Geoffrey Boynton - US grants

The first objective of this research proposal is to test a model of human pattern vision mechanisms with psychophysical results from a contrast matching task. A high-contrast background lowers the apparent contrast of a texture patch. A model of neural contrast normalization may explain this illusion, known as simultaneous contrast-contrast. The model posits that each cell in primary visual cortex is suppressed by the pooled activity of a large number of other cells. If perceived contrast depends on responses from these cortical cells, the normalization model predicts that background contrast should have a divisive effect on the perceived contrast of the central patch. The second objective is to search for the neural substrate of contrast perception. Recently researchers at several institutions have begun using a new kind of brain imaging technique, functional magnetic resonance imaging (fMRI) which, when coupled with a perceptual matching paradigm can bridge the gap to link perceptual appearance with brain activity. Neurons in the posterior calcarine have receptive fields in the central visual field. If contrast appearance is mediated by cells in primary visual cortex, then the fMR signals will reflect the changes in perceived contrast induced by the simultaneous contrast illusion.

Consider creating (a) a computer system to help physicians make a diagnosis using all of a patient's medical data and images along with millions of case histories; (b) intelligent buildings and cars that are aware of their occupants activities; (c) personal digital assistants that watch and learn your habits -- not only gathering information from the web but recalling where you had left your keys; or (d) a computer tutor that watches a child as she performs a science experiment. Each of these scenarios requires machines that can see and learn, and while there have been tremendous advances in computer vision and computational learning, current computer vision and learning systems for many applications (such as face recognition) are still inferior to the visual and learning capabilities of a toddler. Meanwhile, great strides in understanding visual recognition and learning in humans have been made with psychophysical and neurophysiological experiments. The intellectual merit of this proposal is its focus on creating novel interactions between the four areas of: computer and human vision, and human and machine learning. We believe these areas are intimately intertwined, and that the synergy of their simultaneous study will lead to breakthroughs in all four domains.

Our goal in this IGERT is therefore to train a new generation of scientists and engineers who are as versed in the mathematical and physical foundations of computer vision and computational learning as they are in the biological and psychological basis of natural vision and learning. On the one hand, students will be trained to propose a computational model for some aspect of biological vision and then design experiments (fMRI, single cell recordings, psychophysics) to validate this model. On the other hand, they will be ready to expand the frontiers of learning theory and embed the resulting techniques in real-world machine vision applications. The broader impact of this program will be the development of a generation of scholars who will bring new tools to bear upon fundamental problems in human and computer vision, and human and machine learning.

We will develop a new curriculum that introduces new cross-disciplinary courses to complement the current offerings. In addition, students accepted to the program will go through a two-week boot camp, before classes start, where they will receive intensive training in machine learning and vision using MatLab, perceptual psychophysics, and brain imaging. We will balance on-campus training with summer internships in industry.

IGERT is an NSF-wide program intended to meet the challenges of educating U.S. Ph.D. scientists and engineers with the interdisciplinary background, deep knowledge in a chosen discipline, and the technical, professional, and personal skills needed for the career demands of the future. The program is intended to catalyze a cultural change in graduate education by establishing innovative new models for graduate education and training in a fertile environment for collaborative research that transcends traditional disciplinary boundaries. In this sixth year of the program, awards are being made to institutions for programs that collectively span the areas of science and engineering supported by NSF

This award provides funds to permit the University of Washington to acquire a 3T magnetic resonance imaging (MRI) scanner for basic-science brain imaging research. This instrument will be housed in a dedicated MRI research facility in the Health Sciences building where it will support a primary group of over 30 basic-science MRI users at UW whose ongoing research success depends critically on access to a 3T MRI scanner. These multidisciplinary researchers originate from multiple different departments that include Psychology, Speech and Hearing Science, Music, Computer Science, Physiology and Biophysics, Linguistics, Neurosurgery, Neurology, Bioengineering and Radiology. Thus, this 3T facility will provide campus-wide support to neuroscience researchers in the Schools of the Arts and Sciences, Education, Engineering, Medicine, the Health Sciences, the School of Public Health and the Graduate School (Psychology). In addition to the current MR researchers, there are many other neuroscientists at UW who have expressed interest in using the 3T facility for their research when available. Thus, addition of a 3T MRI facility will expand the functional brain MRI research efforts at UW in the future. It will also be a resource accessible to researchers from other regional institutions.

