a cortical depth of 500 pim with a spacing of approximately 500 nm. The injections covered a 5 mm x 5 mm area and a total of 7-8 ul of tracer was injected. These multiple injections created an aggregate injection that saturated a large portion of 21a area with tracer. Histology The visual cortex was unfolded and flattened tangentially. Sections were sliced and reacted for CO and C T X - A u in a one-in-two series. For cases that also had LS injections (see Chapter 6) a one-in-three series was conducted. Data Analysis Details of the following procedures can be found in Chapter 2. A 2D Gaussian smoothing was created for each case to aid in visualization of the bands. From the resulting image, transects were drawn across the bands, perpendicular to the axis of elongation and the labeling density of the bands was calculated (Figure 5.4). The peaks in the profile plots, generated from the transects perpendicular to elongated axis of the 101 bands, were measured to give an estimate of band spacing (Figure 5.3). Also, to get a measurement of band width, multiple transects, which were perpendicular to the elongated axis of the bands, were made progressing from medial to lateral (e.g. Figure 5.4D). These were then averaged. Transects were also taken along the mediolateral axis of the bands to examine changes in labeling density (e.g. Figure 5.4C). Some of this work has appeared in a previous publication (Stewart et al., 2000). 102 Results Injection Placement Correct placement of injections into area 21a was assessed by several means. Sulcal patterns were used to establish the border between areas 19 and 21a (Heath and Jones, 1970; Tusa et al., 1979; Tusa and Palmer, 1980) and alternate tissue sections stained for C O were used to identify the border between areas 18 and 19, as C O staining Figure 5.1 A tangential section of flattened cortex stained for C O . The area 18/19 border, marked by the thick dashed line, is identified by the edge of a dark C O oval, which encompasses areas 17 and 18. In this section, the genu at the posterior end of the suprasylvian sulcus is darkly stained by C O and is marked by the white arrow. Above the genu the multiple pipette tracts in area 21a are visible. The thin dashed lines correspond to the lateral sulcus (Lat) and the suprasylvian sulcus (SS). A=anterior; L=lateral; LS=lateral suprasylvian area. 103 in area 19 is lighter than that observed for areas 17 and 18 (Figure 5.1) (Boyd and Matsubara, 1994a). The area 18/19 border established the location of area 19. Area 21a was identified by the fact it adjoins area 19 at the posterior crown of the suprasylvian gyrus (Figure 5.1) (Tusa and Palmer, 1980). As corresponding visuotopic areas are commonly connected (Sherk, 1988; Dreher et al., 1996c), injections of retrograde tracer in area 21a, which has a representation limited to the central 20 degrees of visual field (Tusa and Palmer, 1980), would be expected to label regions of central representation in other areas. Cell labeling from area 21a injections was confined to the region of central representation in areas 17 and 18. Also, within area 19, labeling was found in the region of central representation, which is located in the posterior half of this area (Tusa et al., 1979). Furthermore, an extension of the 21a injection into the neighboring lateral suprasylvian (LS) area would have resulted in labeling of the peripheral representation in other areas, since the LS border adjacent to 21a contains peripheral representation (Palmer et al., 1978; Tusa and Palmer, 1980; Mulligan and Sherk, 1993). However, there was no label found in the peripheral representation regions of areas 17,18 and 19, further confirming the injection was restricted to area 21a. Tangential Organization Retrograde labeling of cells revealed that the projection from area 19 to area 21a was organized into discrete, irregular bands. A band was defined as an elongated module interspersed by regions of cortex that had little or no label. In most cases, these bands were elongated roughly in the mediolateral direction, but there was some variation in the Figure 5.2 A typical example of labeling in the visual cortex following a bulk injection of retrograde tracer into area 21a. Areas 17 and 18 have a distinctive patchy pattern. Labeled cells in area 19 form irregular bands that are elongated along the mediolateral axis. Arrows within the box indicated the different modules of label within area 19. M = medial; P = posterior. Scale bar = 5mm. The tracer in this case was CTX-Au. 105 mediolateral orientation of the bands. Figure 5.2 is a typical example of label following a bulk injection of C T X - A u into area 21a. Notice the patches of label in areas 17 and 18 that that are known to align with the CO blobs (Conway et al., 2000). In area 19, the spacing of the labeled modules increased and the modules were shaped into rough bands (indicated by arrows). The bands were found in the posterior half of area 19 and ranged in number from 3-7 (Table 5.1). Variation in band number may have been due to differences in the size of the effective injection site in area 21a. They may also have been a result of the complexity of some of the bands, which formed bridges (arrows in Figure 5.4). This irregular structure occasionally made assigning patches to a specific band difficult. Although the density of labeled cells within bands sometimes fluctuated, the presence of labeling between regions of high density, created visibly discrete bands. Table 5.1 Measurements from separate cases of area 19 efferent bands. Animal No. of No. of Mean Band Mean Band Figure No. Number Charted Bands Spacing Width (mm) Sections (mm) T3 4 4 3.56 0.98 5.5 T7 2 3 2.70 0.73 — T14 2 5 3.25 1.03 5.1 T17 3 6 1.76 0.79 5.3 T18 3 4 2.48 0.95 5.4 S5 1 7 2.04 1.18 5.2 S8 1 5 2.00 1.29 — The first column contains the tracking number of each animal. The second column represents the number of alternate sections that were stained for retrograde labeling and charted to create a single graphical image used for the analysis. The number of area 19 efferent bands found in each animal varied between two and seven, as can be seen from column three. The next two columns represent the mean spacing and width values for the bands found in each case. The figure number corresponding to each individual case is found in the last column. 106 Spacing of Tangential Bands Transects measuring changes in labeling density were used to calculate the mean spacing (See Chapter 2 for details, transects shown in Figure 5.3). The mean spacing of the area 19 tangential bands was 2.5 mm. The standard deviation was relatively high at 0.26, indicating that considerable variation existed (x2 overall). Despite this irregularity in spacing, in no case did the separate bands merge to create uniform labeling across area 19. Area 19 clearly demonstrated fluctuations in labeling density in which areas of low or zero density separated areas of high density. Width and Mediolateral density of Tangential Bands The width and mediolateral density of the bands was also assessed using procedures similar to those used to measure spacing. Examples of the profile plots used to measure the mediolateral density and width of each band are shown in Figures 5.4C and D. The width was defined as the length of a transect, perpendicular to the elongated axis, traveling from the edge of a band where there was zero or near zero labeling density and ending on the opposite edge when the labeling density returned to zero or near zero values. The average band width was 1.0 mm (standard deviation 0.08). The profile plots revealed variations in density, particularly along the mediolateral axis of the bands (Figure 5.4C). Furthermore, after multiple sections were charted, aligned and collapsed onto one plane it was evident that there were patches of increased staining within the band, occurring on average at 1.0 mm intervals. This fluctuation in labeling density was visible to some degree in individual sections. However, clusters of dense areas within the bands were not segregated into individual discrete patches. 107 0 1 2 3 4 length (mm) Figure 5.3 A n example of a cell chart used to measure spacing of the area 19 efferent bands. A ) A cell chart confined to labeled cells in area 19 after an injection of C T X - A u in 21a. The clusters of labeled cells are approximately perpendicular to the area 18/19 border and are elongated along the mediolateral axis. Variation in the orientation of the bands was common between animals. The bands tend to stay perpendicular to the injection site. This is due to the bend in the tissue from the suprasylvian sulcus. B) The same cell chart that has undergone a Gaussian smoothing. Two transects (dashed lines) were made perpendicular to axis of elongation of the bands. C) Profile plots of the change in density of label along the transects. A = anterior; L= lateral; scale bar = 2 mm. 108 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 0 °-5 1 length (mm) length (mm) Figure 5.4 A n example of cell charts used to measure the mediolateral density and width of area 19 efferent bands labeled with C T X - A u . A ) A cell chart of the retrograde labeling in area 19 after injections in area 21a. Notice the branching of some of the bands (small arrows). B) The same cell chart as in A but after a Gaussian smoothing. Three transects along the mediolateral axis of a band are illustrated (1,2 and 3). Three transects along the width of the bands are illustrated (A, B and C). C) The profile plots generated by transects 1,2 and 3 in B. Note the uneven density along the transect. This was a common feature found in all cases. D) The profile plots generated by transects A , B and C in B. The density of labeled cells along the transect peaks in a smooth Gaussian fashion. This pattern of density along the anterio-posterior axis of the bands was found in all cases. A = anterior; L = lateral; scale bar = 2 mm. 109 Rather, within the bands there appeared to be a low level of continuous labeling, with zones of denser label. This indicted the possibility the bands are comprised of closely aligned or partially overlapping patches that are on a smaller scale than the bands. Tangential Organization of Area 18 In the process of examining area 19 it was noticed that area 18 also consistently demonstrated clusters of labeled neurons on the medial side of the posterior half of the area 18/19 border. These clusters of cells were evident in all tangential cases. As seen in Figure 5.5 (short arrow) these clusters were elongated mediolaterally and had an approximate spacing of 0.9 mm. The spacing of these area 18 clusters was similar to the spacing of patches of label in area 17 (0.6-0.9 mm) that appeared after a retrograde tracer injection in area 21a. These patches of label in area 17 are known to co-localize with CO blobs and are a characteristic feature of the striate projections to area 21a (Conway et al., 2000). 110 B Figure 5.5 A chart of labeled cells following an injection of C T X - A u in area 21a. A ) Note the clusters of labeled cells just medial to the area 18/19 border (short arrow). These patches were slightly elongated in the mediolateral direction with spacing of approximately 0.9 mm. The spacing was smaller than the largely spaced bands of label in area 19 that are indicated by the long arrows. The area boundaries are shown by dashed lines. A thin solid line marks a tear in the tissue. B) A 2D Gaussian filter was applied to the cells in area 18. The patches of dense label in this area are visible. The legend on the right specifies the number of cells per pixel. A = anterior; L= lateral; Scale bar = 2 mm i l l Discussion Tangential Organization Efferent neurons in area 19 that project to area 21a are organized in a pattern of irregularly shaped, mediolaterally elongated bands (Figures 5.2-5.5). This pattern of mediolaterally elongated bands may be a general organizational feature of area 19, since LS projecting neurons in area 19 are also arranged in mediolaterally elongated bands (see Chapter 6). Results from double label studies indicate that within area 19, LS projecting bands roughly interdigitate with 21a projecting bands (see the next chapter for details). The spacing between the intrinsic patches (described in Chapter 4) and the 21a bands is the same (2.5 mm). This suggests a the possibility of a correlation between the bands and the intrinsic patches. It may be that when an injection is placed in a 21a efferent band that it preferentially labels intrinsic patches that are located in other 21a projecting bands. It is also possible that the irregularity in spacing between the 21a bands mirrors the irregularity in the spacing of the intrinsic patches. Nevertheless, it may be the case that the irregular spacing of the intrinsic patches is a result of the intrinsic patches being located in both the 21a efferent bands and the LS efferent bands. Further double label experiments are needed to specify the relationship between the intrinsic patches and the area 19 efferent bands. The 21a projecting band pattern in area 19 is more complex than the patchy organization of efferent and callosal neurons found in areas 17 and 18 (Gilbert and Wiesel, 1983; Symonds and Rosenquist, 1984b; Sherk, 1986a; Shipp and Grant, 1991; Ferrer et al., 1992; Boyd and Matsubara, 1999; Conway et al., 2000). One possible reason for the irregular structure of the bands may be non-uniform uptake of retrograde 112 tracer by cells in the injection site. This seems unlikely as the patchy pattern of labeled in area 19 cells is consistent over a large area and between different animals, as well as between different tracers. Also, the patches of staining in area 17 are consistent with other studies using retrograde tracer (Conway et al., 2000), indicating that the large aggregate injections used in this study are uniform. Consequently, differences in labeling density in area 19 are a true reflection of the organizational pattern of 21a inputs and not uneven distribution of tracer at the injection site. Another potential reason for the irregular structure of the bands may be retinotopy. This also seems unlikely, as both 18 and 19 have similar retinotopy and magnification factors (Tusa et al., 1979), yet they demonstrate very different anatomical structures (i.e. patches versus bands). The complexity of the bands may be linked to the thalamocortical connections in area 19. Area 19 receives input from the W-cells of the L G N while areas 17 and 18 receive thalamocortical input from X - , Y - and W-cells (Stone and Dreher, 1973; Dreher et al., 1980). Variations in the type (i.e. X , Y and/or W) of input between these areas may contribute to the differences in efferent organization. Furthermore, W- and Y - cell thalamic inputs terminate in the CO blob columns in area 17 (Boyd and Matsubara, 1996). Since CO blob columns also contain 21a efferent neurons it would be worthwhile to determine if area 19 also receives direct W-cell thalamic input which is restricted to the 21a efferent bands. Projections to area 19 from the C-layers of the L G N terminate in patches in layers 3 and 4 along the coronal plane (Kawano, 1998) and intraocular injections of W G A - H R P result in patchy tangential label in area 19 with an approximate spacing of 2.5 mm (Anderson et al., 1988). Therefore, it may be that 113 termination sites of the L G N afferents contribute to the area 19 efferent bands projecting to area 21a. Comparison to Coronal Reports Patchy columns of cells that project to 21a have been reported previously in the coronal plane (Symonds and Rosenquist, 1984a; Dreher et al., 1996c; Stewart et al., 2000). The area 19 to 21a projection is classed as ascending by Felleman and VanEssen (1991) who based their findings on laminar data from Symonds and Rosenquist's work (Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b). Scanell et al. (1995) collates connectional data from numerous studies and also categorizes this projection as ascending based upon multiple quantitative criteria. Based upon the coronal findings the tangential sections should have shown small patchy labeling, not elongated band labeling. One possible reason for this discrepancy is the T M B reaction product used for the coronal experiments is unstable (Palmer et al., 1978; Zaborszky and Heimer, 1989). Although the tissue underwent a stabilization procedure, some of the lighter stained cells may have missed detection by fading as a consequence of dehydration or time. Primate Comparison Like the cat, extrastriate organization in the primate is more complex than the organization in the primary visual cortex, with extrastriate areas demonstrating greater size and spacing of their periodic structures. For example, in V I there are punctate CO blobs which project to thin CO stripes in V 2 and interblob regions of V I project to the pale and thick stripes in V 2 (Sincich and Horton, 2002a). Even within V 2 stripes, there is 114 some suggestion the bands are composed of multiple smaller patches. For instance, V 2 projections to V 4 originate in patchy compartments within the thin and pale CO stripes (Felleman et al., 1997). Clusters of both MT projecting cells and cells staining for Cat 301 - a monoclonal antibody which tends to label magnocellular structures - are found in the thick CO stripes (DeYoe et al., 1990; Olavarria and Van Essen, 1997). These clusters represent a smaller modular organization within these stripes. Furthermore, rather than being uniform, CO stripes in V 2 consist of irregular aggregates of dark CO patches (Tootell and Hamilton, 1989). The MT-projecting cells within the CO stripes co-localize with patches of dense CO staining within the band (Shipp and Zeki, 1989). Like area 19 of the cat, V 2 has a band organization and these bands may be comprised of multiple aligned patches. Despite some irregularity in the structure of these primate modules, the neural connections between modules in different areas is an important anatomical component of visual parallel processing in this species. This may also be true in the cat, with different periodic structures representing different processing streams. Conclusions Area 19 efferent neurons projecting to area 21a are organized into a series of irregular, mediolaterally elongated bands. The irregular structure and spacing of the bands is consistent with the increased organizational complexity witnessed in other visual extrastriate areas (Payne, 1993) and may be related to the type of input and/or activity processed by area 19. This experiment support hypothesis three (p29) in that the results then establishes the band structure of the 21a efferent neurons. The 21a efferent bands also have the same spacing as the intrinsic clusters and two types of modules may be 115 related. The portion of the hypothesis that looks at the relationship between 21a and LS efferent neurons in area 19 is examined in Chapter 6. The similarity between the V2 stripes in the primate and the areal9 efferent bands suggests that the elongation of modules may be a fundamental mammalian feature of extrastriate organization. 116 6. The Relationship Between 21a and LS Efferent Neurons in Area 19 Introduction One of the most consistent features of the cortex is its modular organization. The type of modular organization varies across different areas. While the mammalian visual cortex is divided up into different areas based upon retinotopy, functional response and connectional characteristics, frequently there can be found a further subdivision based on function and/or anatomy. The modular organization of efferent cells is one such modular subdivision. The organization (e.g. spacing, alignment, size) of efferent cell populations can help differentiate between different visual pathways. Area V 2 of the primate is an excellent example of an area with differentiated efferent modules. The thick CO bands have patches of cells that project to extrastriate area M T (Shipp and Zeki, 1989), which is particularly sensitive to motion (Rodman and Albright, 1989; Stoner and Albright, 1992; Albright and Stoner, 1995; L i et al., 2001) and is included in the dorsal or parietal pathway (Ungerleider and Mishkin, 1982). The thin and pale stripes of V 2 have patches of efferent cells that project to different modules within V 4 (Felleman et al., 1997; Xiao et al., 1999), an area believed to be specialized for object recognition (Gallant et al., 1993; Kobatake and Tanaka, 1994; Gallant et al., 1996; Pasupathy and Connor, 1999) and is part of the ventral or inferotemporal pathway (Ungerleider and Mishkin, 1982). Thus, areas V 4 and MT act as gateways to the two upstream pathways. The differentiation between bands that project to V 4 and bands that 117 project to MT is indicative of the fact that analysis in V 2 is not only performed in parallel, but also affects both the parietal and inferotemporal pathways. This may also be the case in the cat. Area 19 receives the bulk of its input from the primary visual cortex and in turn projects to other extrastriate areas, including LS and 21a (Heath and Jones, 1970; Bullier et al., 1984a; Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b; Price, 1985b; Mulligan and Sherk, 1993). The functional characteristics of LS and 21a are distinct and this has precipitated the conclusion that they are functionally separate areas, possibly analogous to MT and V4 respectively (Payne, 1993;Dreher et al., 1996a; Dreher et al., 1996c). It has also been suggested that 21a and LS are gateway areas to the cat temporal and parietal pathways (Lomber, 2001). In light of the functional distinction between LS and area 21a, it may be that the efferent cells projecting to the two areas are grouped in separate modules within area 19, like V2 . Previous physiological research suggests area 19 is subdivided into functional modules. Response properties of cells within area 19 form two distinct populations (Saito et al. 1988). One group of cells is highly end-stopped with moderate direction and orientation tuning while a second group of cells has strong orientation tuning but is weakly end-stopped (Toyama et al., 1994). Cells within area LS typically display strong end-stopping and direction selectivity (Hubel and Wiesel, 1962; Spear and Baumann, 1975; Smith and Spear, 1979; Yin and Greenwood, 1992; Toyama et al., 1994; Dreher et al., 1996a) while cells within area 21a demonstrate strong orientation tuning but poor direction tuning (Mizobe et al., 1988; Wimborne and Henry, 1992; Dreher et al., 1993; Toyama et al., 1994; Dreher et al., 1996a). It may be that the two physiologically 118 different populations of cells within area 19 are related to the input that area 19 sends to LS and 21a. Saito and colleagues (1988) also divides up cells within area 19 into two physiological categories, similar to those mentioned above, and reports that they occur in columnar patches. It is notable that, in the coronal plane, the efferent (Symonds and Rosenquist, 1984b; Price, 1985b), afferent (Gilbert and Kelly, 1975; Bullier et al., 1984a; Symonds and Rosenquist, 1984b; Price and Blakemore, 1985a; Price and Blakemore, 1985b; Mulligan and Sherk, 1993) and callosal connections of area 19 are patchy (Segraves and Rosenquist, 1982a). Thalamic input to area 19 is also patchy (Kawano, 1998), with elongated patches that are visible following an injection of anterograde tracer in the eye (Anderson et al., 1988). Research presented in this thesis has also shown that area 19 is subdivided into modules. The afferent input from the CO blobs and interblobs is divided up in area 19 and the intrinsic connections form a modular patchy network. Furthermore, efferent 21a projecting cells form a distinctive pattern of rough bands, which appear to comprise multiple patches(Stewart et al., 2000). It may be that the projections to LS are similar to those to 21a and are organized in complementary large bands. The organization of LS projecting cells in area 19 was examined using retrograde tracers. It was hypothesized (as previously stated on pg. 29) that area 19 projections to LS in the tangential plane are organized in a banded pattern and these bands, like the different band types in V2 , would interdigitate with the 21a efferent bands. In order to compare projections in area 19 bulk injections of a tracer were made in LS while large parts of 21a were saturated with another tracer. 