Elements of neurogeometry. Functional architectures of vision. Translated from the French by the author (Q2399908)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Elements of neurogeometry. Functional architectures of vision. Translated from the French by the author
scientific article

    Statements

    Elements of neurogeometry. Functional architectures of vision. Translated from the French by the author (English)
    0 references
    0 references
    24 August 2017
    0 references
    This is the first of a two-volume work dedicated to a presentation of progresses in the domain of using specific geometric structures in the neuroscience of vision. The field of investigations is named neurogeometry. As the author states in the preface: neurogeometry deals with the internal and immanent geometric algorithms that allow the visual system to build the external and transcendent geometry of our surrounding world. A characteristic of the work is that in all chapters, at all steps theoretical results are supported by experimental data provided by an extended number of contributions of the best known specialists in the field. Many of the presented neurophysiological experimental data justify the used modeling procedures. History, context, acknowledgements, goal as well as limits of this investigation are discussed in the first chapter. The second chapter focuses on the origin of spatial representation and explains some basic concepts as perceptual geometry, neurogeometry, gestalt geometry, geometry's twofold way, logical idealities, neurophysiology, functional neuroanatomy, functional architecture, cognitive science, geometric models. Some examples as ``Kanizsa's illusory contours'' and ``the cut locus of a figure'' complete the chapter. The third chapter is devoted to receptive fields and receptive profiles of visual neurons. In the sections visual neurons as filters and vision and wavelets, the relation to filters and wavelet theory is described. The geometric formatting of the signal by means of a certain kind of wavelet, is a way of optimizing the compression of natural images. The outline on wavelets and the discussion on receptive profiles and information theory are completed with a presentation of many experimental results. The forth chapter covers a large set of experimental data on the functional architecture of area V1-- the first of the primary visual areas -- and its pinwheel structure. The functional architecture includes two main components, actually two main classes of connections. The first is the class of retino-geniculo-cortical ``vertical'' connections, discussed in this chapter and the second is the class of cortico-cortical ``horizontal'' connections, discussed in the fifth chapter. The forth chapter starts with an explanation of areas of the visual cortex, illustrated by a set of pictures from experimental data. Follows a discussion on V1 as a mesoscopic fibration. Further one concentrates on the pinwheel structure of V1 and presents a number of aspects and results concerning pinwheel, for instance pinwheel as phase fields, pinwheel singularities as well as pinwheels and blow-ups. An important part of the chapter is devoted to the fact that pinwheels of V1 implement a dimensional reduction of the bundle of 1-jets. After reduction, V1 appears to be a field whose singularities are the centers of the pinwheels. These singularities, their normal forms, the distortions and defects of their networks are studied. A way of modeling the orientation maps is to treat them as phase fields which are superpositions of solutions to the Helmholtz equation, whose wave number depends on the mesh of the pinwheel lattice. As illustration an example due to Michael Berry is shown. The chapter ends with the description of data on two aspects of neurobiology: the relation between the cerebral hemispheres through callossal connections and the primary processing of colour in the blobs of V1. One reads 137 references to this chapter. The fifth chapter is devoted to the study of connectivity between various hypercolumns which will help to understand how V1 can integrate local data into global configurations. This is an important geometrical property of the functional architecture of V1. The discussion supported by data and illustrations from a large number of articles, refers to results on: horizontal intracortical connections, integration of contours, contact structure, curvature detectors and 2-jets, structure of the V2 area, colour and area V4, models of direction and singularities of functions and neural morphogenesis and its genetic control. In the final chapter, the author gives a brief preview of the themes to be covered in the second volume and indicates and develops some necessary mathematical prerequisites. An outstanding book.
    0 references
    association field
    0 references
    contact structure
    0 references
    fibre bundle
    0 references
    functional architecture
    0 references
    Gaussian field
    0 references
    geodesic
    0 references
    gestalt
    0 references
    Heisenberg group
    0 references
    Helmholtz equation
    0 references
    Frobenius integrability
    0 references
    jet
    0 references
    Legendrian lift
    0 references
    orientation (hyper)column
    0 references
    parallel transport
    0 references
    pinwheel
    0 references
    receptive profile
    0 references
    sub-Riemannian geometry
    0 references
    transversality
    0 references
    universal unfolding of singularities
    0 references
    wavelet
    0 references

    Identifiers