Neural plasticity underlies the capacity of the central nervous system to encode new information, develop new abilities and adapt to the environment. Plasticity is required for learning and is modulated during development and by disorders of the brain. Recent advances in experimental methodology have led to new insights on the biological mechanisms underlying neural plasticity. The topics if the papers chosen for review will center on recent experimental and theoretical studies of topics such as synaptic plasticity, developmental and activity-dependent changes in sensory and motor maps.

This course deals with the science and engineering of computer vision, that is, the analysis of patterns in visual images of the world with the goal of reconstructing and understanding the objects and processes in the world that are producing them. The emphasis is on physical, mathematical, and information processing aspects of vision, but biological and psychological perspectives will also be considered. Topics covered include image formation and representation, multi-scale analysis, segmentation, contour and region analysis, reconstruction of depth based on stereo, texture shading and motion, and analysis and recognition of objects and scenes using statistical and model-based techniques.

10-701 Machine Learning: 12 units
(Cross-listed as 15-781 for CS PhD students only.)

15-785 Computational Perception and Scene Analysis: 12 units
(Cross-listed as Psychology 85-485/785)

Instructor: Mike Lewicki
Date/Time: Tue & Thu, 10:30 AM - 11:50 AM
Location: Porter Hall A19
Prerequisites: CS 15-385 (undergraduate computer vision course), Psych 85-370 (undergraduate perception course), or permission of the instructor.
The goal of this course is to teach how to reason scientifically about problems and issues in perceptual cognition, how to extract the essential computational properties of those abstract ideas, and finally how to convert these into explicit mathematical models and computational algorithms. The course teaches advanced aspects of perception, scene analysis, and recognition in both the visual and auditory modalities, concentrating on those aspects that allow us and animals to behave in natural, complex environments. Both the experimental approaches of scientific disciplines and the computational approaches of engineering disciplines are emphasized.

Each topic in the course begins by studying the ethology of natural behaviors, analyzing and decomposing these to identify the essential components that are required for the total behavior in a natural environment. This aspect of the course follows the lines of scientific reasoning and key experiemental results that lead to our current understanding of the important computational problems in perception and scene analysis. The course then surveys the most important solutions to these problems, focusing on the idealizations and simplifications that are sensory coding, perceptual invariance, spatial vision and sound localization, visual and auditory scene segmentation, many aspects of attention, and the basics of objects and speech recognition.

This course will provide an overview of parallel-distributed processing models of aspects of perception, memory, language, knowledge representation, and learning. The course will consist of lectures describing the theory behind the models as well as their implementation, and students will get hands-on experience running existing simulation models on workstations.

This course allows the student to explore ways in which the mind shapes language and language shapes the mind. Why are humans the only species with a full linguistic system? Some of the questions to be explored are: What kinds of mental abilities allow the child to learn language? What are the cognitive abilities needed to support the production and comprehension of sentences in real time? How do these abilities differ between people? Are there universal limits on the ways in which languages differ? Where do these limitations come from cognition in general or the specific language facility? Why is it so hard to learn a second language? Are there important links between language change and cultural change that point to links between language and culture?

This seminar will examine how the brain executes higher level cognitive processes, such as problem-solving, language comprehension, and visual thinking. The topic will be addressed by examining what recent brain imaging studies can tell us about these various kinds of thinking. This new scientific approach has the potential of providing important information about how the brain thinks, indicating not only what parts perform what function, but also how the activity of different parts of the brain are organized to perform some thinking task, and how various neurological diseases (e.g. aphasia, Alzheimer's) affect brain activity. A variety of different types of thinking will be examined, including short-term working memory storage and computation, problem solving, language comprehension, visual thinking. Several different technologies for measuring brain activity (e.g. PET and functional MRI and also some PET imaging) will be considered, attempting to relate brain physiology to cognitive functioning. The course will examine brain imaging in normal subjects and in people with various kinds of brain damage. Graduate Students Only.

85-785 Computational Perception and Scene Analysis: 12 units
(Cross-listed as Computer Science 15-785)

The famous psychologist George Miller once said that Psychology should "give itself away." The goal of this course is to look at cases where we have done so -- or at least tried. The course focuses on applications that are sufficiently advanced as to have made an impact outside of the research field per se. That impact can take the form of a product, a change in practice, or a legal statute. The application should have a theoretical base, as contrasted, say, with pure measurement research as in ergonomics. Examples of applications are virtual reality (in vision, hearing, and touch), cognitive tutors based on models of cognitive processing, phonologically based reading programs, latent semantic analysis applications to writing assessment, and measurses of consumers' implicit attitudes. The course will use a case-study approach that considers a set of applications in detail, while building a general understanding of what it means to move research into the applied setting. The questions to be considered include: What makes a body of theoretically based research applicable? What is the pathway from laboratory to practice? What are the barriers - economic, legal, entrenched belief or practice? The format will emphasize analysis and discussion by students.

This course covers the mechanisms of human motor systems and control, using arm movements as an example. The course starts from the anatomy of muscles, sensors, spinal cord, and brain; then functional analysis of these system components will follow. After system analysis, all components are integrated to study feedback control dynamics. Using physiological studies such as psychophysical and lesion experiments, the course covers classic to modern theories of how the nervous systems may control movements. Advance topics include adaptation, representation, coordinate systems, cognitive involvement, and rehabilitation techniques for motor impaired patients.

PSY 2476 Seminar in Cognitive Psychology - Computational Modeling CR HRS: 2.0-3.0

This course will provide an introduction to the methodology of computational modeling in cognitive psychology. The main objectives of this course are to foster a basic understanding of the various approaches to modeling and an appreciation of the many practical and philosophical issues that modelers must consider when they develop their models. The first part of the course will focus on several of these issues, including (but not limited to) the following questions: (1) What are computational models of cognition? (2) What are the major approaches (e.g., production systems, PDP, etc.) that have been used to model cognitive processes? (3) How are models designed and tested? (4) How are models evaluated? The second part of the course will address these issues in a more concrete manner by way of two comparative analyses of models that have been developed to account for phenomena in two areas of cognitive research: general episodic memory (e.g., recognition, recall, etc.), and eye-movement control during reading. The third part of the course will consist of student-led discussions of seminal modeling papers from the students' areas of research. During the final part of the course, students will be required to complete a modeling project or to write a comprehensive review of existing models within their area of research. Although this class assumes no prior knowledge of modeling, basic math proficiency and computer programming skills are necessary.

In this course we will examine the functioning of neurons and synapses, the basic units responsible for fast communication within the nervous system. The course will focus on the elegant use of electrical mechanisms by the nervous system, and on the powerful quantitative approach to scientific investigation that is fundamental to neurophysiology. Topics that will be addressed include: principles of electric current flow exploited by the nervous system; the basis of the resting potential of neurons; the structure and function of voltage-gated and neurotransmitter-gated ion channels; generation and propagation of action potentials; the physiology of fast synaptic communication.

This course is a component of the introductory graduate sequence designed to provide an overview of neuroscience. This course provides an introduction to the structure of the mammalian nervous system and to the functional organization of sensory systems, motor systems, regulatory systems, and systems involved in higher brain functions. It is taught primarily in a lecture format with some laboratory work. The course covers in detail the major sensory, motor and behavioral regulatory systems of the brain. The course satisfies the CNBC core requirement in neuroanatomy.