119 Methods Animal and Surgical Procedure A l l animals were prepared for surgery as specified in Chapter 2. A craniotomy and duratomy were performed at stereotaxic co-ordinates A P -7 to +6 and M L mm 7-15 to expose both 21a and LS for injections. Bulk injections of either CTB, C T X - A u or WGA-HRP were made in LS. One tracer was injected in LS and a second tracer was used in the 21a injections. The LS injections were placed along the medial bank of the lateral sulcus. The pipettes were angled 30°- 45° from the vertical to follow the slope of the sulcus. Seven to ten penetrations, spaced 500 um apart, were made at a depth of 4-5 mm with tracer being injected at 1 mm intervals along the sulcus. These multiple injections created an aggregate injection that saturated a large portion of area LS with tracer. Approximately 10 uJ of tracer was injected in total. Injections in area 21a were as specified in Chapter 5. Following the injections the craniotomies were packed with saline soaked gelfoam and the overlying fascia and skin was sutured. Histology After perfusion the visual cortex was unfolded and flattened. For the fifteen most superficial sections of the tissue a l-in-3 series was processed in which sections were reacted for either one of the two tracers or for the enzyme CO. The remaining sections were reacted for in a l-in-2 series for the appropriate two tracers. In one animal coronal sections of the visual cortex were cut at 50 pirn on a freezing microtome. Once the sections were reacted, every second slide was counterstained with cresyl violet to help distinguish laminar boundaries. To further verify injection placement, the L G N 120 \ ipsilateral to the injection was also cut coronally and processed for visualization of the appropriate tracer. Data Analysis The position of the labeled cells in area 19 was charted relative to the fiduciary landmarks and entered into the computer for each case. Two- dimensional Gaussian filters were applied to the cell charts and transects were made to measure the spacing of the LS projecting bands (see Chapter 2 for details). To examine the overlap between the 21a and LS projections, a 2D density plot was created for the label of one tracer (e.g. 21a projecting cells) and the charted cells from the other tracer (e.g. the LS projecting cells) were overlaid. Examples of this method are found in Figures 6.5 and 6.6. Cluster analyses were also run. A one dimensional correlation was also performed. Two dimensional cross correlations were not done since the large number of cells (300 000+) resulted in a uniformly black cross correlation. The one dimensional cross correlation compared the density of one type of labeled cell at different distances from another type of labeled cell to establish if the pattern of label from the two types of tracer was correlated (e.g. interdigitated). 121 Results LS Modules Bulk injections of a retrograde tracer in LS resulted in numerous labeled cells in area 19. These labeled cells were clustered into large modules. The modules were elongated along the mediolateral axis forming bands of label that were orthogonal to the area 18/19 border and traversed area 19. Typically, there were several bands of LS projecting cells (as indicated by the arrows in Figure 6.1), but in one case there was only a single band of label that was located in the anterior portion of area 19. In Figure 6.2, the labeled cells were located in the more central region of area 19 and formed up to six bands (indicated by small arrows). The profile plots show mild fluctuations in labeling density along the transect. Also commonly found was a strip of label that traveled parallel to the area 18/19 border and was bisected by this border. The long arrow in Figure 6.2 is points to an example of this strip of label. Comparison of 21a and LS bands The LS projection bands tended to be found more in the anterior half of area 19, as opposed to the 21a projection bands that were localized to a more posterior location. In cases that had 21a labeled cells and LS labeled cells occurring in the same zone, the LS label was more diffusely organized, and hence the banded structures were difficult to identify. In fact the organization of the LS projecting cells in these cases is better described as modules. The structure of the LS bands that occurred posteriorly was more 122 B 1 0 1 2 3 4 0 1 2 3 4 Figure 6.1 A) A cell chart of the typical pattern of labeled cells in area 19 following a bulk injection of retrograde tracer into L S . The arrows are pointing to the bands of labeled cells that are formed. The CI value for the bands in this case was 1.95. B) A Gaussian smoothing was applied to the bands and a transects (dashed lines 1 and 2) have been made along the width of the bands. Differences in labeling density within the bands are visible C) The profile plots from the transects drawn in B . There are changes in labeling density along the transect, however the labeling density does not fall to zero or near zero. Scale bar = 2mm. The tracer used in this case was C T B . 123 Figure 6.2 A ) A cell chart of the L S projecting bands in area 19. The arrows are pointing out 6 different bands in this case. This was the largest number of L S projecting bands found. It was more common to find 3-4 bands. The long arrow is indicating a typical strip of labeled cells that bisected the area 18/19 border. The CI value for bands in area 19 was 2.23. B) A 2D histogram demonstrating the relative labeling density of the different bands. Scale bar = 2mm. The tracer used in this case was C T X - A U 124 irregular and the spacing smaller than for cases with LS bands located more anteriorly in area 19. Examples of more diffuse organization in posteriorly located LS projecting cells can be seen in Figures 6.4 and 6.5. Clustering ofLS and 21a Projection Bands The LS bands clearly showed a clustered distribution. Typically there were regions of cortex between the bands that contained no or few labeled cells. This led to a mean cluster index (CI) for the LS bands that was quite high at 1.26 (the details are listed in Table 6.1). This CI value was significantly greater than for a random distribution (p<0.0002, t-test). The mean CI value for the 21a bands (1.48) was comparable to the CI value for the LS bands. This CI values was also significantly greater than for a random distribution too (p<0.0003, t-test). As was the case with the LS bands, the high definition of the 21a bands interdigitated with regions of unlabeled cortex resulted in this relatively high CI value. Table 6.1 A Breakdown of Measurements for the different cases. Case# No. of Modules Mean Spacing (mm) Cluster Index (CI) SI 1 S3 3 4.50 1.95 S4 3 4.28 2.04 S5* 3 3.70 0.77 S8* 3 1.59 0.86 S9 3 3.40 1.55 T10 6 2.95 2.23 T i l 4 3.54 1.68 T14* 2 2.06 0.87 T17* 3 1.41 0.62 Cases used for comparison of 21a and LS projection bands. 125 B 12 H -5 0 5 10 15 mm c 0 5 10 15 20 Figure 6.3 A ) A cell chart of the L S projecting bands (arrows) in area 19. Note the bands are not evenly spaced. The CI value for this case was 1.67. B) A 2D Gaussian filter was applied to the chart and a transect (dashed line) measuring labeling density was drawn across the figure. The transect is parallel to the area 18/19 border and follows the natural curvature of this border. C) The profile plot from the transect in B . The bands on the posterior side of area 19 were relatively evenly spaced. The band on the anterior side of area 19 was located almost up to 10 mm away. The spacing of this one band increased the mean spacing for this case. M = medial; P = posterior. Scale bar = 2mm. The tracer used in this case was C T B . 126 Spacing ofLS versus 21a Projection Bands The spacing of the LS projecting bands was relatively large with a mean spacing of 2.6 mm. However, there was a large range of different spacing found between individual bands (1.3 mm - 5.5 mm). Even within a single case there was variability in the spacing between different bands. This is shown in Figure 6.3A and B. A transect which is drawn through the different bands illustrates the closer spacing of the bands located more posterior while the most anterior LS projecting band is located almost 10 mm away. However, this variability was not dependent on the location of labeled cells within areal9; there was wide and small spacing of the LS projecting bands found in all parts of area 19. Yet the cases that had more diffusely LS labeled bands in the same cortical region as the 21a labeled cells had a mean spacing that was somewhat smaller at 1.7 mm. The spacing of the 21a bands was quite similar to the overall mean spacing of the LS projecting bands at 2.5 mm (see Chapter 5 for details). Interdigitation ofLS and 21 Projection Bands The 21a projecting bands showed a rough interdigitation with the LS projecting bands. However, this interdigitation was not absolute, despite the similarity in the spacing of the two bands (LS = 2.6 mm; 21a = 2.5 mm). There were regions of substantial overlap (Figures 6.4,6.5,6.6). This may be a result of the more diffusely labeled LS projecting cells in the central region of area 19. This overlap between the LS and 21a bands occurred at points where the labeling density was less. So the densest regions of both types of bands avoided each other. One can see this pattern when a 2D histogram that showed relative labeling density was applied to either the 21a bands or the 127 LS bands. The darker pixels, which represent regions of greater density, have fewer cells from the other band type overlapping than the regions that are less densely labeled. That the most densely labeled regions of the LS and 21a tend to nestle between each other is visible in Figure 6.6. In this figure, a Gaussian filter was applied to both the 21a label and LS label. As in Figure 6.5, the region that shows the most overlap in Figure 6.6 is bordering area 21a (top of the Figure). One-dimensional spatial correlations between the distance from the 21a projecting cells and the density of the LS projecting cells also show the tendency for the two to interdigitate. These are illustrated in Figure 6.7. Notice there is a rise in the number of LS projecting cells at further distances away from the 21a projecting cells. The peak of the increase in the number of LS projecting cells occurs at different distances from the 21a projecting cells, which underscores the variability in spacing between the different cases. 128 Figure 6.4 A) A cell chart of the 21a projecting cells (CTX-Au ,blue dots) and the LS projecting cells (WGA-HRP, red dots). In this case, the 21a projecting bands are evident but the LS labeled cells are more. B)an enlargement of A after a 2D density plot of the chart was made. The 21a labeled cells (blue dots) have been overlaid. Note that the regions of overlap are mostly confined to the pixels that have a lower labeling density (lighter grey). The 21a projecting bands are nestled in between the regions that have denser labeling following a LS injection. Scale bar = 2mm. 129 Figure 6.5 A) A cell chart of labeled cells in area 19 that project to 21a (CTB, white dots) and LS (WGA-HRP, black dots). In this case the 21a and LS labeling is more diffuse. Most of the LS projecting label roughly interdigitates with the 21a projecting cells except in the region bordering 21a. The CI values for this case were 0.87 for the LS labeled cells and 0.84 for the 21a labeled cells. B) A 2D histogram was applied to the 21a labeled cells and the LS labeled cells (red dots) were overlaid. Once again note that the overlap between the LS projecting cells and the 21 projecting cells tend to occur in pixels that are less densely labeled (lighter grey). The exception to this is the region bordering 21a, where there are numerous LS projecting cells, even in regions that are densely labeled by 21a projecting cells. A= anterior; M= medial. Scale bar = 2mm. 130 Figure 6.6 A 2D Gaussian smoothing was applied to both the 21a (CTX-Au labeled appear green) and LS labeled cells (CTB labeled appear red). The regions of overlap are indicated by a yellow colour. In this case as in Figure 6.6 the area that has the most overlap is located at the border of 21a (top of the Figure). The density is indicated by the saturation of colour, so the densest regions appear to have the brightest colour. The regions densely labeled with LS and 21a projecting cells (i.e. colour is highly saturated and appears bright) tended to interdigitate. The CI values for this case were 0.86 for the LS projecting cells and 1.02 for the 21a projecting cells. 131 280-c 260-£ E \ jn "5 240-O 220 -200-o.o 85-i 8 0 -7 5 -70-0.2 1 65 H 6 0 -5 5 -0.4 0.6 distance (mm) 0.8 1.0 r~r-r 0.0 0.2 0.4 0.6 distance (mm) 0.8 1.0 50 H 45 <3 40-35 -4 0.0 0.2 0.4 0.6 distance (mm) 0.8 1.0 F i g u r e 6.7. One-dimensional spatial correlations between distance from the 21a labeled cells and the density of LS labeled cells. Each graph represents a different case. The number of cells from one category (e.g. LS projecting cells) rise as the distances from the cells from the other category (e.g. 21a projecting cells) increases. This demonstrates that despite the regions of overlap there is an overall tendency for the 21a and LS bands to interdigitate. Note that the distance at which the graphs peak differs. This reflects the variable spacing found between cases. The first case corresponds to Figure 6.4, the second case corresponds to Figure 6.5 and the third case corresponds to Figure 6.6. 132 Laminar Pattern ofLS Projecting Cells In one case, coronal sections were cut to examine the laminar distribution of the L S projecting cells. A combination of the laminar staining of C T B and cresyl violet staining for Nissl was used to identify the laminar boundaries. In Figure 6.8, which has been labeled for C T B , notice the high background staining in layer 4. A l l of the labeled cells from an LS injection were found in columnar register in layer 2/3. Labeling sometimes occurred in patches, as is the case in Figure 6.8, but it was also found uniformly spanning area 19 (within layer 2/3). Figure 6.8 also has one labeled cell located in layer 4 (large arrow), however this was unusual. Figure 6.8 A coronal slice of area 19 that was reacted for C T B and then stained with cresyl violet. Staining for C T B label results in high background staining in layer 4. Location of the different layers is indicated on the right. The thin arrows are pointing out labeled cells in layer 2/3. One labeled cell in layer 4 is visible (short arrow). In this case, the labeled cells appear to form two patches in area 19. In other cases, the labeled cells formed a continuous strip of label across layer 2/3. 133 Discussion The cells in area 19 that project to LS were organized in large modules. These modules were similar to the 21a projecting modules in that they were elongated along the mediolateral axis and formed a rough band pattern. Table 6.2 compares the results from the 21a versus LS projection bands. There was also a strip of labeled cells that bisected the area 18/19 border, like labeled cells after a 21a injection. This strip likely corresponded to the cortical representation of the vertical meridian (Tusa et al., 1979). The input that area 19 sends to areas 21a and LS was segregated into modules which roughly interdigitated with overlapping regions. The interdigitation occurred for portions of the bands that were densely labeled. Less densely labeled portions of the 21a and LS efferent bands tended to overlap each other. Area Mean # of Bands Mean Spacing (mm) Mean Cluster Index (CI) Relative Location within Area 19 21a 5 2.5 1.48 posterior L S 3 2.6 1.26 anterior-posterior Methodological Considerations One of the challenges in comparing patterns of label resulting from LS and 21a injections was to have the labeled cells from the two injections occurring in the same region of cortex. The visual field representation in area 21a is restricted to the upper 20 degrees (Tusa and Palmer, 1980). Therefore, injections of retrograde tracer in area 21a only labeled this portion of the central upper visual field in area 19. On the other hand, LS has a complete retinotopic map (Palmer et al., 1978). Labeled cells resulting from injections of retrograde tracer in LS ranged throughout area 19. However, with the 134 substantial loca l and between animal variat ion i n the L S map o f v i sua l space (Grant and Shipp , 1991; M u l l i g a n and Sherk, 1993) it was diff icul t to consistently label a region o f area 19 that contained 21a projecting cel ls . In cases that labeled 21a projecting cel ls and L S projecting cel ls i n the same cor t ical region, the label was restricted to the central port ion o f area 19, medial to 21a. T h e anterior L S bands were identif ied as being i n area 19 o n several basis (the relative locat ion o f the different areas are illustrated i n Figure 6.9). Fi rs t o f a l l , the dark C O ova l that marked the border o f area 18 and area 19 ( B o y d and Matsubara , 1996) served as a useful landmark for p lac ing the media l border o f area 19 and its anterior extent. N o labeled bands occurred anterior to this dark C O o v a l ; this region is somatosensory cortex ( M y a s n i k o v et a l . , 1997). It was also un l ike ly that these patches belonged to area 7 as this area is located lateral to area 19, next to area 21a, meaning a labeled module i n area 7 w o u l d have to be located relat ively far f rom the 18/19 border. Figure 6.9. A schematic o f a flattened tangential section o f v isual cortex. The relative locat ion o f different areas are indicated and their boundaries marked by dashed lines. T h e shaded regions represent su lc i . S S ; suprasylvian. L S ; lateral sulcus. M ; media l . A ; anterior. 