Functional brain imaging (fMRI) is an important new research tool used to define how the brain operates to control motor and cognitive tasks and how these functions may be affected by abnormal development or disease. The 3T MRI will be utilized for many fMRI studies some of which include visual system analysis to understand how the brain operates to simultaneously process diverse types of visual input that includes spatial information (location) as well as feature information (color, shape, motion, direction, etc) to interpret a visual scene. Studies of fMRI involving language and cognition will investigate how the brain codes successes and errors during trial-and-error learning to see whether these neural coding events can be used to predict how well people learn. Neuroimaging using structural MRI and MR spectroscopy (MRS) to measure brain metabolite changes will be used to compare brains of young children with normal development to those with Autism. Quantitative MRI/MRS analyses of grey matter and regional volumetric and chemical changes over time (age 3 to age 10) will be correlated with behavioral developmental changes to address questions regarding the time course of brain development and its relationship to typical and atypical behavior.

The University of Washington is strongly committed to interdisciplinary science and education, as evidenced by strong cross-campus research ties, interdisciplinary training programs such as the Graduate Program in Neurobiology and Behavior, and discipline-spanning research centers such as the Integrated Brain Imaging Center (IBIC), Institute for Learning and Brain Science (ILABS), Virginia Merrill Bloedel Hearing Research Center (VMBHRC), Center on Human Development and Disability (CHDD), and Washington National Primate Research Center (WaNPRC). The proposed 3T facility will operate in this tradition, providing new opportunities not only for disciplinary and cross-disciplinary research, but also for graduate and post-graduate training, educational outreach, recruitment of new faculty in neuroimaging, and involvement by female students and researchers, and those from underrepresented groups.

DESCRIPTION (provided by applicant): Visual attention is the process of selecting a subset of incoming visual information for further processing. Attention is normally allocated either to a particular location in space (spatial attention), or to a specific feature, such as color or motion, at that location (feature-based attention), or to some combination of the two. While research has tended to focus on how the brain enhances or modifies the representation of visual stimuli within the spatial focus of attention, recent electrophysiological and neuroimaging evidence shows that attention can also affect the neuronal response to an ignored visual stimulus, if it shares common features with an attended stimulus. Also, preliminary studies show that attention to an auditory stimulus can also have a different effect on the brain's response to an ignored visual stimulus. The brain's processing of unattended information therefore depends strongly on what is being attended elsewhere. The goal of the proposed research is to use a combination of functional MRI and behavioral methods to obtain a better understanding of how the brain's representation of unattended stimuli depends on the location, feature, and modality attended to elsewhere. Findings from the proposed research could have significant impact on our understanding of attentional disorders, such as autism and ADHD, which may involve a disruption in the ability to ignore distracting sensory information.

DESCRIPTION (provided by applicant): Humans are excellent at selecting the relevant part of a cluttered visual scene or the relevant conversation at a noisy party. In contrast, humans are often not so successful at dividing attention over multiple stimuli. One cannot read two books at once and is it is not wise to talk on a phone and drive at the same time. Much has been learned about the effects of attention on physiological responses in the human and monkey visual cortex. However nearly all of this work has addressed selective attention, which is when attention is directed to one source of information over another. In general, studies of selective attention have shown that activity in many areas of the brain is greater for a stimulus that is relevant to the current task compared to a stimulus that is not relevant. Surprisingly, very littl is known about the effects of divided attention - paying attention to more than one thing on a time - on neuronal responses. This lack of a physiological literature is particularly surprising given the long history of research on the effects of divided attention on behavioral performance. Interestingly. these behavioral studies show a wide range of effects: for discrimination of simple features there can little cost to attended to multiple stimuli at a time, whereas for higher-level perceptual tasks such as reading words it may impossible to attend to more than one stimulus at a time. Here we propose a series of behavioral and imaging studies to examine the physiological basis of divided attention. We will (1) examine what factors in a task result in a cost when dividing attention. In particular we will examine whether it is the complexity of the stimulus or the task that is the critical factor for both a simple grating task (Specific Aim 1) an complex lexical task (Specific Aim 2). Second we will determine the cause of reduced neural responses and impaired behavioral performance when attentional capacity is limited. In particular, we will determine whether attentional limitations are due to attenuation of attentional gain, a shift to serial processing or suppressive interactions between stimuli. Finally we will examine the spatial profile of attentional modulations during divided attention: whether it is spread broadly across space and/or features or allocated discretely. This gap in the literature is of clinical importance. Individuals with autism spectrum disorder and ADHD show differential divided attention effects: a deeper understanding of the mechanism underlying divided attention is likely to prove critical in linking these behavioral differences to underlying neurophysiologica mechanisms.

