Work

Specialised Visual Representations and Long-range Communication Pathways in the Mouse Visual Cortex

Year: 2021
Type: dissertation
Abstract: Animals, from insects to humans, rely on vision to sense, understand and interact with the external world.In mammals, even the simplest visual sensation activates millions of nerve cells embedded in hundreds of long-range neural circuits that process and distribute visual information across the brain.A fundamentally unresolved question is how these long-range circuits are organized functionally and anatomically to generate neural representations and guide behavior.In this thesis, I studied this question in the visual cortex of the mouse.The visual cortex plays key roles in many visual perception and behaviors.It is composed of multiple visual areas that form distinct representations of the visual world.These visual areas receive visual information via parallel information processing streams running across the cortex and via the thalamus.However, the function of these intracortical and thalamocortical pathways, and their contribution to visual processing remain poorly understood.This study had two main aims: (1) to characterize the diversity and distribution of visual information encoded within and across mouse visual areas, (2) to characterize the visual information carried by the intracortical and thalamocortical pathways.To achieve these aims, I developed a cellular optical imaging approach to characterize cortical visual representations and visual information pathways that make up the mouse visual cortex.I applied this approach first to study the visual response properties of neurons in eight distinct visual cortical areas (the primary visual cortex V1, and higher visual areas LM, AL, RL, A, AM, PM, and LI).I then applied this approach to examine the function of cortical and thalamic neurons that project to three representative higher visual areas (AL, PM, and A).In terms of cellular populations, I focused on layer 2/3 and layer 5 cortical excitatory neurons and neurons in the lateral posterior nucleus in the thalamus (pulvinar), which convey visual information across visual areas.In terms of visual response properties, I focused on tunings for the spatiotemporal frequency and orientation, which characterize visual information processing channels and could allow linking to the animal's visual perceptual abilities.SUMMARY II Using rich visual stimuli, in vivo cellular imaging, and GCaMP transgenic mice, I first characterized visual spatiotemporal tunings of layer 2/3 excitatory neurons in eight visual cortical areas and found high functional diversity and specificity in all areas.To determine the underlying organization of visual processing channels, I applied an unsupervised clustering approach to group neurons with similar tuning properties.I found all areas encode diverse visual information but differ widely in the numerosity of neurons tuned to particular spatiotemporal frequencies and spatial elongations, suggesting visual cortical areas differentially partake in distinct visual processing channels.These data also revealed a global division of spatiotemporal frequency preferences between anterior and posterior higher visual areas, as well as a division of spatial elongation preferences between lateral and dorsal areas, suggesting segregated cortical processing streams underpinning distinct visual functions.To understand the function of cortical and thalamocortical information pathways in establishing distinct visual representations in higher visual areas, I examined the specificity of visual information flows from cortical and thalamic neurons to higher visual areas.To study cortical pathways, I used retrograde AAV to label cortical neurons projecting to specific higher visual areas in combination with cellular calcium imaging to measure their response properties.To study thalamocortical pathways, I used AAV to label neurons in the lateral posterior thalamic nucleus with calcium indicators and used calcium imaging to measure the response properties of their axonal projections in higher visual areas.Looking at the distribution of cortical neurons projecting to higher visual areas, I found each area samples from distinct sets of cortical areas, the strength of anatomical connectivity reflects the similarity of visual representations between the source and target areas.Looking at their response properties, I found a moderate level of specificity, with projecting neurons often differing significantly from the general population in the source areas and showing similar response properties as the target areas.Each cortical pathway provides slightly different visual information and contributes differentially to visual representations.In comparison, I found the lateral posterior nucleus in the thalamus provides highly specific information to visual cortical areas, suggesting a primary role of thalamocortical pathways in the establishment of visual representations in the cortex.SUMMARY III Together, these results revealed the high diversity and specificity of spatiotemporal tunings in mouse visual cortical areas, as well as a functional target-specific logic of long-range connectivity.I propose that specialized long-range connections might link functionally alike neurons across brain areas and form multiple distributed subnetworks for parallel processing of spatiotemporal information.Moreover, our data suggest that highly specific thalamocortical pathways play a primary role in determining visual representations in the cortex, which are concurrently shaped by intracortical pathways with various degrees of functional specificity.Overall, this work provides novel insights into the organization of visual channels underlying specialized visual representations, as well as on the rules of cortical and thalamic connectivity and hierarchical sensory processing.This study also provides the groundwork for future studies in exploring specific pathways underlying perception and behaviors.Map of the thesis Chapter 1 introduces the research questions, the current state and gap of knowledge, and research objectives.Chapters 2-4 provide large-scale cellular functional characterization of mouse visual cortical neurons and demonstrate different visual representations across areas for spatial elongation, spatiotemporal frequencies and visual motion speed.Chapter 5 provides a categorization of visual processing channels based on the multidimensional tuning measurements, and demonstrates the wide yet asymmetric distribution of visual channels across visual cortical areas.Chapter 6 demonstrates the functional specificity of intracortical communication pathways using cellular imaging and retrograde neural tracing.Chapter 7 presents the functional specificity of thalamocortical projections using axonal imaging.The experiments related to Chapters 2-5 and Chapters 6-7 are reported in two research articles.Chapter 8 summarizes the main results and provides discussion and general conclusions.SAMENVATTING IV SAMENVATTING Alle dieren, van insecten tot mensen, gebruiken hun zicht om de externe wereld te waar te nemen, te begrijpen en ermee te interageren.Bij zoogdieren activeert de eenvoudigste visuele prikkel miljoenen zenuwcellen, die vervat zijn in honderden uitgebreide neurale netwerken die informatie verwerken en rondsturen doorheen heel het brein.Een fundamenteel onopgeloste vraag is hoe deze netwerken anatomisch en functioneel georganiseerd zijn om neurale representaties op te bouwen en gedrag te sturen.In deze thesis, heb ik deze vraag bestudeerd in de visuele cortex van de muis.De visuele cortex is van vitaal belang voor visuele perceptie en vele gedragingen.Dit hersengebied is samengesteld uit verschillende subgebieden die unieke representaties vormen van de visuele wereld.Deze visuele gebieden ontvangen visuele informatie via parallelle informatie verwerkingsstromen die doorheen de cortex en via de thalamus lopen.Over de functie van deze intracorticale en thalamocorticale paden en hun bijdrage tot visuele verwerking is echter nog weinig bekend.
Author: Han, Xu
Institution: KU Leuven
Language: English
Open Access status: green
OpenAlex ID: W7062092266
Cites: 187
Cited by:
Related to: 10