135 The LS bands, which had medial borders adjoining the 18/19 border did not demonstrate continuous label that extended into area 7. So the anterior LS bands were identified in area 19. Laminar Distribution of Efferent Cells The laminar distribution of LS projecting cells had some similarities with that of the 21a projecting cells (Stewart et al., 2000). In the case of LS projecting cells, the majority of the efferent cells were found in layers 2/3 in columnar register. This has been reported previously (Symonds and Rosenquist, 1984a; Dreher et al., 1996c; Norita et al., 1996). Dreher et al. (1996a) reported that cells in area 19 that project to 21a and LS have similar laminar distributions and are present in all layers. However, the current study found that LS projecting cells showed no labeling in layers 5 and 6, and almost no labeling in layer 4. This is unlike the 21a projecting cells, which have labeled cells in all layers but layer 1 (Symonds and Rosenquist, 1984a; Dreher et al., 1996c; Stewart et al., 2000). The lack of patchy labeling in the coronal plane for LS projecting cells has already been noted (Sherk, 1986a). The laminar location of efferent cells and/or their axon terminals is frequently used to mark the relative hierarchy of cortical areas. Efferent cells located in the supragranular layers are indicative of an ascending (feed forward) pathway. In other words, the receiving area is considered to be at a higher processing level than the area containing the cells of origin (Rockland and Pandya, 1979; Tigges et al., 1981; Maunsell and Van Essen, 1983). The hierarchy of an area, with a bilaminar distribution of efferent cells found in both supragranular and infragranular layers, is equivocal unless the laminar 136 distribution of the efferent terminals is also known (Maunsell and Van Essen, 1983). In this study, area 19 efferent neurons that project to LS were located in the supragranular layers this means that the area 19 projection to LS is an ascending projection. Both the area 19 to 21a projections and the area 19 to LS projections are classed as ascending by Felleman and VanEssen (1991) who based their findings on laminar data from Symonds and Rosenquist's work (1984 a, b) as well as Scannell et al. (1995) who collated connectional data from numerous studies. Although both of these studies used different criteria to determine hierarchy status they both place 21a higher in the visual hierarchy than LS. Implications for Separate Parallel Pathways The pattern of rough interdigitation with regions of overlap for the LS and 21a projecting cells was reported before in the coronal plane cells (Dreher et al., 1996c). In that study it was also noted that the regions of overlap had very few double labeled cells. This pattern of partial segregation is not specific to area 19 but occurs at the level of the geniculate as well. Within the L G N , the X , Y and W inputs are partially divided so that layers A and A l contain X and Y input, layer C receives Y and W input and layers C l -C3 are composed of W input (Leventhal et al., 1985). The segregation of these pathways extends into the primary visual cortex. The W cell projections from C1-C3 terminate in patches in layer 3 that coincide with CO blobs in the primary visual cortex (LeVay and Gilbert, 1976; Boyd and Matsubara, 1996; Kawano, 1998). Also terminating in the layer 4a blobs are Y terminal patches. Since Layers 4a and 4b receive mixed X and Y terminals from layers A and A l of the L G N , it is difficult to determine if there is any X / Y 137 segregation (Boyd and Matsubara, 1996). A further segregation is the afferent inputs from the CO blob and interblob regions in the striate cortex that project to different compartments within area 19 is suggested (see Chapter 3 for details). This present study suggests that segregation, which is an indicator of parallel pathways, is not limited to the X , Y and W geniculate pathways and area 17 CO blob/interblob pathways but may includes the extrastriate efferent cells as well. The partial segregation of efferent cells in LS and 21a further supports the notion that parallel pathways exist in extrastriate area 19. As the input from the CO blob and interblobs appears to be kept spatially separate in area 19, it would be interesting to see if there was any connection between the afferent input and the efferent bands. For example, if the CO interblob recipient regions corresponded to the LS projecting bands and the CO blob recipient regions were related to the 21a projecting bands, or vice versa. It is interesting to speculate on this because if there is a relationship between the efferent and afferent modules then either LS or 21a is receiving an indirect projection from the interblobs via area 19 and a direct projection from the blobs in area 17. Much of the physiological data supports the notion that LS is a major contributor to the "high-order motion stream" (Dreher, 1986; Spear, 1991) possibly involved in analyzing optic flow (Sherk et al., 1995; Kim et al., 1997; Mulligan et al., 1997; L i et al., 2000; Brosseau-Lachaine et al., 2001; L i et al., 2001). Overall, cells in LS are end-stopped with large receptive fields, a contralateral bias, good direction selectivity and poor orientation tuning . Specifically speaking, the accommodation region of LS is known to receive the bulk of its inputs from area 19 (Maekawa and Ohtsuka, 1993). So the LS bands in area 19 may also be related to lens accommodation responses. Area 21a 138 on the other hand, has been implicated in form analysis. Unlike LS, area 21a cells have an ipsilateral bias and are highly orientation selective but not direction selective (Dreher et al., 1993; Toyama et al., 1994; Dreher et al., 1996c). The segregation may not be limited to function but also found anatomically. Although areas 21a and LS receive input from the same thalamic nuclei and cortical layers, this input originates from two non-overlapping populations of cells (Dreher et al., 1996c). Also, in area 17 the 21a projecting cells found in the CO blobs are sandwiched in between two tiers of LS projecting cells (Conway et al., 2000). Thus, it would seem the segregation of LS and 21a projecting cells is area 19 is no exception. The intrinsic connections of area 19 form patches that are not organized into a band pattern, but this does not mean they are not related to the efferent bands. The spacing of the intrinsic connections on a very similar scale (i.e. intrinsic spacing = 2.5mm; LS spacing = 2.6mm; 21a spacing = 2.5mm). Therefore, it is tempting to speculate that the intrinsic patches that result from a single injection may preferentially label the type of efferent band that received the injection. Future double label studies are needed to determine if this is the case. Primate Comparison Area V 2 of the primate has a well-defined stripe organization that is defined by CO staining. In V2, the efferent cells that project to extrastriate areas V4 and M T are occur in spatially segregated stripes (Shipp and Zeki, 1985,1989). This is similar to area 19, which has segregated efferent populations projecting to 21a and LS. However, the pattern of efferent cells in area 19 is somewhat more complicated in that less densely 139 labeled zones overlap. Within V4 of the primate visual cortex there are modules that receive segregated projections from the thin and pale CO stripes in V 2 (Felleman et al., 1997) but are also organized in a more complex fashion. The modules that receive input from the pale stripes are larger than those that that receive input from the thin stripes. Both types of the modules (thin stripe recipient and pale stripe recipient) form irregular primary foci with several secondary smaller patches of label accompanying the primary Figure 6.10. A schematic of the projections from V 2 to V4 adapted from Xiao et al. (1999). The red represents modules labeled after an injection in a pale stripe. The blue represents modules labeled after an injection in a thin stripe. Each injection produces a primary foci (the 2 large modules) surrounded by secondary foci (the small modules). modules. Overlapping of the two types of modules in V4 only occurs in regions with less labeled cells. The most densely labeled region of the pale stripe recipient module do not overlap with the most densely labeled region of the pale recipient module (Xiao et al., 1999). Despite the fact that the efferent LS bands differs from this scenario, in that there was not one large labeled module accompanied by small patches of label, it shares a similarity in that there is segregation of outputs at the densest part of the band/module and an overlap of outputs in the less densely labeled interband region. Conclusion The results of this experiment show that hypothesis three of this thesis (pg. 29) needs to be amended to include a partial segregation between the LS and 21a projecting bands. The efferent organization of area 19 reinforces a reoccurring theme in cat visual 140 cortical organization: partial segregation with regions of overlap. The LS projecting cells are clustered into a series of mediolateral bands in area 19. The LS projecting bands and 21a projecting bands have similar spacing and clustering values. The 21a bands and LS bands show a rough interdigitation with overlap in regions of less dense label. On the other hand, the partial overlap of label within area 19 also supports the growing evidence that parallel pathways are not totally segregated but contains portions of overlap. 141 7. General Discussion Area 19 receives projections from both the CO blobs and interblobs. In some cases the retrogradely filled cells are more numerous in the CO blobs and in other cases they are more numerous in the CO interblobs. This difference suggests the possibility of different afferent compartments in area 19. The patchy intrinsic connections within area 19 have a much larger longitudinal spread than those found in area 17. The efferent 21a and LS modules within area 19 interdigitate with less densely labeled regions overlapping. The spacing of the intrinsic patches and the efferent bands is very similar suggesting the possibility of a relationship between these modules. Table 7.1 A comparison between the different area 19 modules Module Module Type Mean Spacing Mean Cluster Index (CI) 21a Projecting Band 1.5 1.48 L S Projecting Band 1.6 1.26 Intrinsic Patch 1.5 1.55 Species Justification Cats have traditionally been the species of choice for vision research. Historically a tremendous body of research on this species already exists on which subsequent research can be built upon. Hubel and Wiesel's groundbreaking physiology in 1960s focused on cats (Hubel and Wiesel, 1998). The discovery and subsequent research in the 1970s of the three parallel systems were also made on cats (Stone et al., 1979). In the early 1980s cytochrome oxidase blobs (Horton and Hubel, 1981; Carroll and Wong-Riley, 1984) and stripes (Tootell et al., 1983) were discovered in primates. Blobs were believed to be absent in cats (Wong-Riley and Riley, 1983) until they were discovered a 142 decade later (Murphy et al., 1995; Boyd and Matsubara, 1996). The belief that cats lacked CO blobs, and thus was somehow fundamentally different from the primate, resulted in a shift in research to the primate. However, since that time, research on both animal models continue to provide important comparative information on the visual system In addition to historical reasons to use cats for vision research, there are a number of reasons that make them a practical choice for an animal model. Unlike rodents, cats have an important feature of human and primate vision: frontally placed eyes with binocular overlap. This allows for stereoscopic vision (Barlow et al., 1967). The visual cortex of cats has also proven to be much more comparable to the primate and human than that of the rodent (Malach, 1989; Felleman and Van Essen, 1991; Payne, 1993). Primate research is more costly to conduct than that of cat research. Also, in terms of time efficiency cats are a more prudent choice over primates for developmental studies. Primates have a longer gestation period than cats with only a single offspring (Mitchell, 1989). The critical period that precedes the adult state of the visual cortex is also longer in the primate (LeVay et al., 1980) compared to cats (Cynader et al., 1980; Wilkinson, 1980). Thus, for various practical reasons the cat is a good choice for vision research. The choice of the cat as a model of the visual system does not limit one to the role of simply confirming primate/human visual organization. Instead, the combination of similarities and differences between species allows for questions of evolution to be addressed. While it is beyond the scope of this current study, the use of this species for vision research is important in establishing information that can be used for a truly comparative study of different species. For example, one major difference between 143 primates and cats only have two cones (Hammond, 1978; Loop et al., 1979), while primates have a well-developed system for detecting color (De Valois et al., 1974). Nevertheless, cats have CO blobs. Thus either the original suggestion that CO blobs are for color analysis (Livingstone and Hubel, 1988) is incorrect or response to color is a byproduct of some feature specific to the CO blobs (Allman and Zucker, 1990). Ancestors of these two species were likely nocturnal predators (Cartmill, 1992) so if CO blobs evolved convergently in both species then it maybe that blobs are used for vision in low-level lighting (Allman and Zucker, 1990; Preuss and Kaas, 1996; Preuss, 2000). So although the visual system of the cat may differ from the primate, the cat is still a valid visual model. Differences between species are important from a comparative point of view to help determine species-specific evolutionary processes (Preuss, 2000). Methodological Considerations A contentious issue concerning the visual cortex is the designation of a visual area. A region of cortex is determined to be an individual visual area if it has a retinotopic map, a distinctive function, identifiable architectonic organization and a unique set of connections (Felleman and Van Essen, 1991; Zeki, 2003). Even with this definition of a visual area some extrastriate regions (i.e. the middle suprasylvian sulcus) are difficult to divide up, owing primarily to their redundant distorted retinotopic maps (Palmer et al., 1978; Grant and Shipp, 1991; Mulligan and Sherk, 1993). To add to this confusion, Zeki (2003) recently disputed the existence of several extrastriate areas in the primate, based upon the fact that these areas have only a partial map of visual space. He points out that there are no visual behavioral data to support the idea that certain visual 144 features are perceived only in a single quadrant (i.e. upper or lower visual field) of visual space. Thus he concludes that the existence of areas that represent a limited retinotopic map are "improbable" and closer scrutiny shows these "improbable" areas actually belong to neighboring areas. If true, this contention is problematic for the designation of some cat visual areas 21a and 21b (not for area 19, which has a very well ordered complete retinotopic map). Both of these areas contain an upper visual field representation only. Also, areas 20a and 20b are almost exclusively dedicated to the upper visual field, whereas the visual field representations in lateral suprasylvian areas are concentrated in the lower visual fields (Palmer et al., 1978; Tusa and Palmer, 1980). At first sight Zeki's assertion that areas that contain partial visual field maps belong to neighboring areas supports Sherk's notion that part of 21a should be included in LS (Sherk, 1986a; Sherk, 1986b), thus combining the two areas into a single area that contains both an upper and lower visual field. However, Sherk's hypothesis results in a dissected 21a, which would consequently have even a smaller representation of the upper visual field than the traditionally designated 21a. So Sherk's redefinition of the LS area is not more amenable to Zeki's contention that a visual area must have a complete visual field representation. Nevertheless, it may be that some species, including the cat, have visual areas with partial representations of visual space. Evidence to support this comes from the fact that there exists a substantial amount of physiological and anatomical data that 21a and LS are indeed distinct functional areas (see introduction for details). One possible reason for the discrepancy between the cat and the primate examples Zeki provided is that these examples did not encompass visual areas beyond V4. In the primate many extrastriate areas of the inferotemporal and parietal lobes supposedly 145 contain partial visual field maps and complex retinotopy. This fuels the dispute as to the actual extrastriate area borders (Sereno et al., 1994). While Zeki contends that any visual area that has only a partial representation of visual space is an "improbable" area, the evidence needed to initiate a new parceling of the primate and cat extrastriate areas remains to be worked out. One of the challenges of working with cortex is determining the spatial organization of a structure that is folded up into a pattern of gyri and sulci. Unfolding and flattening the cortical mantle (Olavarria and Van Sluyters, 1985) provides several advantages. Traditionally, neuroanatomists attempted to look at tangential patterns by doing reconstructions of coronal sections. However, as Anderson et al. (1988) pointed out, this took a large number of sections and accurate determination of patterns was difficult. Furthermore, these reconstructions did not allow for estimates of relative labeling density (Shipp and Grant, 1991). Also, in the coronal plane subtle patterns are difficult to visualize. For example, in the cat, CO blobs were not found in the coronal plane (Kageyama and Wong-Riley, 1986a; Kageyama and Wong-Riley, 1986c) and were generally believed not to exist in this species until they were discovered in tangential sections (Murphy et al., 1995). Also, the 21a projection bands in area 19 were not detected in coronal sections (Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b; Dreher et al., 1996c). Looking at tangential sections allows multiple areas to be viewed at the same times. This is convenient for between area comparisons of different anatomical structures. Because of these advantages, most of the data generated in this study came from examining the pattern of labeled cells in the tangential plane. 146 Multiple retrograde tracers (WGA-HRP, C T X - A U , and CTB) were used in this study. The tracer WGA-HRP is a plant lectin and the other two tracers are derived from a bacteria toxin. Both types of these tracers bind to glycoconjugates on the axon terminal (Gerfen and Sawchenko, 1984). They are actively transported (up to 100 mm a day) via vesicles to the cell soma where they accumulate (Kobbert and Thanos, 2000). One of the disadvantages of WGA-HRP is the halo of label that surrounds the effective injection site. However, it has been shown that cells located within the halo have not transported the label. Nevertheless, this halo of label effectively occludes the cells that occur within it, including labeled cells, which have transported tracer from the effective injection site (Mesulam, 1978; Grant and Shipp, 1991). For this reason, WGA-HRP was not generally used to study the intrinsic connections. However, each experiment did use at least two different tracers and they all produced similar results. The tracer CTB did result in high background staining in layer 4 that was likely a result of the immunohistochemical procedure. This may have interfered with the detection of labeled cells, but as the majority of the afferent, intrinsic and efferent cells were located in the supragranular layers, the effect of the high background in layer 4 was negligible. Variability Within Area 19 Both the intrinsic connections and the efferent bands in area 19 showed considerable variability between animals in their spacing and shape. There are several possible reasons for this variability. Extrastriate regions in the cat contain more complex retinotopic maps, larger spacing of connections and more variable patterns of connectivity than the primary visual cortex (Tusa et al., 1979; Sherk, 1986a; Shipp and 147 Grant, 1991). Through statistical evaluation of tracing experiments, Scannell et al. (1999) has shown that variability can be decreased slightly by having a sample size (N) of 10 but that increasing the sample size past this number does not decrease the substantial variation found in these studies. Scannell et al. points out that with such considerable variability it is important to remember that no individual case will match the 'average pattern' of connections. This increased variation in organization is probably established during maturation of the extrastriate areas. Corticocortical connections are frequently established after eye opening, making the organization of these projections in theory more susceptible to individual visual experience (Clarke and Innocenti, 1986; Dehay et al., 1988; MacNeil et al., 1997). Also during development, activity may play a role in determining modular size and organization (Shatz and Stryker, 1988; Horton and Hocking, 1997) and this activity in extrastriate areas may differ from primary visual cortex for two reasons. First, extrastriate areas are more vulnerable to changes in organization since input in these higher areas is altered by previous visual processing in other visual areas. Second, the correlation between incoming inputs decreases in extrastriate regions (Kaas, 1997). Therefore, the increased variation and complexity witnessed in the area 19 intrinsic connections and efferent bands may be the result of an organization based upon individual experience as well as the type of input and processing occurring in this area. Cortical Column The columnar structure of the cortex is a fundamental feature of cortical organization. Many anatomical modules follow this columnar structure. Columnar 148 modules can be broken down into basic units called minicolumns. Minicolumns are a vertical string of cells that extend through layers 2 to 6 and are invariant in size across different species (Mountcastle, 1997). The radial formation of minicolumns has been studied developmentally using neural clones (for review see Rakic, 1995). This is not the case for cortical columns themselves, which vary in size depending on the parameter examined (Lund et al., 2003). Two cells in a single cortical column can share similar preferred orientation and spatial frequency but show quite dissimilar response for temporal parameters. Tanaka (2003) proposes that this grouping of similar features, but not identical ones, allows the visual system to either dismiss differences or amplify them based on the environmental context. Many cortical columns can be identified on anatomical grounds as well. For example, the 21a projection bands in area 19 extend in columnar register through layers 2 to 6 (Stewart et al., 2000). The projection to LS from area 19 was found in columns in layers 2/3 . Despite the prevalence of columns in the cortex, it is