With support from the NSF Major Research Instrumentation Program to the University of Washington (UW), Drs. Geoffrey M. Boynton in the Psychology Department and Patricia K. Kuhl, co-director of the Institute for Learning & Brain Sciences, will purchase a Siemens MAGNETOM Prisma 3-Tesla Magnetic Resonance Imaging (MRI) scanner for human neuroimaging research. MRI is the primary tool for investigating the human brain. This instrument will be shared and available for scientists from a wide range of departments including Psychology, Speech and Hearing, Radiology, Psychiatry and Computer Science. This state-of-the-art device has four functions for studying the human brain: structural imaging, functional imaging, diffusion weighted imaging and magnetic resonance (MR) spectroscopy. Structural imaging is using the MRI scanner to acquire images of brain structure which can be used to study, for example, how the brain structure differs across different populations, or how age and experience affect the growth of the human brain. Functional imaging, or "functional magnetic resonance imaging, (fMRI)" is used to measure where and when brain activity occurs as human subjects perform a task or experience sensory stimulation. fMRI relies on the MRI scanner's ability to measure changes in the level of oxygenated vs. deoxygenated blood that are caused by changes in brain activity. Diffusion weighted imaging can detect the major neuronal pathways in the brain by measuring the way in which water molecules naturally diffuse along these pathways. Finally, MR spectroscopy measures the levels of neurotransmitter concentration, such as the inhibitory neurotransmitter GABA throughout the brain. Different levels of GABA have been found to vary, for example, between subjects with autism and neurotypical subjects.

This new MRI scanner will be the primary device for studying a wide range of human neuroscience problems. These include studies of linguistic function and dysfunction, cognition, developmental cognitive neuroscience, sensory neuroscience, neurological disorders, and studies of social cognition including autism. The four functions described above will be used to study a wide variety of subjects including children, adolescents, subjects with autism, attention deficit disorder (ADD), Alzheimer's, blindness, dyslexia, and Parkinson's disease. For example, fMRI will be used to study the brain's response while subjects with ADD attempt to ignore a visual stimulus. In another example, diffusion MRI will be used to study changes in prefrontal executive function networks in children during language development. The new scanner will form the centerpiece of a thriving and growing human interdisciplinary neuroscience community and will further strengthen an existing culture of training in neuroimaging data acquisition and analysis methods. Through a program of courses and hands-on training, we will provide the training that is needed for students and postdocs across all disciplines to master the advanced imaging techniques necessary to launch successful independent careers in the human neurosciences.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

ABSTRACT The field of sight restoration has made dramatic progress over the last decade. Two types of retinal implants have been commercially approved, and several other designs are in development worldwide. In addition, two groups are actively implanting and developing cortical electronic implants. The first optogenetic clinical trial has begun, with many others likely in the next two years. Within a decade, many blind individuals are likely to be offered a wide range of options for sight restoration that depend on widely different technologies. Interactions between implant electronics and the underlying neurophysiology of the retina or cortex mean that the vision provided by most of these technologies will differ substantially from normal sight. The question of this proposal is ? What role can cortical plasticity play in helping patients make use of this artificial visual input? Over the past 15 years our research group has been generating computational models, developed using a combination of physiological and psychophysical data, which can predict the percepts that patients might experience for a variety of sight recovery technologies. We propose to use these models to simulate, within visually normal participants, four critical neurophysiological distortions inherent in sight restoration technologies: Aim 1. Abnormal neuronal population responses during retinal stimulation: Simultaneous stimulation of on and off cells. Aim 2. Spatial distortions: Stimulation of retinal ganglion cell axons. Aim 3. Abnormal cortical neuronal population responses: Distortions induced by the V1 neural architecture. Aim 4. Temporal blurring due to slow optogenetic kinetics. Our goal is to use normally sighted participants, viewing distorted visual input, as ?virtual patients? to learn which spatiotemporal distortions can be compensated for by plasticity, and which must be compensated for in device design. This will provide device manufacturers with a more nuanced understanding of the abilities and limits of visual perceptual adaptability. Finally, this work will provide novel insights regarding the fundamental mechanisms of cortical plasticity by asking whether, in adulthood, it is possible to reconfigure the fundamental building blocks of visual perception?