important to note that they do not appear to be a necessary feature for function. The rat contains neurons that are as sensitive to orientation as the primate's but these neurons are not grouped into orientation pinwheels, instead preferences are scattered randomly through V I (Girman et al., 1999). It may be that columns are one way of decreasing the metabolic demands of neurons. Grouping cells that have similar function, and thus increased communication, allows for shorter connections that are more efficient both temporally and metabolically (Kaas, 1997). Even i f this is true, grouping cells into cortical columns is a form of segregation and the cortical columns in V I and V 2 are hallmarks of parallel pathways in the primate. In area 19 there is the presence of segregated efferent bands that are columnar in structure (evidenced by the multiple 149 sections that contributed to each cell chart). The LS projecting cells were in columnar register in layers 2/3. The presence of anatomical cortical columns in area 19 is indicative that physiological columns may exist in this area as well. Hierarchy A method for determining placement of an area within the visual hierarchy was established by Maunsell and Van Essen (1983). It was proposed in this model that connections that are feed forward originate in the superficial layers, while projections that are feedback originate in the supergranular layers. Efferent cells that demonstrate a bilaminar pattern (located in the supergranular and infragranular layers) can be either feed forward, lateral or feedback connections depending on their termination pattern. Using data from the present study the projection from area 19 to LS is clearly feed forward as the cells are found primarily in layers 2/3. The cells projecting to 21a are concentrated in layers 2/3 and 5 and 6 (Stewart et al., 2000). Without knowing the laminar pattern of the terminations, this bilateral distribution of 21a projecting cells in area 19 does not provide enough information to asses its hierarchical status. Based upon these rules and using Symonds and Rosenquist's data (Symonds and Rosenquist, 1984a; Symonds and Rosenquist, 1984b), Felleman and Van Essen proposed that area 19 was the third visual area in the cat hierarchy and that 21a was located two levels above area 19. They divided LS up using Palmer et al.'s (1978) stoichiometry so that it occurred on several levels. However, PMLS, which forms a significant portion of LS (Grant and Shipp, 1991) was located on a level between area 19 and 21a. Felleman and Van Essen consulted the projection patterns of thirteen different visual areas and from that data were 150 able to place area 21a despite the ambiguous projection from area 19. Area 17 cells projecting to area 19 on the other hand were located in the CO blobs and interblobs, and thus were found in the superficial layers making this a feed forward connection. Symonds and Rosenquist (1984b) have reported the same laminar locations of efferent 17 and 19 cells. The hierarchical organization, determined according to data from the present study, places 19 between the initial processing areas 17 and 18 and a higher up processing area, LS. The relative position of area 21a remains to be defined. Scannell et al. (1999) has also looked at the hierarchy of the visual system in the cat. Scannel et al. determined that one of the main factors for determining cortical hierarchy was a simple wiring rule. It is assumed that the cortical location of areas are arranged to minimize the volume of connections, so that areas which are conducting related visual analysis are physically located close to each other. Using non-parametric cluster analysis they place areas 19 and 21a on a level above the primary visual cortex and LS on the level above that. Figure 7.1 compares the different hierarchy schemes. It Figure 7.1. A schematic of the different proposed hierarchy for areas 17, 19, 21 and LS. In the present study the location of area 21a remains to be defined. Determining the laminar location of the 21a afferents would place area 21a in the hierarchy developed by Maunsell and Van Essen (1983). is interesting that area 19 and 21a are placed on similar levels as 21a is a functionally defined area involved in form processing and which receives very specific inputs, while area 19 receives a soft patterned input from the primary visual cortex and may be involved in global redistribution of information. Scannel et al. suggests that Felleman 151 LS 19 17 21a 21a Present Study LSI J21a 19 \1U Felleman and Van Essen 21a LS 19 17 Scannel et al. and Van Essen's model is not ideal for the cat, as some connections do not fit his criteria for determining hierarchical position. For example the input to area 19 arises from the superficial layers of areas 17 and terminates in the superficial layers of area 19 (Symonds and Rosenquist, 1984a; Price and Zumbroich, 1989). This pattern of connections is not dealt with in Felleman and Van Essen's model. It may be that the connectional rules used to determine hierarchy in the cat visual system might differ from the primate and still remain to be defined. Parallel Pathways Parallel pathways were first discovered in the cat (Enroth-Cugell and Robson, 1966). Since that time the contributions of the X , Y and W pathways to area 17 have been worked out. The W cells of the L G N input directly into layer 3 blobs whereas the Y terminals from layer C terminate in layer 4a blobs. The contribution of the X projections to the layer 4a blobs and interblobs remains to be determined (Boyd and Matsubara, 1996). So like the primate, the CO blobs represent a divisible input of the parallel processing pathways. Since area 19 receives an input from the CO blobs that is spatially segregated from the terminals of cells that are in the CO interblobs, the continuation of the parallel pathways in this extrastriate area is highly likely. The partial segregation of the 21a and LS efferent bands to two functionally distinct areas is suggestive of the fact that parallel pathways may be traveling through area 19. Further research is necessary to tie the afferent input to the efferent modules before any definitive link in area 19 to parallel pathways is made. Also, the segregation between the two types of afferent bands is not absolute. The partial overlap between the 152 two types of efferent bands is consistent with the increasing amount of data on the partial convergence between the retino-geniculo-cortical pathways in the primate (Malpeli et al., 1981; Ferrera et al., 1994; Nealey and Maunsell, 1994; Yoshioka et al., 1994). Even the traditional view of the parvocellular and magnocellular inputs (Livingstone and Hubel, 1988) into the different CO compartments is being questioned. It is now known that V I interblobs provide the major input to both pale and thick stripes (Sincich and Horton, 2002a) and that the different CO stripe compartments in V 2 are interconnected (Levitt et al., 1994a; Malach et al., 1994). This convergence suggests that the parallel pathways of the extrastriate areas are not defined simply as magnocellular and parvocellular or Y and X , as they are in the geniculate, but rather the parallel pathways represent some other more complex functional division. For example, based on the motion sensitivity of area LS it has been assumed that this area receives its principal input from the Y pathway (Berson, 1985; Rauschecker et al., 1987). However, when the Y-channel is pressure blocked, LS cells show normal receptive field properties with only a small deficit in responding to fast stimuli (Wang et al., 1997). It would be more accurate to say that LS is part of a motion sensitive pathway and 21a is included in an object recognition pathway rather than to tie the pathways to the Y and X channels respectively. The results from chapter 3 demonstrate that area 19 is receiving input from both the CO blob and interblob pathways. Also, the partial segregation of efferent bands in area 19 suggests a specialized output to area 21a and LS. While the connection between the afferent and efferent modules remains to be worked out the presence of segregation suggests the presence of parallel pathways in area 19. 153 Function of Area 19 The classic Aristotelian belief that form follows function allows for functional propositions to be forwarded based on the anatomical make up of an area. After all, it is logical that the function of an area would depend upon the information it receives. The physiological studies that have been done on area 19 have not been able to determine a single function for this area. Based upon the fact that area 19 has similar receptive field properties as area 17 and 18 it, has been suggested that area 19 is performing a redundant analysis (Kimura et al., 1980; Duysens et al., 1982a; Bergeron et al., 1998). The different connectional properties and modular organization of area 19 argue against this hypothesis. It could simply be that the physiological studies that looked at neuronal response properties in area 19 did not use appropriate stimuli. It was recently shown that complex shapes (e.g. arcs, angles) could elicit a much greater response from cells in V2 than simple stimuli like oriented bars and sinusoidal gratings (Hegde and Van Essen, 2000). Perhaps area 19 is also selective for complex stimuli. Lesion studies suggest a dual role for area 19 in both form (Doty, 1971; Sprague et al., 1977; Hughes and Sprague, 1986; Kruger et al., 1988) and motion analysis (Dinse and Kruger, 1990). The partial segregation of area 19 efferents to LS (a motion sensitive area) and 21a (an object recognition area) support this assertion. Based upon findings from the present study that area 19 receives both CO blob and interblob input into spatial distinct compartments and sends input to 21a and LS in a semi-segregated fashion it appears that area 19 may work as a distribution center, possibly sorting form and motion information. 154 Species Comparison The notion that fundamental aspects of visual cortical structure are shared by mammalian species underlies the animal-model based approach to visual systems (Preuss, 2000). And indeed there have been many similarities found between the primate and cat species to support this. Diagrams in Figure 7.2 and 7.3 outline the primate and cat cortical organization for comparison. Based upon similarities, homologies between the different visual areas have been cited, foremost being the striate cortex. The primary visual cortex of both primate and cat receive input from three different processing streams (Norton and Casagrande, 1982; Irvin et al., 1986; Norton et al., 1988; Casagrande, 1994; Freund et al.,1985; Humphery et al., 1985a,b; Boyd and Matsubara, 1996; Kawano, 1998; Hendry and Reid, 2000), have CO blobs (Horton and Hubel, 1981; Horton, 1984; Murphy et al., 1995; Boyd and Matsubara, 1996), orientation pinwheels (Bonhoeffer and Grinvald, 1991; Bartfeld and Grinvald, 1992) ocular dominance columns (Anderson et al., 1988; Levay et al., 1975; Florence and Kaas, 1992) and show many similar visual field properties (for review see Payne, 1993). Therefore, primate V I and cat area 17 resemble each other on a variety of different physiological and anatomical levels and are frequently referred to as homologous structures. Nevertheless, there are also major visual differences between these species (e.g. colour vision) which should not be ignored. These differences are useful from a comparative point of view for looking at the evolution of species-specific adaptations. Thus, establishing homologies is not a clear-cut process. A variety of different analytical approaches (i.e. molecular, physiological, anatomical) should be conducted and used as criteria to confirm a true homology between areas. It is also argued that a broader cladistic approach that 155 examines multiple species (rather than just cat and primate) is also necessary to determine homology (Harvey and Pagel, 1991; Kaas and Lyon, 2001). As the present study is focused on the anatomy of area 19 of a single species, rather than make claims of homology I will take a comparative approach and suggest possible analogous structures in the primate. Primate V 2 and cat area 18 are both the second areas in their respective visual hierarchies and are historically considered homologues for each other (Payne, 1993). Nonetheless, there are reasons to dispute this assertion. The foremost is that area 18 receives a substantial geniculate input (Garey and Powell, 1971; Gilbert and Kelly, 1975; LeVay and Gilbert, 1976; Hollander and Vanegas, 1977) while V2 does not (Benevento and Yoshida, 1981; Fries, 1981; Yukie and Iwai, 1981; Bullier and Kennedy, 1983). Second, the patchy structure of anatomical and functional modules in area 18 does not differ significantly from area 17; the patches simply occur on a larger scale (Anderson et al., 1988; Boyd and Matsubara, 1996; Shmuel and Grinvald, 1996). On the other hand, the modular structure of V 2 (bands) (Tootell and Hamilton, 1989) differs significantly from that of V I (patches) (Horton, 1984). Thus, the case for homology between area 18 and V 2 is not strongly supported. In fact, area 19 shows more similarities to V 2 than area 18. For example, within the visual hierarchy both primate V 2 and cat area 19 are the first extrastriate areas encountered (Felleman and Van Essen, 1991). These areas are considered extrastriate areas since they receive a substantially reduced input from thalamus, compared to the primary visual cortex (Benevento and Yoshida, 1981; Dreher et al., 1980). Both areas segregate input from CO blob and interblob columns into separate compartments (Livingstone and Hubel, 1987; Sincich and Horton, 2002a). 156 I (Felleman etal . , 1997) (Mausell and >Jf >|rvan Essen, 1983) Inferotemporal Cortex Parietal Cortex The "What" Pathway The "Where" Pathway Figure 7.2 A schematic of parallel pathway organization in the primate. The shaded regions represent C O rich pathway. Input that originate from cells in layers 2 /3 ,4A and 4 B of the C O blob columns terminates in the thin C O stripes of V 2 . Cells in layer 2 /3 ,4A and 4B that are part of the interblob column terminate in both the thick and pale C O stripes of V 2 . The thin and pale stripes project to different modules within V 4 that partially overlap. The thick stripes project to M T . In turn, area V 4 and M T project to the inferotemporal and parietal processing streams respectively. 157 Intrinsic patches in V 2 are found in all three CO band compartments, but unlike the intrinsic patches of area 19, they have a smaller spacing than the V2 stripes (Levitt et al., 1994a; Malach et al., 1994). The banded pattern itself in V 2 is similar to the structure of the efferent bands in area 19; both are orthogonal to their respective borders with the primary visual cortex (Tootell and Hamilton, 1989). Area V 2 has a patchy substructure within the CO bands (Ts'o et al., 2001) similar to the patchy substructure seen in the 21a projecting bands in area 19. Despite these similarities, V 2 and area 19 are not absolutely comparable. The biggest difference between these two areas is the absence of CO bands in area 19. However, this absence is also noted in V 2 of several prosimian primates (Condo and Casagrande, 1990; Krubitzer and Kaas, 1990; Preuss et al., 1993; Rosa et al., 1997) and is not considered a criterion for establishing V 2 in the primate. Another difference is that area 19 has a larger thalamic input than V2. This input is enough to sustain activity in area 19 following lesions to area 17 and 18 (Dinse and Kruger, 1990), while lesions to V I effectively silences V 2 (Schiller and Malpeli, 1977; Girard and Bullier, 1989). Furthermore, while V 2 has functionally grouped cells that align with the CO bands (Hubel and Livingstone, 1987; Ts'o et al., 1990; Levitt et al., 1994b; Malach et al., 1994; Roe and Ts'o, 1995; Roe and Ts'o, 1999) no such grouping has been reported for area 19. Yet, the absence of such reports actually stems from the fact that no systematic study of the spatial distribution of function has been conducted in area 19. Functionally distinct classes of cells in area 19 have been reported but their spatial distribution was not ascertained (Saito et al., 1988; Toyama et al., 1994). So in summary, area 19 does show a number of similar characteristics as primate area V2. However, these similar characteristics can also be viewed as general extrastriate features 158 Primary Visual Cortex Cat Cortical Pathways Area 19 Temporal Cortex "Form" Pathway Parietal Cortex "Motion" Pathway Figure 7.3 A schematic of the organization of the parallel pathways that travel through area 19 of the cat visual cortex. While the relationship between the different C O recipient afferent modules and the L S and 21a efferent modules is not yet determined, it is interesting to speculate that the modules are related so that the 21a bands are receiving C O blob input and the L S bands are receiving C O interblob input, or vice versa. Further studies are needed to determine the exact nature of the relationship between the afferent and efferent modules in area 19. 159 (segregated input from CO compartments, enlarged, extended banded modules, patchy intrinsic connections), which can be found in other primate extrastriate structures such as V 4 - see below. Thus, rather than conclude that area 19 is an analogous structure to V 2 all these similarities can enforce is that there are common features to extrastriate organization. Area V3 of the primate has not been extensively studied owing to the contentious debate over the actual location and retinotopic organization of this area (Kaas and Lyon, 2001). Area V3 is smaller than V 2 and the center of area V3 is discontinuous so that it is actually composed of two segments (Lyon and Kaas, 2002). Area 19 was traditionally believed to be homologous to V3 simply based on its position in the cortical mantle (Payne, 1993). Area V3 does share some similarities with area 19 but these similarities are common to V 2 as well. For example, all three areas show greater variability in the lower visual field maps (Tusa et al., 1979; Gattass et al., 1981; Gattas et al., 1988). A l l three areas send input to two functionally distinct areas (in the primate MT and V4; in the cat 21a and LS) (Zeki and Shipp, 1988; Felleman et al., 1997). A l l three areas lack ocular dominance columns (Tieman and Tumosa, 1983; Friedman et al., 1989; Tootell and Hamilton, 1989; Ts'o et al., 1990). So while on the surface these similar characteristics support the homology between area 19 and V3 , the general nature of these characteristics along with the unusual retinotopy of V3 really does not argue for an area specific homology. In fact Kaas and Lyon (2001) describe the naming of V3 as "more confusing than descriptive" since it naturally leads to a comparison with VIII of the cat, or area 19. It may be that the cat does not have an analogous structure to V3 as some prosimians do not even have a V3. After all, it is to be expected that not all of the areas 160 found in the primate have homologies in the cat as the cat has fifteen visual areas (Symonds and Rosenquist, 1984b) while the primate has thirty-one (Felleman and Van Essen, 1991). Primate area V 4 is unlikely to be a candidate for an area 19 analogy. This is because V4 is a specialized area for form processing (Gallant et al., 1993; Kobatake and Tanaka, 1994; Gallant et al., 1996; Pasupathy and Connor, 1999), where as the physiological properties of area 19 along with its continuous input from areas 17 and 18 suggest that it is involved in more general analysis. Nevertheless, V 4 is useful for comparison when examining the structure of the afferent modules of this area. The thin and pale stripes of V 2 send input to different modules in V 4 (see Figure 7.2; Felleman et al., 1997). These modules are more irregular in shape with partial overlap occurring at the borders. These overlapping regions occur at points where the labeling density of the modules is decreased. Thus, like the efferent LS and 21a modules in area 19 there is a pattern of segregation with regions of convergence (Xiao et al., 1999). Further research comparing the different modules of extrastriate areas may find this to be an inherent feature of extrastriate cortex. Two other species whose visual cortices have been researched are the rat and the ferret. The number of visual areas in the rat and mouse has long been debated (Montero et al., 1973; Olavarria and Montero, 1989; Malach, 1989; Coogan and Burkhalter, 1993; Montero, 1993; Rumberger et al., 2001). But because the most fundamental visual area, V I , does not resemble the primary visual cortex of the cat or primate, in that it shows no discernible functional or columnar organization (Girman et al., 1999) and no patchy intrinsic connections (Malach, 1989; Rumberger et al., 2001), it is unlikely that another 161 visual area would show a resemblance to area 19. The ferret on the other hand is more closely related to the cat. The border between areas 18 and 19 in the ferret is marked by a dramatic decrease in CO staining, just as in the cat. Also, callosal connections in area 19 of the ferret and cat are both patchy (Innocenti et al., 2002). The cortical location and retinotopic organization of areas 17,18 and 19 in ferret strongly suggests that area 19 is homologous in the ferret and cat (Manger et al., 2002a; Manger et al., 2002b). However, research on the anatomy of the extrastriate areas of this species is just emerging and the information available is not adequate enough to draw any parallels between this study and ferret area 19. Future Exploration There are several questions that arise out of this study, which need to be addressed. The first is identifying the afferent recipient compartments in area 19. The research presented here demonstrates that within area 19 the projections from the CO blobs terminate in spatially separate regions from the projection of the CO interblobs. However, the CO blob recipient compartments and the CO interblob recipient compartments need to be identified and characterized. Are they the same size and structure of the 21a and LS efferent bands? Do they form interdigitating patches? One way to approach this question would be to make a very small focal injection with an anterograde tracer, such as biocytin, in area 17 that would be localized to a single blob or interblob. This is possible as the center-to-center spacing of the blobs is just under 1 mm (Murphy et al., 1995; Boyd and Matsubara, 1996; Murphy et al., 2001) and small injections of biocytin can have a radius as small as 200 pm (Lund et al., 1993). However, 162 the necessity of making an injection small enough that it is limited to a single CO blob or interblob means that only a small region of area 19 will be labeled. Thus the extent and boundaries of the modules would still remain undefined. Using two different tracers such as biotinylated dextran amine (BDA) and Phaseolus-Vulgaris Leukoagglutinin (PHA-L) (Veenman et al., 1992) would reveal the relationship between a CO blob recipient compartments and a CO interblob recipient compartments. Also, it would be worth while to make a large injection or multiple small injections in area 17 with an anterograde tracer to determine if all of area 19 receives projections from area 17 or are there modules in area 19 that don't receive a projection. The relationship between the afferent and efferent modules also still remains to be worked out. For example, it would be interesting to determine if the CO blob recipient modules overlap with a single type of efferent band. The best approach to look at this would be to combine retrograde and anterograde tracers. Thus, a retrograde tracer would be injected into either 21a or LS to label efferent bands and an anterograde tracer injected into area 17 to label the blob or interblob projection. Alternatively a small injection of a mixture of an anterograde and retrograde tracer in area 19 would reveal the if that region of cortex received input specific to CO blob or interblob and what area the efferents projected too. This would also allow for the intrinsic patches to be linked to a specific efferent band and compared. For example, if an injection in area 19 labeled afferents primarily in 21a , then it would be possible to compare the intrinsic patches in that case to a case that had afferent labeling in LS. Also, a mixed anterograde/retrograde injection would determine the relationship between projecting and terminating intrinsic clusters in area 19. 163 Both 21a and LS receive a direct projection from the CO blobs, albeit from different cell populations within the blobs (Boyd and Matsubara, 1999; Conway et al., 2000). It would also be interesting to determine if one or both of the extrastriate areas receives an indirect projection from the CO interblobs via the CO interblob recipient compartments in area 19. However, as there is a degree of overlap even with the 21a and LS projecting bands, it is unlikely that the efferent compartments would co-localize absolutely with one type of CO recipient compartment. The relationship between the 21a and LS projection bands in area 19 could also be further explored. A double label study, in which LS and 21a were injected with different tracers and area 19 examined for double labeled cells. This would quantify the percentage of cells that project to both LS and 32a. Alternatively, an injection of anterograde tracer in area 19 would demonstrate the projections to LS and 21a and to what extent there are projections to both areas. This experiment would also give us information about where area 21a stands in the cortical hierarchy by defining the laminar terminations of the area 19 projecting cells. One possible way of further defining the modular mosaic of area 19 would be to examine the spatial distribution of the afferent and efferent connections between the pulvinar and superior colliculus to area 19. In the primate, the pulvinar projections are centered on the thick and thin CO stripes in V2. These projections are found concentrated in layer 3, which is also the layer that contains the strongest CO activity (Levitt et al., 1995). It has recently been put forward that in V 2 it is the pulvinar input which defines the cycles of CO bands (Shipp and Zeki, 2002; Sincich and Horton, 2002b). A further distinction between the CO bands of V 2 can be made by the fact that 164 cells that project to the superior colliculus are concentrated in the thick CO stripes (Abel et al., 1997). The link between superior colliculus and dense CO staining is indirectly bolstered by the fact that the K layers of the L G N , which input exclusively into the CO blobs, are the only L G N laminae innervated by the superior colliculus (Halting et al., 1978; Halting et al., 1991). Area 19 shows many similarities with V 2 so it may be that either pulvinar afferents or efferent cells projecting to the superior colliculus can also define the different modules in area 19. It has already been established that area 19 receives a pulvinar projection (Graybiel, 1972; Symonds et al., 1981; Raczkowski and Rosenquist, 1983), as well as sends terminals to the superior colliculus (Updyke, 1977; Halting et al., 1992). A l l visual cortical areas are visible in a single tangential section following the unfolding and flattening of the visual cortex. It was obvious that retrograde tracer injections into areas 19, LS or 21a labeled a number of extrastriate areas (for illustration see Figure 5.2). One such example is the consistent labeling of area 7 following bulk injections in 21a. This was also noted by Grant and Shipp (1991). This labeling of multiple extrastriate areas was expected. It was previously reported that area 19 receives projections from seven other extrastriate areas, area 21a receive projections from four other extrastriate areas and PMLS (the portion of LS usually injected in the present study) receives input from six extrastriate areas (Symonds and Rosenquist, 1984b). However, the details and tangential organization of these areas remains to be worked out. As almost nothing is known of these extrastriate areas, electrophysiological verification of the exact location and boundaries of some of these areas will be necessary. 165 It is well established that the layers of the cortex have distinct architecture, connections and physiological responses (Lund et al., 2003). So it is not surprising that studies have begun to look for functional segregation within the different cortical layers. Recent studies of the effects of cooling in LS have shown that cooling the superficial layers results in motion detection deficits. Cooling the superficial and infragrandular layers results in deficits in motion detection and attention (Lomber and Payne, 2000; Lomber, 2001). Raizada and Grossberg ( 2003) propose that feedback circuits from layers 4 to 6 of the visual cortex largely control attention and attentive modulation. A reexamination of functional differences in V 2 lead to Shipp and Zeki (2002) proposing that segregation and specialized pathways were a feature of layers 3 to 5, while integrated feedback and modulation were specific to layers 1, 2 and 6. This new direction of research would be interesting to explore in area 19. Are there functional differences between layers that are specific to area 19? Determining the different laminar contributions will be necessary to creating a complete model of area 19. Future physiological studies that examine the spatial distribution of functional characteristics and their correlation with parallel processing streams and the anatomical modules in area 19 are needed. There are several experiments that hint at a functional division within area 19. Both Saito et al. (1988) and Toyama et al. (1994) report finding two physiologically different classes of cells, one for processing motion of discontinuities and the other was suited for determining the orientation of continuous contours. It would be interesting to see if the population of cells that responds to motion of discontinuous elements is concentrated in the LS projecting bands, and if the population of cells that processes orientation of continuous contours overlaps with the 21a projecting bands. A n 166 ideal method for determining grouping of function is optical imaging of intrinsic signals. Optical imaging detects differences in cortical reflectance following metabolic changes that result from local neural activity (Frostig et al., 1990). The posterior portion of area 19 is exposed on the suprasylvian gyrus making it ideal for imaging and optical imaging can penetrate the cortex up to 1000 jim (Shoham et al., 1997). As terminals from the primary visual cortex (see figure 7 in Symonds and Rosenquist, 1984a) and the 21a and LS efferent bands are located in layers 2/3, any functional differences within them would be visible with optical imaging. However, when orientation pinwheels were imaged in area 21a no signals were mapped in area 19 (Ohki et al., 1998). The primary reason for this is likely due to the sluggish response of neurons in area 19 under anesthesia (Feldon et al., 1978; Saito et al., 1988; Sherk, 1990). One possibility would be to do optical imaging with kittens and young adult cats as they tend to give the best optically imaged maps (Godecke et al., 1997). However, Feldon et al., even goes so far as to suggest that barring infusions with caffeine, the only way to reliably get a response from area 19 neurons is to record in awake behaving animals. Despite this assertion, there are studies, which have been able to do extensive extracellular recording from 19, without caffeine (Guillemot et al., 1993a; Guillemot et al., 1993b; Tardif et al., 1996; Bergeron et al., 1998). Thus, a systematic mapping of specific physiological traits should be possible. 167 8. Summary Extrastriate area 19 of the cat visual cortex is divided into different modules that form a complex modular mosaic. A schematic of this modular mosaic is illustrated in Figure 8.1. Although the relationship between all the afferent, intrinsic and efferent connections of area remain to be worked out, this study defined the type of organization found in area 19. The major findings from this study are listed below. • The afferent inputs from area 17 to area 19 are soft patterned; the cells from both the CO blobs and interblobs project to area 19. • Within area 17, area 19 projecting cells tended to show a preference for either CO blobs or interblobs, suggesting that the afferents in area 19 are segregated. • The intrinsic connections of area 19 form patches. • The intrinsic patches demonstrate an impressive variability in their size, structure and spacing • Both LS and 21a projecting cells form elongated band modules in area 19. • The cells in the LS projection bands are localized to layer 2/3. • The spacing of these two efferent band types (~2.5 mm) is comparable to that for the intrinsic patches (2.5 mm). • The 21a and LS projection bands interdigitate, with less densely labeled regions overlapping. Although area 19 is an extrastriate area, it too, like the primary visual cortex, it is divided up into segregated modules that form the modular mosaic in area 19. 168 Area 17 & 18 Area 19 CO blob pathway CO interblob pathway Figure 8.1 A schematic of the segregated afferent and efferent modules in area 19. The fact that cells in area 17 are preferentially located in the CO blobs or in the interblobs suggests that the afferent input to area 19 may be segregated into compartments. In this diagram the afferent modules are hypothesized as interdigitating CO blob recipient and CO interblob recipient compartments. The organization of the area 19 afferent compartments remains to be worked out. Efferent bands of cells projecting to 21a and LS roughly interdigitate with each other, while the borders of the bands that contain less label overlap. The relationship between the afferent efferent compartments is not known Conclusion Area 19 receives a global soft patterned input from area 17 (e.g. both CO blobs and interblobs project to area 19). The terminology soft and hard patterned are applied as specified by Shipp and Grant (1991; see Chapter 4 for details). This differs from the two 169 functionally distinct areas, 21a and LS, which both receive a hard patterned input from the CO blobs. This restricted input is a characteristic of functionally specific areas in the primate (i.e. M T and V4). However, the soft patterned input to area 19 is spatially arranged accordingly so there are regions in area 19 that receive a preferential input from the CO blobs and other regions that receive a preferential input from the interblobs. This suggests that, like V2, which also receives a global input, area 19 may be functionally divided into different afferent compartments. Evidence of functionally distinct modules in area 19 and their relationship to the anatomical modular mosaic defined in this study remains to be worked out. Still, further support of a division of labour within area 19 comes from the fact that input from 21a and LS is sequestered into banded modules. The partial overlap of these two types of bands in regions that have low labeling density may be a general feature of mammalian extrastriate organization that allows for parallel processing as well as communication between the two pathways. The spacing of the intrinsic connections is very similar to the efferent bands, which suggests a specific relationship between the two modules. It is tempting to speculate that an injection in one type of efferent band preferentially labels other bands of that same type. An example of this hypothesis is shown in Figure 8.2. In conclusion, extrastriate area 19 is organized into different modules: The afferent input from area 17 that appears to be at least partially segregated and the presence of projection bands to LS (a motion area) and 21a (a form area). Although the relationship between the afferent and efferent modules remains defined , the presence of segregated modules in area 19 strongly suggests the presence of at least two different parallel pathways. The overlapping portions of the 21a and LS 170 efferent bands also raise the intriguing possibility of convergence between the different modules in area 19. A R E A 19 | LS Projecting Band 21a Projecting Band Figure 8.2 A schematic of the hypothesized relationship between area 19 efferent bands and intrinsic connections. A t the top of the figure is an illustration of an injection. The red halo represents the ring of non-specific label that surrounds an injection site. The red circles represent intrinsically labeled patches. In this figure, the injection is being placed in a 21a projecting band. The patches of label are drawn so they colocalized with the 21a projecting bands. Alternatively i f the injection was placed in an L S band it is hypothesized that the intrinsic patches would be found preferentially in the L S bands. 171 References Abel PA, O'Brien BJ, Lia B , Olavarria JF (1997) Distribution of neurons projecting to the superior colliculus correlates with thick cytochrome oxidase stripes in macaque visual area 2. J Comp Neurol 377:313-323. Adams DL, Zeki S (2001) Functional organization of macaque V3 for stereoscopic depth. J Neurophysiol 86:2195-2203. Akase E, Inokawa H , Toyama K (1998) Neuronal responsiveness to three-dimensional motion in cat posteromedial lateral suprasylvian cortex. Exp Brain Res 122:214-226. Albright TD (1992) Form-cue invariant motion processing in primate visual cortex. Science 255:1141-1143. Albright TD, Stoner GR (1995) Visual motion perception. Proc Natl Acad Sci U S A 92:2433-2440. Albus K (1975) A quantitative study of the projection area of the central and the paracentral visual field in area 17 of the cat I. The precision of the topography. Exp Brain Res 24:159-179. Albus K , Beckmann R (1980) Second and third visual areas of the cat: interindividual variability in retinotopic arrangement and cortical Location. 299:247-276. Allman J, Zucker S (1990) Cytochrome oxidase and functional coding in primate striate cortex: an hypothesis. Cold Spring Harbor Sympia on Quantitative Biology 55:979-982. Amir Y , Harel M , Malach R (1993) Cortical hierarchy reflected in the organization of intrinsic connections in macaque monkey visual cortex. J Comp Neurol 334:19-46. Anderson PA, Olavarria J, Van Sluyters RC (1988) The overall pattern of ocular dominance bands in cat visual cortex. J Neurosci 8:2183-2200. Angelucci A , Levitt JB, Walton EJ, Hupe J M , Bullier J, Lund JS (2002) Circuits for local and global signal integration in primary visual cortex. J Neurosci 22:8633-8646. Bando T, Tsukuda K , Yamamoto N , Maeda J, Tsukahara N (1981) Cortical neurons in and around the Clare-Bishop area related with lens accommodation in the cat. Brain Res 225:195-199. Bando T, Toda H , Awaji T (1988) Lens accommodation-related and pupil-related units in the lateral suprasylvian area in cats. Prog Brain Res 75:231-236. Bando T, Hara N , Takagi M , Yamamoto K , Toda H (1996) Roles of the lateral suprasylvian cortex in convergence eye movements in cats. Prog Brain Res 112:143-156. Barlow HB, Blakemore C, Pettigrew JD (1967) The neural mechanism of bincular depth discrimination. J Physiol (Lond) 193:327-342. Bartfeld E, Grinvald A (1992) Relationships between Orientation-Preference pinwheels, cytochrome oxidase blobs, and Ocular-Dominance columns in primate striate cortex. Proc Natl Acad of Sci USA 89:11905-11909. Beck PD, Kaas JH (1998) Cortical connections of the dorsomedial visual areas in prosimian primates. J Comp Neurol 398:162-178. Benevento L A , Yoshida K (1981) The afferent and efferent organization of the lateral geniculo- prestriate pathways in the macaque monkey. J Comp Neurol 203:455-474. Bergeron A , Tardif E, Lepore F, Guillemot JP (1998) Spatial and temporal matching of receptive field properties of binocular cells in area 19 of the cat. Neuroscience 86:121-134. 173 Berman NE, Payne BR (1983) Alterations in connections of the corpus callosum following convergent and divergent strabismus. Brain Res 274:201-212. Berson D M (1985) Cat lateral suprasylvian cortex: Y-cell inputs and corticotectal projection. J Neurophysiol 53:544-556. Berson D M (1987) Retinal W-cell input to the upper superficial gray layer of the cat's superior colliculus: a conduction-velocity analysis. J Neurophysiol 58:1035-1051. Blasdel G, Lund JS (1983) Termination of afferent axons in macaque striate cortex. J Neurosci 3:1389-1413. Bonhoeffer T, Grinvald A (1991) Iso-orientation domains in cat visual cortex are arranged in pinwheel- like patterns. Nature 353:429-431. Boussaoud D, Ungerleider L G , Desimone R (1990) Pathways for motion analysis: cortical connections of the medial superior temporal and fundus of the superior temporal visual areas in the macaque. J Comp Neurol 296:462-495. Boycott B B , Wassle H (1974) The morphological types of ganglion cells of the domestic cat's retina. J Physiol 240:397-419. Boyd JD, Matsubara JA (1991) Intrinsic connections in cat visual cortex: a combined anterograde and retrograde tracing study. Brain Research 560:207-215. Boyd JD, Matsubara JA (1994b) Modular organization of corticocortical inputs and outputs of area 19. In: Soc. Neurosci. Abstr., p 1742. Miami Beach, Florida: Society for Neuroscience. Boyd JD, Matsubara JA (1994a) The tangential organization of callosal connectivity in the cat's visual cortex. J Comp Neurol 347:197-210. 174 Boyd JD, Matsubara JA (1996) Laminar and columnar patterns of geniculocortical projections in the cat: relationship to cytochrome oxidase. J Comp Neurol 365:659-682. Boyd JD, Matsubara JA (1999) Projections from V I to lateral suprasylvian cortex: an efferent pathway in the cat's visual cortex that originates preferentially from C O blob columns. Vis Neurosci 16:849-860. Boyd JD, Casagrande V A (1999) Relationship between cytochrome oxidase (CO) blobs in primate visual cortex (VI) and the distribution of cells projecting to the middle temporal area (MT). J Comp Neurol 409:573-591. Boyd JD, Mavity-Hudson JA, Casagrande V A (2000) The connections of layer 4 subdivisions in the primary visual cortex (VI) of the owl monkey. Cereb Cortex 10:644-662. Brosseau-Lachaine O, Faubert J, Casanova C (2001) Functional sub-regions for optic flow processing in the posteromedial lateral suprasylvian cortex of the cat. Cerebral Cortex 11:989-1001. Bullier J, Kennedy H (1983) Projection of the lateral geniculate nucleus onto cortical area V 2 in the macaque monkey. Exp Brain Res 53:168-172. Bullier J, Kennedy H , Salinger W (1984a) Branching and laminar origin of projections between visual cortical areas in the cat J Comp Neurol 228:329-341. Bullier J, Kennedy H , Salinger W (1984b) Bifurcation of subcortical afferents to visual areas 17,18, and 19 in the cat cortex. J Comp Neurol 228:309-328. Burke W, Dreher B , Wang C (1998) Selective block of conduction in Y optic nerve fibres: significance for the concept of parallel processing. Eur J Neurosci 10:8-19. Callahan EC, Haberly L B (1987) A n extensive intrinsic association fiber system within cat area 7 revealed by anterograde and retrograde axon tracing methods. J Comp Neurol 258:125-137. Carroll E, Wong-Riley M T T (1984) Quantitative light and electron microscopic analysis of cytochrome oxidase-rich zones in the striate cortex of the squirrel monkey. J Comp Neurol 22:1-17. Cartmill M (1992) New views on primate origins. Evolutionary Antropology 1:105-111. Casagrande V A , Norton TT (1991) Lateral geniculate nucleus: a review of its physiology and function. In: The Neural Basis of Visual Function (Leventhal A G , ed), pp 41-84. Houndmills, Basingstoke, Hampshire, and London, U.K.: The Macmillan Press. Casagrande V A (1994) A third parallel visual pathway to primate area V I . Trends Neurosci 17:305-310. Chen B, Hu X J , Pourcho R G (1996) Morphological diversity in terminals of W-type retinal ganglion cells at projection sites in cat brain. Vis Neurosci 13:449-460. Clare M H , Bishop G H (1954) Responses from an association area secondarily activated from optic cortex. J Neurophysiol 17:271-277. Clarke S, Innocenti G M (1986) Organization of immature intrahemispheric connections. J Comp Neurol 251:1-22. Cleland B G , Levick WR, Rushton W A (1975) Proceedings: Visual signals in the cat. J Physiol 252:42P-43P. Cleland B G , Levick WR, Morstyn R, Wagner HG (1976) Lateral geniculate relay of slowly conducting retinal afferents to cat visual cortex. J Neurophysiol 255:299-320. 176 Collins CE , Stepniewska I, Kaas JH (2001) Topographic patterns of v2 cortical connections in a prosimian primate (Galago garnetti). J Comp Neurol 431:155-167. Condo GC, Casagrande V A (1990) Organization of cytochrome oxidase staining in the visual cortex of nocturnal primates (Galago crassicaudatus and Galago senegalensis): I. Adult patterns. J Comp Neurol 293:632-645. Conway B, Boyd JD, Stewart T H , Matsubara JA (2000) The Projection from VI to Extrastriate Area 21a: A Second Patchy Efferent Pathway that Colocalizes With The CO Blob Columns in Cat Visual Cortex. Cerebral Cortex 10:149-159. Coogan T A , Burkhalter A (1993) Hierarchical organization of areas in rat visual cortex. J Neurosci 13:3749-3772. Cornwell P, Overman W, Ross C (1980a) Lesions of visual cortex impair discrimination of hidden figures by cats. Physiol Behav 24:533-540. Cornwell P, Overman W, Campbell A (1980b) Subtotal lesions of the visual cortex impair discrimination of hidden figures by cats. J Comp Physiol Psychol 94:289-304. Crabtree JW, Spear PD, McCall M A , Jones K R , Kornguth SE (1986) Contributions of Y -and W-cell pathways to response properties of cat superior colliculus neurons: comparison of antibody- and deprivation- induced alterations. J Neurophysiol 56:1157-1173. Cragg B G (1969) The topography of the afferent projections in the circumstriate visual cortex of the monkey studied by the Nauta method. Vision Res 9:733-747. Crair M C , Ruthazer ES, Gillespie DC, Stryker M P (1997) Ocular dominance peaks at pinwheel center singularities of the orientation map in cat visual cortex. J Neurophysiol 77:3381-3385. 177 Crockett DP, Maslany S, Harris SL, Egger M D (1993) Enhanced cytochrome-oxidase staining of the cuneate nucleus in the rat reveals a modifiable somatotopic map. Brain Research 612:41-55. Cusick C G , Kaas JH (1988) Cortical connections of area 18 and dorsolateral visual cortex in squirrel monkeys. Vis Neurosci 1:211-237. Cynader M , Timney B N , Mitchell D E (1980) Period of susceptibility of kitten visual cortex to the effects of monocular deprivation extends beyond six months of age. Brain Res 191:545-550. De Valois RL , Morgan HC, Poison M C , Mead WR, Hull E M (1974) Psychophysical studies of monkey vision. I. Macaque luminosity and color vision tests. Vision Res 14:53-67. Dehay C, Kennedy H , Bullier J (1988) Characterization of transient cortical projections from auditory, somatosensory, and motor cortices to visual areas 17,18, and 19 in the kitten. J Comp Neurol 272:68-89. DeYoe E A , Van Essen DC (1985) Segregation of efferent connections and receptive field properties in visual area V 2 of the macaque. Nature 317:5861. DeYoe E A , Van Essen D C (1988) Concurrent processing streams in monkey visual cortex. TINS 11:219-226. DeYoe E A , Hockfield S, Garren H , Van Essen D C (1990) Antibody labeling of functional subdivisions in visual cortex: Cat-301 immunoreactivity in striate and extrastriate cortex of the macaque monkey. Visual Neurosci 5:67-81. DeYoe E A , Glickman S, Wieser J (1992) Clustering of visual response properties in cortical area V 4 of macaque monkeys. Society for Neuroscience Abstracts 18:592. 178 Ding Y , Casagrande V A (1997) The distribution and morphology of L G N K pathway axons within the layers and CO blobs of owl monkey V I . Visual Neurosci 14:691-704. Ding Y , Casagrande V A (1998) Synaptic and neurochemical characterization of parallel pathways to the cytochrome oxidase blobs of primate visual cortex. J Comp Neurol 391:429-443. Dinse HR, Kruger K (1990) Contribution of area 19 to the foreground-background-interaction of the cat: an analysis based on single cell recordings and behavioural experiments. Exp Brain Res 82:107-122. Donaldson IML, Whitteridge D (1977) The nature of the boundary between cortical visual area II and III in the cat. Proc R Soc Lond B 199:445-462. Doty RW (1971) Survival of pattern vision after removal of striate cortex in the adult cat. J Comp Neurol 143:341-369. Dreher B, Sefton A J (1979) Properties of neurons In cat's dorsal lateral geniculate nucleus: a comparison between medial interlaminar and laminated parts of the nucleus. J Comp Neurol 183:47-64. Dreher B, Leventhal A G , Hale PT (1980) Geniculate input to cat visual cortex: a comparison of area 19 with areas 17 and 18. J Neurophysiol 44:804-826. Dreher B (1986) Thalamocortical and corticocoitical interconnections in the cat visual system: relation to the mechanisms of information processing. In: Visual Neurscience (Pettigrew JD, Sanderson K J , Levick WR, eds), pp 290-314. Cambridge: Cambridge University Press. Dreher B, Michalski A , Ho R H , Lee CW, Burke W (1993) Processing of form and motion in area 21a of cat visual cortex. Vis Neurosci 10:93-115. 179 Dreher B, Djavadian RL , Turlejeski K J , Wang C (1996c) Area PMLS and 21a of cat visual cortex are not only functionally but also hodologically distinct Prog Brain Res 112:251-276. Dreher B , Wang C, Turlejski K J , Djavadian R L , Burke W (1996a) Areas PMLS and 21a of cat visual cortex: two functionally distinct areas. Cereb Cortex 6:585-599. abs.html. Dreher B , Wang C, Burke W (1996b) Limits of parallel processing: excitatory convergence of different information channels on single neurons in striate and extrastriate visual cortices. Clin Exp Phar Physiol 23:913-925. Duffy CJ, Wurtz R H (1991) Sensitivity of MST neurons to optic flow stimuli. II. Mechanisms of response selectivity revealed by small-field stimuli. J Neurophysiol 65:1346-1359. Duysens J, Orban G A , van der Glas HW, De Zegher F E (1982a) Functional properties of area 19 as compared to area 17 of the cat. Brain Res 231:279-291. Duysens J, Orban G A , Van der Glas HW, Maes H (1982b) Receptive field structure of area 19 as compared to area 17 of the cat Brain Research 231:293-308. Dyck R H , Cynader MS (1993) A n interdigiated columnar mosaic of cytochrome oxidase, zinc, and neurotransmitter-related molecules in cat and monkey visual cortex. Proc Nat'l Acad Sci USA 90:9066-9069. Enroth-Cugell C, Robson JG (1966) The contrast sensitivity of retinal ganglion cells of the cat. J Physiol 187:517-552. Eysel UT, Worgotter F, Pape H C (1987) Local cortical lesions abolish lateral inhibition at direction selective cells in cat visual cortex. Exp Brain Res 68:606-612. 180 Eysel UT, Muche T, Worgotter F (1988) Lateral interactions at direction-selective striate neurones in the cat demonstrated by local cortical inactivation. J Physiol 399:657-675. Eysel UT, Crook J M , Machemer HF (1990) GABA-induced remote inactivation reveals cross-orientation inhibition in the cat striate cortex. Exp Brain Res 80:626-630. Feldon SE, Andrews BW, Pollen D A (1978) Periodic complex cells in cortical area 19 of the cat. Vision Res 18:347-350. Felleman DJ, Burkhalter A , Van Essen D C (1987) Cortical connections of areas V3 and Vp of macaque monkey extrastriate visual cortex. J Comp Neurol 379:21-47. Felleman DJ, Van Essen DC (1991) Distributed hierarchical processing in the primate cerebral cortex. Cerebral Cortex 1:1-47. Felleman DJ, Xiao Y , McClendon E (1997) Modular organization of occipito-temporal pathways: cortical connections between visual area 4 and visual area 2 and posterior inferotemporal ventral area in macaque monkeys. J Neurosci 17:3185-3200. Ferrer JMR, Price DJ, Blakemore C (1988) The organization of corticocortical projections from area 17 to area 18 of the cat's visual cortex. Proc R Soc Lond B 223:77-98. Ferrer JMR, Kato N , Price DJ (1992) Organization of association projections from area 17 to areas 18 and 19 and to suprasylvian areas in the cat's visual cortex. J Comp Neurol 316:261-278. Ferrera VP , Nealy TA, Maunsell JHR (1994) Responses in macaque visual area v4 following inactivation of the parvocellular and magnocellular L G N pathways. J Neurosci 14:2080-2088. 181 Florence SL, Kaas JH (1992) Ocular dominance columns in area 17 of Old World macaque and talapoin monkeys: complete reconstructions and quantitative analyses. Visual Neurosci 8:449^162. Freund TF, Martin K A C , Whitteridge D (1985) Innervation of cat visual areas 17 and 18 by physiologically identified X - and Y-type thalamic afferents. I. Arborization patterns and quantitative distribuyion of postsynaptic elements. J Comp Neurol 242:263-274. Freund TF, Martin K A C , Soltesz I, Somogyi P, Whitteridge D (1989) Arborization pattern and postsynaptic targets of physiologically identified thalamocortical afferents in striate cortex of the macaque monkey. J Comp Neurol 289:315-336. Friedlander M J , Lin C-S, Stanford LR, Sherman S M (1981) Morphology of functionally identified neurons in lateral geniculate nucleus of the cat J Neurophysiol 46:80-129. Friedman HR, Bruce CJ, Goldman-Rakic (1989) Resolution of metabolic columns by a double-label 2-DG technique: interdigitation and coincidence in visual cortical areas of the same monkey. J Neurosci 9:411-4121. Fries W (1981) The projection from the lateral geniculate nucleus to the prestriate cortex of the macaque monkey. Poc R Soc Lond B 213:73-80. Frostig RD, Lieke EE, Ts'o D Y , Grinvald A (1990) Cortical functional architecture and local coupling between neuronal activity and the microcirculation revealed by in vivo high-resolution optical imaging of intrinsic signals. Proc Natl Acad Sci U S A 87:6082-6086. Fukuda Y , Stone J (1974) Retinal distribution and central projections of Y - , X - , and W-cells of the cat's retina. J Neurophysiol 37:749-772. 182 Gabbott P L A , Martin K A C , Whitteridge D (1987) Connections between pyramidal neurons in layer 5 of cat visual cortex (area 17). J Comp Neurol 259:364-381. Gallant JL, Braun J, Van Essen I X ! (1993) Selectivity for polar, hyperbolic, and Cartesian gratings in macaque visual cortex. Science 259:100-103. Gallant JL, Connor CE , Rakshit S, Lewis JW, Van Essen DC (1996) Neural responses to polar, hyperbolic, and Cartesian gratings in area V4 of the macaque monkey. J Neurophysiol 76:2718-2739. Galuske RA, Singer W (1996) The origin and topography of long-range intrinsic projections in cat visual cortex: a developmental study. Cereb Cortex 6:417-430. abs.html. Garey LJ , Powell TPS (1971) A n experimental study of the termination of the lateral geniculo-cortical pathway in the cat and monkey. Proc R Soc Lond B 179:41-63. Gattas R, Sousa A P B , Gross C G (1988) Visuotopic organization and extent of V3 and V 4 of the macaque. J Neurosci 8:1831-1845. Gattass R, Gross C G , Sandell JH (1981) Visual topography of V 2 in the macaque. J Comp Neurol 201:519-539. Gegenfurtner K R , Kiper DC, Levitt JB (1997) Functional properties of neurons in macaque area V3. J Neurophysiol 77:1906-1923. Geisert EE (1985) The projection of the lateral geniculate nucleus to area 18. J Comp Neurol 238:101-106. Gerfen CR, Sawchenko PE (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and terminals: 183 immunohistochemical localization of an axonally transported plant lectin, Phaseolus vulgaris Leucoagglutin (PHA-L). Brain Res 290:219-238. Ghose G M , Ohzawa I, Freeman RD (1994a) Receptive-field maps of correlated discharge between pairs of neurons in the cat's visual cortex. J Neurophysiol 71:330-346. Ghose G M , Freeman RD, Ohzawa I (1994b) Local intracortical connections in the cat's visual cortex: postnatal development and plasticity. J Neurophysiol 72:1290-1303. Ghose G M , Ohzawa I, Freeman R D (1995) A flexible PC-based physiological monitor for animal experiments. J Neurosci Methods 62:7-13. Ghose G, Ts'o D Y (1997) Form processing modules in primate area V4. J Neurophysiol 77:2191-21%. Gilbert CD, Kelly JP (1975) The projection of cells in different layers of the cats visual cortex. J Comp Neurol 163:81-106. Gilbert C D , Wiesel T N (1979) Morphology and intracortical projections of functionally characterised neurones in the cat visual cortex. Nature 280:120-125. Gilbert C D (1983) Microcircuitry of the visual cortex. Ann Rev Neurosci 6:217-247. Gilbert CD, Wiesel T N (1983) Clustered intrinsic connections in cat visual cortex. J Neurosci 3:1116-1133. Gilbert C D (1985) Horizontal integration in the neocortex. TINS 8:160-165. Gilbert C D , Wiesel T N (1989) Columnar specificity of intrinsic horizontal and corticocortical connections in cat visual cortex. J Neurosci 9:2432-2442. Gilbert C D , Hirsch JA, Wiesel T N (1990) Lateral interactions in visual cortex. Cold Spring Harbour Symposia on Quantitative Biology 55:663-677. 184 Girard P, Bullier J (1989) Visual activity in area V 2 during reversible inactivation of area 17 in the macaque monkey. J Neurophysiol 62:1287-1302. Girman SV, Sauve Y , Lund RD (1999) Receptive field properties of single neurons in rat primary visual cortex. J Neurophysiol 82:301-311. Godecke I, K im DS, Bonhoeffer T, Singer W (1997) Development of orientation preference maps in area 18 of kitten visual cortex. Eur J Neurosci 9:1754-1762. Graham J (1977) A n autoradiographic study of the efferent connections of the superior colliculus in the cat. J Comp Neurol 173:629-654. Grant S, Shipp S (1991) Visuotopic organization of the lateral suprasylvian area and of an adjacent area of the ectosylvian gyrus of cat cortex: a physiological and connectional study. Visual Neurosci 6:315-338. Gray C M , Konig P, Engel A K , Singer W (1989) Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties. Nature 338:334-337. Graybiel A M (1972) Some ascending connections of the pulvinar and nucleus lateralis posterior of the thalamus in the cat Brain Res 44:99-125. Graziano MS, Andersen R A , Snowden RJ (1994) Tuning of MST neurons to spiral motions. J Neurosci 14:54-67. Gross C G , Cowey A , Manning FJ (1971) Further analysis of visual discrimination deficits following fovea! prestriate and inferotemporal lesions in rhesus monkeys. J Comp Physiol Psychol 76:1-7. Guillemot J-P, Richer L , Ptito M , Lepore F (1993a) Disparity coding in the cat: a comparison between areas 17-18 and area 19. Prog Brain Res 95:179-187. 185 Guillemot JP, Paradis M C , Samson A , Ptito M , Richer L , Lepore F (1993b) Binocular interaction and disparity coding in area 19 of visual cortex in normal and split-chiasm cats. Exp Brain Res 94:405-417. Hammond P (1978) The neural basis for colour discrimination in the domestic cat. Vision Res 18:233-235. Hailing JK, Guillery R W (1976) Organization of retinocollicular pathways in the cat. J Comp Neurol 166:133-144. Harting JK, Casagrande V A , Weber JT (1978) The projection of the primate superior colliculus upon the dorsal lateral geniculate nucleus: autoradiographic demonstration of the interlaminar distribution of tectal axons. Brain Research 150:593-599. Harting JK, Huerta M F , Hashikawa T, van Lieshout DP (1991) Projection of the mammalian superior colliculus upon the dorsal lateral geniculate nucleus: organization of tectogeniculate pathways in nineteen species. J Comp Neurol 304:275-306. Harting JH, Updyke B V , Van Lieshout DP (1992) Corticotectal projections in the cat: anterograde transport studies of twenty-five cortical areas. J Comp Neurol 234:379-414. Harvey A R (1980) A physiological analysis of subcortical and commissural projections of areas 17 and 18 of the cat. J Physiol 302:507-534. Harvey PH, Pagel M D (1991) The Comparative Method in Evolutionary Biology. Oxford: Oxford University Press. Heath CJ, Jones E G (1970) Connexions of area 19 and the lateral suprasylvian area of the visual cortex of the cat. Brain Research 19:302-305. Hegde J, Van Essen DC (2000) Selectivity for complex shapes in primate visual area V2. J Neurosci 20:RC61. 186 Hendrickson A E , Wilson JR, Ogren M P (1978) Neuroanatomical organization of pathways between the dorsal lateral geniculate nucleus and visual cortex in old world and new world primates. J Comp Neurol 182:123-136. Hendry SHC, Yoshioka T (1994) A neurochemically distinct third channel in the macaque dorsal lateral geniculate nucleus. Science 264:575-577. Hendry SH, Reid RC (2000) The koniocellular pathway in primate vision. Annu Rev Neurosci 23:127-153. Hernandez-Gonzalez A , Cavada C, Reinoso-Suarez F (1994) The lateral geniculate nucleus projects to the inferior temporal cortex in the macaque monkey. Neuroreport 5:2693-2696. Heywood C A , Cowey A (1987) On the role of cortical area V4 in the discrimination of hue and pattern in macaque monkeys. J Neurosci 7:2601-2617. Heywood C A , Cowey A (1992) The role of the 'face-cell' area in the discrimination and recognition of faces by monkeys. Philos Trans R Soc Lond B Biol Sci 335:31-37; discussion 37-38. Hickey T L , Guillery R W (1974) A n autoradiographic study of retinogeniculate pathways in the cat and the fox. J Comp Neurol 1156:239-254. Hilgetag CC, Grant S (2000) Uniformity, specificity and variability of corticocortical connectivity. Philos Trans R Soc Lond B Biol Sci 355:7-20. Hoffmann K P (1973) Conduction velocity in pathways from retina to superior colliculus in the cat: a correlation with receptive-field properties. J Neurophysiol 36:409-424. Hoffmann KP, Schoppmann A (1975) Retinal input to direction selective cells in the nucleus tractus opticus of the cat. Brain Res 99:359-366. 187 Hollander H , Vanegas H (1977) The projection from the lateral geniculate nucleus onto the visual cortex in the cat A quantitative study with horseradish-peroxidase. J Comp Neurol 173:519-536. Horton JC, Hubel D H (1981) Regular patchy distribution of cytochrome oxidase staining in primary visual cortex of macaque monkey. Nature 292:762-764. Horton JC (1984) Cytochrome oxidase patches: a new cytoarchitectonic feature of monkey visual cortex. Philos Trans R Soc Lond B Biol Sci 304:199-253. Horton JC, Hocking DR (1997) Timing of the critical period for plasticity of ocular dominance columns in macaque striate cortex. J Neurosci 17:3684-3709. Hubel D H , Wiesel T N (1962) Receptive fields of cells in striate cortex of very young, visually inexperienced kittens. Journal of Neurophysiology 26:994-1002. Hubel D H , Wiesel T N (1963) Shape and arrangement of columns in cat's striate cortex. J Physiol 165:559-568. Hubel D H , Wiesel T N (1965) Receptive fields and functional architecture in two nonstriate visual areas (18 and 19) of the cat. J Neurophysiol 28:229-289. Hubel D H , Wiesel T N (1969) Visual area of the lateral suprasylvian gyrus (Clare-Bishop area) of the cat. J Physiol 202:251-260. Hubel D H , Wiesel T N (1972) Laminar and columnar distribution of geniculo-cortical fibers in the macaque monkey. J Comp Neurol 146:421-450. Hubel D H , Livingstone MS (1987) Segregation of form, color, and stereopsis in primate area 18. Journal Neurosci 7:3378-3415. Hubel D H , Wiesel T N (1998) Early exploration of the visual cortex. Neuron 20:401-412. 188 Hubener M , Shoham D, Grinvald A , BonhoefferT (1997) Spatial relationships among three columnar systems in cat area 17. J Neurosci 17:9270-9284. Hughes HC, Sprague J M (1986) Cortical mechanisms for local and global analysis of visual space in the cat. Exp Brain Res 61:332-354. Humphery A L , Sur M , Uhlrich DJ, Sherman S M (1985a) Termination patterns of individual X - and Y-cell axons in the visual cortex of the cat: projections to area 18, to the 17/18 border region, and to both areas 17 and 18. J Comp Neurol 233:190-212. Humphery A L , Sur M , Uhlrich DJ, Sherman S M (1985b) Projection patterns of individual X - and Y-cell axons from the lateral geniculate nucleus to cortical area 17 in the cat. J Comp Neurol 233:159-189. Humphrey A L , Murthy A (1999) Cell types and response timings in the medial interlaminar nucleus and C-layers of the cat lateral geniculate nucleus. Vis Neurosci 16:513-525. Innocenti G M , Manger PR, Masiello I, Colin I, Tettoni L (2002) Architecture and callosal connections of visual areas 17,18,19 and 21 in the ferret (Mustela putorius). Cereb Cortex 12:411-422. Irvin GE, Norton TT, Sesma M A , Casagrande V A (1986) W-like response properties of interlaminar zone cells in the lateral geniculate nucleus of a primate (Galago crassicaudatus). Brain Research 362:254-270. Itoh K , Konishi A , Nomura S, Mizuno N , Nakamura Y , Sugimoto T (1979) Application of coupled oxidation reaction to electron microscopic demonstration of horseradish peroxidase: cobalt-glucose oxidase method. Brain Research 175:341-346. 189 Itoh K , Conley M , Diamond T (1981) Different distributions of large and small retinal ganglion cells in the cat after HRP injections of single layers of the lateral geniculate body and the superior colliculus. Brain Research 207:147-152. Kaas JH (1997) Topographic maps are fundamental to sensory processing. Brain Research Bulletin 44:107-112. Kaas JH, Lyon DC (2001) Visual cortex organization in primates: theories of V3 and adjoining visual areas. Prog Brain Res 134:285-295. Kageyama G H , Wong-Riley M (1986a) Laminar and cellular localization of cytochrome oxidase in the cat striate cortex. J Comp Neurol 245:137-159. Kageyama G H , Wong-Riley M (1986b) The localization of cytochrome oxidase in the L G N and striate cortex of postnatal kittens. J Comp Neurol 243:182-194. Kato N , Kawaguchi S, Miyata H (1984) Geniculocortical projection to layer I of area 17 in kittens: orthograde and retrograde HRP studies. J Comp Neurol 225:441-447. Kato N , Ferrer JMR, Price DJ, Blakemore C (1993) Effects of neonatal ablation of area 18 on corticocortical projections from area 17 to extrastriate areas in the cat Behav Brain Res 54:205-210. Kawamura S, Sprague J M , Niimi K (1974) Corticofugal projections from the visual cortices to the thalamus, pretectum and superior colliculus in the cat. J Comp Neurol 158:339-362. Kawamura S, Fukoshima N , Hattori S, Kudo M (1980) Laminar segregation of cells of origin of ascending projections from the superficial layers of the superior colliculus in the cat. Brain Research 184:486-490. 190 Kawano J (1998) Cortical projections of the parvocellular laminae C of the dorsal lateral geniculate nucleus in the cat: an anterograde wheat germ agglutinin conjugated to horseradish peroxidase study. J Comp Neurol 392:439-457. Kennedy H , Bullier J (1985) A double-labeling investigation of the afferent connectivity to cortical areas V I and V 2 of the Macaque monkey. J Neurosci 5:2815-2830. Kim JN, Mulligan K , Sherk H (1997) Simulated optic flow and extrastriate cortex. I. Optic flow versus texture. J Neurophysiol 77:554-561. Kimura M , Shiida T, Tanaka K , Toyama K (1980) Three classes of area 19 cortical cells of the cat classified by their neuronal connectivity and photic responsiveness. Vision Res 20:69-77. Kisvarday ZF, Eysel UT (1993) Functional and structural topography of horizontal inhibitory connections in cat visual cortex. Eur J Neurosci 5:1558-1572. Kisvarday ZF, K im DS, Eysel UT, Bonhoeffer T (1994) Relationship between lateral inhibitory connections and the topography of the orientation map in cat visual cortex. Eur J Neurosci 6:1619-1632. Kisvarday ZF, Toth E, Rausch M , Eysel UT (1997) Orientation-specific relationship between populations of excitatory and inhibitory lateral connections in the visual cortex of the cat. Cereb Cortex 7:605-618. Kisvarday ZF, Eysel UT (1992) Cellular organization of reciprocal patchy networks in layer III of cat visual cortex (area 17). Neurosci 46:275-286. Kobatake E, Tanaka K (1994) Neuronal selectivities to complex object features in the ventral visual pathway of the macaque cerebral cortex. J Neurophysiol 71:856-867. 191 Kobbert C, Thanos S (2000) Topographic representation of the sciatic nerve motor neurons in the spinal cord of the adult rat correlates to region-specific activation patterns of microglia. J Neurocytol 29:271-283. Kobbert C, Apps R, Bechmann I, Lanciego JL, Mey J, Thanos S (2000) Current concepts in neuroanatomical tracing. Prog Neurobiol 62:327-351. Kolb H , Nelson R, Mariani A (1981) Amacrine cells, bipolar cells and ganglion cells of the cat retina: a Golgi study. Vision Res 21:1081-1114. Krubitzer L , Kaas J (1990) Convergence of processing channels in the extrastriate cortex of monkeys. Visual Neurosci 5:609-613. Kruger J, Gouras P (1980) Spectral selectivity of cells and its dependence on slit length in monkey visual cortex. J Neurophysiol 43:1055-1069. Kruger K , Donicht M , Muller-Kusdian G, Kiefer W, Berlucchi G (1988) Lesion of areas 17/18/19: effects on the cat's performance in a binary detection task. Exp Brain Res 72:510-516. Kulikowski JJ, Walsh V , McKeefry D, Butler SR, Carden D (1994) The electrophysiological basis of colour processing in macaques with V 4 lesions. Behav Brain Res 60:73-78. Lachica E A , Beck PD, Casagrande V A (1992) Parallel pathways in Macaque Monkey striate cortex: anatomically defined columns in layer III. Proc Natl Acad Sci USA 89:3566-3570. Lachica E A , Casagrande V A (1992) Direct W-Like Geniculate Projections to the Cytochrome Oxidase (CO) Blobs in Primate Visual Cortex - Axon Morphology. J Comp Neurol 319:141-158. 192 Lachica E A , Beck PD, Casagrande V A (1993) Intrinsic connections of Layer-Ill of striate cortex in squirrel monkey and bush baby - correlations with patterns of cytochrome oxidase. J Comp Neurol 329:163-187. Lagae L , Maes H , Raiguel S, Xiao D K , Orban G A (1994) Responses of macaque STS neurons to optic flow components: a comparison of areas M T and MST. J Neurophysiol 71:1597-1626. Law MI, Zahs K R , Stryker M P (1988) Organization of primary visual cortex (Area 17) in the ferret. J Comp Neurol 278:157-180. Lennie P (1998) Single units and visual cortical organization. Perception 27:889-935. Levay S, Hubel D H , Wiesel T N (1975) The pattern of ocular dominance columns in macaque visual cortex revealed by a reduced silver stain. J Comp Neurol 159:559-575. LeVay S, Gilbert C D (1976) Laminar patterns of geniculocortical projection in the cat. Brain Research 113:1-19. LeVay S, Wiesel T N , Hubel D H (1980) The development of ocular dominance columns in normal and visually deprived monkeys. J Comp Neurol 191:1-51. LeVay S (1988) The patchy intrinsic projections of visual cortex. Prog Brain Res 75:147-161. Leventhal A G , Rodieck RW, Dreher B (1981) Retinal ganglion cell classes in the old world monkey: morphology and central projections. Science 213:1139-1142. Leventhal A G , Hirsch H V B (1983) Effects of visual deprivation upon the geniculocortical W-cell pathway in the cat: area 19 and its afferent input. J Comp Neurol 214:59-71. 193 Leventhal A G , Rodieck RW, Dreher B (1985) Central projections of cat retinal ganglion cells. J Comp Neurol 237:216-226. Leventhal A G , Thompson K G , Liu D, Zhou Y , Ault SJ (1995) Concomitant sensitivity to orientation, direction, and color of cells in layers 2,3, and 4 of monkey striate cortex. J Neurosci :1808-1818. Levitt JB, Yoshioka T, Lund JS (1994a) Intrinsic cortical connections in macaque visual area V2: evidence for interaction between different functional streams. J Comp Neurol 342:551-570. Levitt JB, Kiper DC, Movshon JA (1994b) Receptive fields and functional architecture of macaque V2. J Neurophysiol 71:2517-2542. Levitt JB, Yoshioka T, Lund JS (1995) Connections between the pulvinar complex and cytochrome oxidase-defined compartments in visual area V 2 of macaque monkey. Exp Brain Res 104:419-430. Levitt JB, Lund JS, Yoshioka T (1996) Anatomical substrates for early stages in cortical processing of visual information in the macaque monkey. Behav Brain Res 76:5-19. L i B , L i BW, Chen Y , Wang L H , Diao Y C (2000) Response properties of PMLS and PLLS neurons to simulated optic flow patterns. Eur J Neurosci 12:1534-1544. L i B , Chen Y , L i BW, Wang L H , Diao Y C (2001) Pattern and component motion selectivity in cortical area PMLS of the cat. Eur J Neurosci 14:690-700. Lia B , Olavarria JF (1996) The distribution of corticotectal projection neurons correlates with the interblob compartment in macaque striate cortex. Vis Neurosci 13:461-466. Liu Y , Gu Q, Cynader MS (1993) An improved staining technique for cytochrome C oxidase. J Neurosci Meth 49:181-184. 194 Livingstone MS, Hubel D H (1982) Thalamic inputs to cytochrome oxidase rich regions in monkey visual cortex. Proc Natl Acad Sci USA 79:6098-6101. Livingstone MS, Hubel D H (1983) Specificity of cortico-cortical connections in monkey visual system. Nature 304:531-534. Livingstone MS, Hubel D H (1984a) Specificity of intrinsic connections in primate primary visual cortex. J Neurosci 4:2830-2835. Livingstone MS, Hubel D H (1984b) Anatomy and physiology of a color system in the primate visual cortex. J Neurosci 4:309-356. Livingstone MS, Hubel D H (1987) Connections between layer 4B of area 17 and the thick cytochrome oxidase stripes of area 18 in the squirrel monkey. J Neurosci 7:3371-3377. Livingstone M , Hubel D (1988) Segregation of form, colour, movement, and depth: anatomy, physiology, and perception. Science 240:740-749. Lomber SG, Payne BR (2000) Translaminar differentiation of visually guided behaviors revealed by restricted cerebral cooling deactivation. Cereb Cortex 10:1066-1077. Lomber SG (2001) Behavioral cartography of visual functions in cat parietal cortex: areal and laminar dissociations. Prog Brain Res 134:265-284. Loop MS, Bruce L L , Petuchowski S (1979) Cat colour vision: the effect of stimulus size, shape and viewing distance. Vision Res 19:507-513. Lowel S, Singer W (1992) Selection of horizontal ceonnections in the visual cortex by correlated neuronal activity. Science 255:209-212. 195 Lowenstein PR, Somogyi P (1991) Synaptic organization of cortico-cortical connections from the primary visual cortex to the posteromedial lateral suprasylvian visual area in the cat. The Journal of Comparative Neurology 310:253-266. Luhmann HJ, Millan L M , Singer W (1986) Development of horizontal intrinsic connections in cat striate cortex. Exp Brain Res 63:443-448. Luhmann HJ, Singer W, Martinez-Millan L (1991) Horizontal interactions in cat striate cortex: I. Anatomical substrate and postnatal development. European Journal of Neuroscience 2:344-357. Lund JS, Yoshioka T (1991) Local circuit neurons of macaque monkey striate cortex: III. Neurons of laminae 4B, 4A, and 3B. J Comp Neurol 311:234-258. Lund JS, Yoshioka T, Levitt JB (1993) Comparison of intrinsic connectivity in different areas of macaque monkey cerebral cortex. Cerebral Cortex 3:148-162. Lund JS, Angelucci A , Bressloff PC (2003) Anatomical substrates for functional columns in macaque monkey primary visual cortex. Cereb Cortex 13:15-24. Lyon DC, Jain N , Kaas JH (1998) Cortical connections of striate and extrastriate visual areas in tree shrews. J Comp Neurol 401:109-128. Lyon DC, Kaas JH (2002) Evidence for a modified V3 with dorsal and ventral halves in macaque monkeys. Neuron 33:453-461. Lysakowski A , Standage GP, Benevento L A (1988) A n investigation of collateral projections of the dorsal lateral geniculate nucleus and other subcortical structures to cortical areas V I and V4 in the macaque monkey: a double label retrograde tracer study. Exp Brain Res 69:651-661. 196 Maciewicz RJ (1975) Thalamic afferents to areas 17,18, and 19 of cat cortex traced with horseradish peroxidase. BRain Research 84:308-312. MacNeil M A , Lomber SG, Payne BR (1997) Thalamic and cortical projections to middle suprasylvian cortex of cats: constancy and variation. Exp Brain Res 114:24-32. Maekawa H , Ohtsuka K (1993) Afferent and efferent connections of the cortical accommodation area in the cat. Neurosci Res 17:315-323. Malach R (1989) Patterns of connections in rat visual cortex. J Neurosci 9:3741-3752. Malach R, Amir Y , Harel M , Grinvald A (1993) Relationship between intrinsic connections and functional architecture revealed by optical imaging and in vivo targeted biocytin injections in primate striate cortex. Proc Natl Acad Sci USA 90:10469-10473. Malach R, Tootell R B H , Malonek D (1994) Relationship between orientation domains, cytochrome oxidase stripes, and intrinsic horizontal connections in squirrel monkey V2. Cerebral Cortex 4:151-165. Malach R, Schirman TD, Harel M , Tootell RB, Malonek D (1997) Organization of intrinsic connections in owl monkey area MT. Cereb Cortex 7:386-393. Malonek D, Tootell RB, Grinvald A (1994) Optical imaging reveals the functional architecture of neurons processing shape and motion in owl monkey area MT. Proc R Soc Lond B Biol Sci 258:109-119. Malpeli JG, Schiller PH, Colby C L (1981) Response properties of single cells in monkey striate cortex during reversible inactivation of individual lateral geniculate laminae. J Neurophysiol 46:1102-1119. Manger PR, Kiper D, Masiello I, Murillo L , Tettoni L , Hunyadi Z, Innocenti G M (2002a) The representation of the visual field in three extrastriate areas of the ferret (Mustela 197 putorius) and the relationship of retinotopy and field boundaries to callosal connectivity. Cereb Cortex 12:423^+37. Manger PR, Masiello I, Innocenti G M (2002b) Areal organization of the posterior parietal cortex of the ferret (Mustela putorius). Cereb Cortex 12:1280-1297. Marshall W H , Talbot SA, Ades H W (1943) Cortical response of the anesthetized cat to gross photic electrical afferent stimulation. Journal of Neurophysiology 6:1-15. Martin K A , Whitteridge D (1984) Form, function and intracortical projections of spiny neurones in the striate visual cortex of the cat. J Physiol 353:463-504. Mastronarde D N (1983) Interactions between ganglion cells in cat retina. J Neurophysiol 49:350-365. Matsubara J, Cynader M , Swindale N V , Stryker M P (1985) Intrinsic projections within visual cortex: evidence for orientation-specific local connections. Proc Natl Acad Sci U S A 82:935-939. Matsubara JA, Cynader MS, Swindale NV (1987) Anatomical properties and physiological correlates of the intrinsic connections in cat area 18. J Neurosci 7:1428-1446. Matsubara JA, Phillips DP (1988) Intracortical connections and their physiological correlates in the primary auditory cortex (AI) of the cat. J Comp Neurol 268:38-48. Matsubara JA, Boyd JD (1992) Presence of GABA-immunoreactive neurons within intracortical patches in area 18 of the cat. Brain Res 583:161-170. Maunsell JH, van Essen DC (1983) The connections of the middle temporal visual area (MT) and their relationship to a cortical hierarchy in the macaque monkey. J Neurosci 3:2563-2586. 198 Merabet L , Minville K , Ptito M , Casanova C (2000) Responses of neurons in the cat posteromedial lateral suprasylvian cortex to moving texture patterns. Neuroscience 97:611-623. Mesulam M M (1978) Tetramethyl benzidine for horseradish peroxidase neurohistochemistry: a non-carcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents. J Histochem Cytochem 26:106-117. Michalski A , Gerstein G L , Czarkowska J, Tarnecki R (1983) Interactions between cat striate cortex neurons. Exp Brain Res 51:97-107. Mitchell D E (1989) Normal and abnormal visual development in kittens: insights into the mechanisms that underlie visual perceptual development in humans. Can J Psychol 43:141-164. Mitchison G, Crick F (1982) Long axons within the striate cortex: their distribution, orientation, and patterns of connection. Proc Natl Acad Sci U S A 79:3661-3665. Mizobe K , Itoi M , Kaihara T, Toyama K (1988) Neuronal responsiveness in area 21a of the cat. Brain Res 438:307-310. Montero V M , Rojas A , Torrealba F (1973) Retinotopic organization of striate and peristriate visual cortex in the albino rat. Brain Res 53:197-201. Montero V M (1980) Patterns of connections from the striate cortex to cortical visual areas in superior temporal sulcus of macaque and middle temporal gyrus of owl monkey. J Comp Neurol 189:45-59. Montero V M (1981) Topography of the cortico-cortical connections from the striate cortex in the cat. Brain Behav Evol 18:194-218. Montero V M (1993) Retinotopy of cortical connections between the striate cortex and extrastriate visual areas in the rat. Exp Brain Res 94:1-15. Morley JW, Yuan L , Vickery R M (1997) Corticocortical connections between area 21a and primary visual cortex in the cat. Neuroreport 8:1263-1266. Morley JW, Vickery R M (1997) Spatial and temporal frequency selectivity of cells in area 21a of the cat. J Physiol 501:405^+13. Mountcastle V B (1957) Modality and topographic properties of single neurons of cat's somatosensory cortex. Journal of Neurophysiology 20:408-434. Mountcasde V B (1997) The columnar organization of the neocortex. Brain 120:701-722. Muller T, Stetter M , Hubener M , Sengpiel F, Bonhoeffer T, Godecke I, Chapman B, Lowel S, Obermayer K (2000) A n analysis of orientation and ocular dominance patterns in the visual cortex of cats and ferrets. Neural Comput 12:2573-2595. Mulligan K , Sherk H (1993) A comparison of magnification functions in area 19 and the lateral suprasylvian visual area in the cat. Exp Brain Res 97:195-208. Mulligan K , Kim JN, Sherk H (1997) Simulated optic flow and extrastriate cortex. II. Responses to bar versus large-field stimuli. J Neurophysiol 77:562-570. Murphy K M , Jones DG, Van Sluyters RC (1995) Cytochrome-oxidase blobs in cat primary visual cortex. J Neurosci 15:4196-4208. Murphy K M , Duffy K R , Jones DG, Mitchell D E (2001) Development of cytochrome oxidase blobs in visual cortex of normal and visually deprived cats. Cereb Cortex 11:122-135. 200 Myasnikov A A , Dykes RW, Leclerc SS (1997) Correlating cytoarchitecture and function in cat primary somatosensory cortex: the challenge of individual differences. Brain Res 750:95-108. Nealey T A , Maunsell JH (1994) Magnocellular and parvocellular contributions to the responses of neurons in macaque striate cortex. J Neurosci 14:2069-2079. Nelson J l , Frost BJ (1985) Intracortical facilitation among co-oriented, co-axially aligned simple cells in cat striate cortex. Exp Brain Res 61:54-61. Newsome WT, Wurtz RH, Dursteler MR, Mikami A (1985) Deficits in visual motion processing following ibotenic acid lesions of the middle temporal visual area of the macaque monkey. J Neurosci 5:825-840. Newsome WT, Pare EB (1988) A selective impairment of motion perception following lesions of the middle temporal visual area (MT). J Neurosci 8:2201-2211. Niimi K , Matsuoka H , Yamazaki Y , Matsumoto H (1981) Thalamic afferents to the visual cortex in the cat studied by retrograde axonal transport of horseradish peroxidase. Brain Behav Evol 18:114-139. Norita M , Kase M , Hoshino K , Meguro R, Funaki S, Hirano S, McHaffie JG (1996) Extrinsic and intrinsic connections of the cat's lateral suprasylvian visual area. Prog Brain Res 112:231-250. Norton TT, Casagrande V A (1982) Laminar organization of receptive-field properties in lateral geniculate nucleus of bush baby (Galago crassicaudatus). J Neurophysiol 47:715-741. 201 Norton TT, Casagrande V A , Irvin GE, Sesma M A , Perry H M (1988) Contrast-sensitivity functions of W-, X - , and Y-like relay cells in the lateral geniculate nucleus of bush baby, Galago crassicaudatus. J Neurophysiol 59:1639-1656. Ohki K, Hagino N , Ihara T, Morita M (1998) Orientation preference maps in area 21a of the cat extrastriate cortex. Society for Neuroscience Abstracts 24. Olavarria J, Van Sluyters RC (1985) Unfolding and flattening the cortex of gyrencephalic brains. J Neurosci Methods 15:191-202. Olavarria J, Montero V M (1989) Organization of visual cortex in the mouse revealed by correlating callosal and striate-extrastriate connections. Vis Neurosci 3:59-69. Olavarria JF, Van Essen DC (1997) The global pattern of cytochrome oxidase stripes in visual area V 2 of the macaque monkey. Cereb Cortex 7:395-404. Olavarria JF (2001) Callosal connections correlate preferentially with ipsilateral cortical domains in cat areas 17 and 18, and with contralateral domains in the 17/18 transition zone. J Comp Neurol 433:441-457. Olson CR, Lawler K (1987) Cortical and subcortical afferent connections of a posterior division of feline area 7 (area 7p). J Comp Neurol 259:13-30. Orban G A , Lagae L , Verri A , Raiguel S, Xiao D, Maes H , Torre V (1992) First-order analysis of optical flow in monkey brain. Proc Natl Acad Sci U S A 89:2595-2599. Palmer L A , Rosenquist A C , Tusa RJ (1978) The retinotopic organization of lateral suprasylvian visual areas in the cat. J Comp Neurol 177:237-256. Pasternak T, Horn K M , Maunsell JH (1989) Deficits in speed discrimination following lesions of the lateral suprasylvian cortex in the cat. Vis Neurosci 3:365-375. 202 Pasupathy A , Connor C E (1999) Responses to contour features in macaque area V4. J Neurophysiol 82:2490-2502. Payne BR, Peters A (1989) Cytochrome oxidase patches and meynert cells in monkey visual cortex. Neurosci 28:353-363. Payne BR, Siwek DF (1991) Visual-field map in the callosal recipient zone at the border between areas 17 and 18 in the cat. 7:221-236. Payne B R (1993) Evidence for visual cortical area homologs in cat and macaque monkey. Cerebral Cortex 3:1-25. Payne BR (2002) The concept of cat primary visual cortex. Toronto: Academic Press. Perry V H , Oehler R, Cowey A (1984) Retinal ganglion cells that project to the dorsal lateral geniculate nucleus in the macaque monkey. Neuroscience 12:1101-1123. Peters A , Sethares C (1991) Organization of pyramidal neurons in area 17 of monkey visual cortex. J Comp Neurol 306:1-23. Pettigrew JD, Dreher B (1987) Parallel processing of binocular disparity in the cat's retinogeniculocortical pathways. Proc R Soc Lond B 232:297-321. Preuss T M , Beck PD, Kaas JH (1993) Areal, modular and connectional organization of visual cortex in a prosimian primate, the slow loris (Nycticebus coucang). Brain Behav Evol 42:321-335. Preuss T M , Kaas JH (1996) Cytochrome oxidase "blobs' and other characteristics of primary visual cortex in a lemuroid primate, Cheirogaleus medius. Brain Behav Evol 47:103-112. Preuss T M (2000) Taking the measure of diversity: comparative alternatives to the model-animal paradigm in cortical neuroscience. Brain Behav Evol 55:287-299. 203 Price DJ, Blakemore C (1985a) Regressive events in the postnatal development of association projections in the visual cortex. Nature 316:721-724. Price DJ, Blakemore C (1985b) The postnatal development of the association projection from visual cortical area 17 to area 18 in the cat. J Neurosci 5:2443-2452. Price DJ (1985a) Patterns of cytochrome oxidase activity in areas 17,18 and 19 of the visual cortex of cats and kittens. Exp Brain Res 58:125-133. Price DJ (1985b) Projections from visual cortical area 19, the posterior medial lateral suprasylvian area and the lateral posterior-pulvinar complex of the thalamus to areas 17 and 18 in young kittens. Neurosci Lett 60:363-368. Price DJ, Zumbroich TJ (1989) Postnatal development of corticocortical efferents from area 17 in the cat's visual cortex. J Neurosci 9:600-613. Ptito M , Lepore F, Guillemot JP (1991) Stereopsis in the cat: behavioral demonstration and underlying mechanisms. Neuropsychologia 29:443-464. Raczkowski D, Rosenquist A C (1980) Connections of the parvocellular C laminae of the dorsal lateral geniculate nucleus with the visual cortex in the cat. Brain Res 199:447-451. Raczkowski D, Rosenquist A C (1983) Connections of the multiple visual cortical areas with the lateral posterior-pulvinar complex and adjacent thalamic nuclei in the cat. J Neurosci 3:1912-1942. Raizada RD, Grossberg S (2003) Towards a theory of the laminar architecture of cerebral cortex: computational clues from the visual system. Cereb Cortex 13:100-113. Rakic P (1995) Radial versus tangential migration of neuronal clones in the developing cerebral cortex. Proc Natl Acad Sci U S A 92:11323-11327. Rathjen S, Schmidt K E , Lowel S (2002) Two-dimensional analysis of the spacing of ocular dominance columns in normally raised and strabismic kittens. Exp Brain Res 145:158-165. Rauschecker JP, von Griinau M W , Poulin C (1987) Thalamocortical connections and their correlation with receptive field properties in the cat's lateral syprasylvian cortex. Exp Brain Res 67:100-112. Read H L , Winer JA, Schreiner C E (2001) Modular oganization of intrinsic connections associated with spectral tuning in cat auditory cortex. Proceedings of the National Academy of Science 98:8042-8047. Rockland KS, Pandya D N (1979) Laminar origins and terminations of cortical connections of the occipital lobe in the rhesus monkey. Brain Res 179:3-20. Rockland KS , Lund JS, Humphery A L (1982) Anatomial banding of intrinsic connections in striate cortex of tree shrews (Tupaia glis). J Comp Neurol 209:41-58. Rockland KS , Lund JS (1982) Widespread periodic intrinsic connections in the tree shrew visual cortex. Science 215:1532-1534. Rockland KS, Lund JS (1983) Intrinsic laminar lattice connections in primate visual cortex. J Comp Neurol 216:303-318. Rockland K S (1985a) Anatomical organization of primary visual cortex (area 17) in the ferret J Comp Neurol 241:225-236. Rockland KS (1985b) A reticular pattern of intrinsic connections in primate area V 2 (area 18). J Comp Neurol 235:467-478. Rodman HR, Albright TD (1989) Single-unit analysis of pattern-motion selective properties in the middle temporal visual area (MT). Exp Brain Res 75:53-64. 205 Roe A W , Ts'o DY (1995) Visual topography in primate V2: multiple representation across functional stripes. J Neurosci 15:3689-3715. Roe A W , Ts'o D Y (1999) Specificity of color connectivity between primate V I and V2. J Neurophysiol 82:2719-2730. Rosa MGP, Casagrande V A , Preuss T, Kaas JH (1997) Visual field representation in striate and prestriate cortices of a prosimian primate (Galago garnetti). J Neurophysiol 77:3193-3217. Rosa M G , Krubitzer L A (1999) The evolution of visual cortex: where is V2? Trends Neurosci 22:242-248. Rowe M H , Stone J (1977) Naming of neurones. Classification and naming of cat retinal ganglion cells. Brain Behav Evol 14:185-216. Rowe M H , Stone J (1980) The interpretation of variation in the classification of nerve cells. Brain Behav Evol 17:123-151. Rowe M H , Drehher B (1982) Retinal W-cell projections to the medial interlaminar nucleus in the cat: implications for ganglion cell classifications. J Comp Neurol 204:117-133. Rudolph K K , Pasternak T (1996) Lesions in cat lateral suprasylvian cortex affect the perception of complex motion. Cereb Cortex 6:814-822. Rumberger A , Tyler CJ, Lund JS (2001) Intra- and inter-areal connections between the primary visual cortex V I and the area immediately surrounding V I in the rat. Neuroscience 102:35-52. Saito H (1983) Pharmacological and morphological differences between X - and Y-type ganglion cells in the cat's retina. Vision Res 23:1299-1308. 206 Saito H , Yukie M , Tanaka K , Hikosaka K , Fukada Y , Iwai E (1986) Integration of direction signals of image motion in the superior temporal sulcus of the macaque monkey. J Neurosci 6:145-157. Saito H-a, Tanaka K , Fukada Y , Oyamada H (1988) Analysis of discontinuity in visual contours in area 19 of the cat. J Neurosci 8:1131-1143. Sakata H , Shibutani H , Ito Y , Tsurugai K (1986) Parietal cortical neurons responding to rotary movement of visual stimulus in space. Exp Brain Res 61:658-663. Sanderson K J (1971) Visual field projection columns and magnification factors in the lateral geniculate nucleus of the cat. Exp Brain Res 13:159-177. Scannell JW, Young M P (1993) The connectional organization of neural systems in the cat cerebral cortex. Curr Biol 3:191-200. Scannell JW, Blakemore C, Young M P (1995) Analysis of connectivity in the cat cerebral cortex. J Neurosci 15:1463-1438. Scannell JW, Burns G A , Hilgetag C C , O'Neil M A , Young M P (1999) The connectional organization of the cortico-thalamic system of the cat Cereb Cortex 9:277-299. Schein SJ, Marrocco RT, de Monasterio F M (1982) Is there a high concentration of color-selective cells in area V 4 of monkey visual cortex? J Neurophysiol 47:193-213. Schiller PH, Mai pel i JG (1977) The effect of striate cortex cooling on area 18 cells in the monkey. Brain Res 126:366-369. Schwark HD, Esteky H , Jones EG (1992) Corticocortical connections of cat primary somatosensory cortex. Exp Brain Res 91:425-434. Segal R L , Beckstead R M (1984) The lateral suprasylvian corticotectal projection in cats. J Comp Neurol 225:259-275. 207 Segraves M A , Rosenquist A C (1982a) The afferent and efferent callosal connections of retinotopically defined areas in cat cortex. J Neurosci 2:1090-1107. Segraves M A , Rosenquist A C (1982b) The distribution of the cells of origin of callosal projections in cat visual cortex. Neurosci 2:1079-1089. Sereno M l , McDonald CT, Allman J M (1994) Analysis of retinotopic maps in extrastriate cortex. Cereb Cortex 4:601-620. Shatz CJ, Lindstrom S, Wiesel T N (1977) The distribution of afferents representing the right and left eyes in the cat's visual cortex. Brain Research 131:103-116. Shatz CJ, Stryker M P (1978) Ocular dominance in layer IV of the cat's visual cortex and the effects of monocular deprivation. J Physiol 281:267-283. Shatz CJ, Stryker M P (1988) Prenatal tetrodotoxin infusion blocks segregation of retinogeniculate afferents. Science 242:87-89. Sherk H (1986a) Coincidence of patchy inputs from the lateral geniculate complex and area 17 to the cat's Clare-Bishop area. J Comp Neurol 253:105-120. Sherk H (1986b) Location and connections of visual cortical areas in the cat's suprasylvian sulcus. J Comp Neurol 247:1-31. Sherk H (1988) Retinotopic order and functional organization in a region of suprasylvian visual cortex: the Clare-Bishop area. Prog Brain Res 75:237-244. Sherk H (1990) Functional organization of input from areas 17 and 18 to an extrastriate area in the cat. J Neurosci 10:2780-2790. Sherk H , K i m JN, Mulligan K (1995) Are the preferred directions of neurons in cat extrastriate cortex related to optic flow? Vis Neurosci 12:887-894. 208 Sherman S M , Spear PD (1982) Organization of visual pathways in normal and visually deprived cats. Physiol Rev 62:739-855. Shipp S, Zeki S (1985) Segregation of pathways leading from area V 2 to areas V4 and V5 of Macaque monkey visual cortex. Nature 315:322-325. Shipp S, Zeki S (1989) The organization of connections between areas V5 and V 2 in macaque monkey visual cortex. Euro J Neurosci 1:333-353. Shipp S, Grant S (1991) Organization of reciprocal connections between area 17 and the lateral suprasylvian area of cat visual cortex. Vis NeuroSci 6:339-355. Shipp S, Zeki S (2002) The functional organization of area V2,1: specialization across stripes and layers. Vis Neurosci 19:187-210. Shmuel A , Grinvald A (19%) Functional organization for direction of motion and its relationship to orientation maps in cat area 18. J Neurosci 16:6945-6964. Shoham D, Hubener M , Schulze S, Grinvald A , Bonhoeffer T (1997) Spatio-temporal frequency domains and their relation to cytochrome oxidase staining in cat visual cortex. Nature 385:529-533. Shostak Y , Ding Y , Mavity-Hudson J, Casagrande V A (2002) Cortical synaptic arrangements of the third visual pathway in three primate species: Macaca mulatta, Saimiri sciureus, and Aotus trivirgatus. J Neurosci 22:2885-2893. Silverman NS, Tootell R B H (1987) Modified technique for cytochrome oxidase histochemistry: increased staining intensity and compatibility with 2-deoxyglucose autoradiography. J Neurosci Meth 19:1-10. Sincich L C , Horton JC (2002a) Divided by cytochrome oxidase: a map of the projections from V I to V 2 in macaques. Science 295:1734-1737. 209 Sincich L C , Horton JC (2002b) Pale cytochrome oxidase stripes in V 2 receive the richest projection from macaque striate cortex. J Comp Neurol 447:18-33. Smith DC, Spear PD (1979) Effects of superior colliculus removal on receptive-field properties of neurons in lateral suprasylvian visual area of the cat. J Neurophysiol 42:57-75. Sonty RV, Juliano SL (1997) Development of intrinsic connections in cat somatosensory cortex. J Comp Neurol 384:501-516. Spear PD, Baumann TP (1975) Receptive-field characteristics of single neurons in lateral suprasylvian visual area of the cat. J Neurophysiol 38:1403-1420. Spear PD, Tong L, Sawyer C (1983) Effects of binocular deprivation on responses of cells in cat's lateral suprasylvian visual cortex. J Neurophysiol 49:366-382. Spear PD (1991) Functions of extrastriate visual cortex in non-primate species. In: The Neural Basis of Visual Function (Leventhal A G , ed), pp 339-370. Houndmills, Basingstoke, Hampshire, and London, U.K.: The Macmillan Press. Sprague J M , Levy J, DiBerardino A , Berlucchi G (1977) Visual cortical areas mediating form discrimination in the cat. J Comp Neurol 172:441-488. Sprague J M (1996) Neural mechanisms of visual orienting responses. Prog Brain Res 112:1-15. Squatrito S, Galletti C, Battaglini PP, Sanseverino ER (1981) An autoradiographic study of bilateral cortical projections from cat area 19 and lateral suprasylvian visual area. Arch Ital Biol 119:21-42. 210 Stanford LR, Friedlander J, Sherman S M (1983) Morphological and physiological properties of geniculate W-cells of the cat: a comparison with X - and Y-cells. Journal of Neurophysiology 50:582-608. Stepniewska I, Kaas JH (1996) Topographic patterns of V2 cortical connections in macaque monkeys. J Comp Neurol 371:129-152. Stewart T H , Boyd JD, Matsubara JA (2000) Organization of efferent neurons in area 19: the projection to extrastriate area 21a. Brain Res 881:47-56. Stewart T H , Boyd JD, Matsubara JA (2003a) Patchy intrinsic connections in extrastriate area 19 of the cat. In preparation. Stewart T H , Boyd JD, Matsubara JA (2003b) CO Blobs and Interblobs Project Separately to Area 19. submitted to Journal of Comparative Neurology . Stone J, Dreher B (1973) Projection of X - and Y-cells of the cat's lateral geniculate nucleus to areas 17 and 18 of visual cortex. J Neurophysiol 36:551-567. Stone JS, Dreher B , Leventhal A (1979) Hierarchical and parallel mechanisms in the organization of visual cortex. Brain Res Rev 1:345-395. Stoner GR, Albright TD (1992) Motion coherency rules are form-cue invariant. Vision Res 32:465-475. Swindale N V , Shoham D, Grinvald A , Bonhoeffer T, Hubener M (2000) Visual cortex maps are optimized for uniform coverage. Nat Neurosci 3:822-826. tef/DynaPage.taf?file=/neuro/joumal/v823/n828/full/nn0800_0822.html taf/ItynaPage.taf?file=/neuro/jouTna^ 211 Symonds L L , Rosenquist A C , Edwards SB, Palmer L A (1981) Projections of the pulvinar-lateral posterior complex to visual cortical areas in the cat. Neuroscience 6:1995-2020. Symonds L L , Rosenquist A C (1984a) Corticocortical connections among visual areas in the cat. J Comp Neurol 229:1-38. Symonds L L , Rosenquist A C (1984b) Laminar origins of visual corticocortical connections in the cat. J Comp Neurol 229:39-47. Takagi M , Toda H , Yoshizawa T, Hara N , Ando T, Abe H , Bando T (1992) Ocular Convergence-Related neuronal responses in the lateral suprasylvian area of alert cats. Neuroscience Research 15:229-234. Tanaka M , Weber H , Creutzfeldt OD (1986) Visual properties and spatial distribution of neurones in the visual association area on the prelunate gyrus of the awake monkey. Exp Brain Res 65:11-37. Tanaka K , Ohzawa I, Ramoa A S , Freeman R D (1987) Receptive field properties of cells in area 19 of the cat. Exp Brain Res 65:549-558. Tanaka K , Saito H (1989) Analysis of motion of the visual field by direction, expansion/contraction, and rotation cells clustered in the dorsal part of the medial superior temporal area of the macaque monkey. J Neurophysiol 62:626-641. Tanaka K , Fukada Y , Saito H A (1989) Underlying mechanisms of the response specificity of expansion/contraction and rotation cells in the dorsal part of the medial superior temporal area of the macaque monkey. J Neurophysiol 62:642-656. Tanaka M , Lindsley E, Lausmann S, Creutzfeldt OD (1990) Afferent connections of the prelunate visual association cortex (areas V 4 and DP). Anat Embryol 181:19-30. 212 Tanaka K (2003) Columns for complex visual object features in the inferotemporal cortex: clustering of cells with similar but slightly different stimulus selectivities. Cereb Cortex 13:90-99. Tardif E, Bergeron A , Lepore F, Guillemot J-P (19%) Spatial resolution and contrast sensitivity of single neurons in area 19 of split-chiasm cats: a comparison with primary visual cortex. Euro J Neurosci 9:1929-1939. Thejomayen D M , Matsubara JA (1993) Confocal microscopic study of the dendritic organization of patchy, intrinsic neurons in area 18 of the cat. Cerebral Cortex 3:442-453. Thompson RF, Johnson RH, Hoopes JJ (1%3) Organization of auditory, somatic sensory, and visual projection to association fields of cerebral cortex in the cat. Journal of Neurophysiology 26:343-364. Tieman SB, Tumosa N (1983) [14C]2-deoxyglucose demonstration of the organization of ocular dominance in areas 17 and 18 of the normal cat. Brain Research 267:35-46. Tigges J, Tigges M , Anschel S, Cross N A , Ledbetter WD, Mcbride R L (1981) Areal and laminar distribution of neurons interconnecting the central visual cortical areas 17,18, 19, and M T in squirrel monkey (Saimiri). J Comp Neurol 202:539-560. Toda H , Yoshizawa T, Takagi M , Bando T (2001) The properties of convergence eye movements evoked from the rostral and caudal lateral suprasylvian cortex in the cat. Neurosci Res 39:359-367. Tong L , Kalil RE, Spear PD (1982) Thalamic projections to visual areas of the middle suprasylvian sulcus in the cat. J Comp Neurol 212:103-117. Tootell RB, Silverman MS, De Valois RL , Jacobs G H (1983) Functional organization of the second cortical visual areas in primates. Science 220:737-739. 213 Tootell RB, Hamilton SL (1989) Functional anatomy of the second visual area (V2) in the macaque. J Neurosci 9:2620-2644. Toyama K , Mizobe K , Akase E, Kaihara T (1994) Neuronal responsiveness in areas 19 and 21a, and the posteromedial lateral suprasylvian cortex of the cat Exp Brain Res 99:289-301. Ts'o D Y , Gilbert CD, Wiesel T N (1986) Relationships between horizontal interactions and functional architecture in cat striate cortex as revealed by cross-correlation analysis. J Neurosci 6:1160-1170. Ts'o D Y , Frostig RD, Lieke EE, Grinvald A (1990) Functional organization of primate visual cortex revealed by high resolution optical imaging. Science 240:417-420. Ts'o D Y , Roe A W , Gilbert C D (2001) A hierarchy of the functional organization for color, form and disparity in primate visual area V2. Vision Res 41:1333-1349. Tusa RJ, Palmer L A , Rosenquist A C (1978) The retinotopic organization of area 17 (striate cortex) in the cat J Comp Neurol 177:213-236. Tusa RJ, Rosenquist A C , Palmer L A (1979) Retinotopic organization of areas 18 and 19 in the cat. J Comp Neurol 185:657-678. Tusa RJ, Palmer L A (1980) Retinotopic organization of areas 20 and 21 in the cat. J Comp Neurol 193:147-164. Ungerleider L G , Mishkin M (1982) Two cortical visual systems. In: Analysis of visual behavior (Ingle DJ, Goodale M A , Mansfield RJW, eds), pp 549-586. Cambridge, M A : MIT Press. 214 Updyke B V (1977) Topographic organization of the projections from cortical areas 17, 18 and 19 onto the thalamus, pretectum and superior colliculus in the cat. J Comp Neurol 173:81-122. Van Essen DC, Newsome WT, Maunsell JHR, Bixby JL (1986) The projections from striate cortex (VI) to areas V 2 and V3 in the macaque monkey: asymetries, areal boundaries, and patchy connections. J Comp Neurol 244:451-480. Van Essen DC, Felleman DJ, DeYoe E A , Olavarria J, Knierim J (1990) Modular and hierarchical organization of extrastriate visual cortex in the macaque monkey. Cold Spring Harbour Symp Quant Biol 55:679-696. Vanduffel W, Vandenbussche E, Singer W, Orban G A (1997) A metabolic mapping study of orientation discrimination and detection tasks in the cat. Eur J Neurosci 9:1314-1328. Veenman C L , Reiner A , Honig M G (1992) Biotinylated dextran amine as an anterograde tracer for single-labeling and double-labeling studies. J Neurosci Meth 41:239-254. Voigt T, LeVay S, Stamnes M A (1988) Morphological and Immunocytochemical observations on the visual callosal projections in the cat. J Comp Neurol 272:450-460. Waleszczyk WJ, Wang C, Burke W, Dreher B (1999) Velocity response profiles of collicular neurons: parallel and convergent visual information channels. Neuroscience 93:1063-1076. Wallace M N , Kitzes L M , Jones E G (1991) Intrinsic inter- and intralaminar connections and their relationship to the tonotopic map in catprimary auditory cortex. Exp Brain Res 86:527-544. 215 Wang C, Dreher B (1996) Binocular interactions and disparity coding in area 21a of cat extrastriate visual cortex. Exp Brain Res 108:257-272. Wang C, Dreher B , Huxlin KR, Burke W (1997) Excitatory convergence of Y and non-Y information channels on single neurons in the PMLS area, a motion area of the cat visual cortex. Euro J Neurosci 9:921-933. Weller RE, Wall JT, Kaas JH (1984) Cortical connections of the middle temporal visual area (MT) and the superior temporal cortex in owl monkeys. J Comp Neurol 228:81-104. Wiesel T N , Hubel D H , Lam D M K (1974) Autoradiographic demonstration of ocular-dominance columns in the monkey striate cortex by means of transneuronal transport. Brain Research 79:273-279. Wikstrom M , Saraste M , Krab K (1981) Cytochrome oxidase, a synthesis. Toronto: Academic Press. Wilkinson FE (1980) Reversal of the behavioural effects of monocular deprivation as a function of age in the kitten. Behav Brain Res 1:101-123. Wilson PD, Stone J (1975) Evidence of W-cell input to the cat's visual cortex via the C-laminae of the lateral geniculate nucleus. Brain Research 92:472-478. Wilson PD, Rowe M H , Stone J (1976) Properties of relay cells in cat's lateral geniculate nucleus: a comparison of W-cells with X - and Y-cells. J Neurophysiol 39:1193-1209. Wimborne B M , Henry G H (1992) Response characteristics of the cells of cortical area-21a of the cat with special reference to orientation specificity. Journal of Physiology -London 449:457-478. 216 Wong-Riley M (1979) Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain Research 171:11-28. Wong-Riley M , Riley D A (1983) The effect of impulse blockage on cytochrome oxidase activity in the cat visual system. Brain Research 261:185-193. Wong-Riley MTT (1989) Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci 12:94-101. Wong-Riley MTT, Tripathi SC, Trusk TC, Hoppe D A (1989a) Effect of retinal impulse blockage on cytochrome oxidase-rich zones in the macaque striate cortex: I quantitative electron-microscopic (EM) analysis of neurons. Visual Neurosci 2:483-497. Wong-Riley MT, Tripathi SC, Trusk TC, Hoppe D A (1989b) Effect of retinal impulse blockage on cytochrome oxidase-rich zones in the macaque striate cortex: I. Quantitative electron-microscopic (EM) analysis of neurons. Vis Neurosci 2:483-497. Xiao Y , Zych A , Felleman DJ (1999) Segregation and convergence of functionally defined V 2 thin stripe and interstripe compartment projections to area V4 of macaques. Cereb Cortex 9:792-804. Yin TCT, Greenwood M (1992) Visual response properties of neurons in the middle and lateral suprasylvian cortices of the behaving cat. Exp Brain Res 88:1-14. Yoshioka T, Levitt JB, Lund JS (1992) Intrinsic lattice connections of macaque monkey visual cortical area V4. J Neuroci 12:2785-2802. Yoshioka T, Levitt JB, Lund JS (1994) Independence and merger of thalamocortical channels within macaque monkey primary visual cortex: anatomy of interlaminar projections. Vis Neurosci 11:467-490. 217 Yoshioka T, Blasdel GG, Levitt JB, Lund JS (1996) Relationship between patterns of intrinsic lateral connectivity, ocular dominance, and cytochrome oxidase-reactive regions in macaque monkey striate cortex. Cerebral Cortex 6:297-310. Young JZ (1981) The life of vertebrates, 3rd Edition. Oxford: Clarendon. Yousef T, Bonhoeffer T, Kim DS, Eysel UT, Toth E, Kisvarday Z F (1999) Orientation topography of layer 4 lateral networks revealed by optical imaging in cat visual cortex (area 18). Eur J Neurosci 11:4291-4308. Yukie M , Iwai E (1981) Direct projection from the dorsal lateral geniculate nucleus to the prestiate cortex in macaque. J Comp Neurol 201:81-97. Zaborszky L , Heimer L (1989) Combination of tracer techniques, especially HRP and PHA-L, with transmitter identification for correlated light and electron microscopic syudies. In: Neuroanatomical Tract-Tracing Methods 2. Recent Progress (Zaborszky L , Heimer L , eds), pp 49-%. New York: Plenum Press. Zeki S M (1%9) The secondary visual areas of the monkey. Brain Res 13:197-226. Zeki S M (1973) Colour coding in rhesus monkey prestriate cortex. Brain Research 53:422-427. Zeki S M (1978a) Uniformity and diversity of structure and function in rhesus monkey prestriate visual cortex. J Physiol 277:273-290. Zeki S M (1978b) The third visual complex of rhesus monkey prestriate cortex. J Physiol 277:245-272. Zeki S, Shipp S (1988) The functional logic of cortical connections. Nature 335:311-317. Zeki S (2003) Improbable areas in the visual brain. Trends Neurosci 26:23-26. 218 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0091380/manifest

Comment

Related Items