Aries Arditi - US grants

The proposed work will study the relationship between field defects in low vision patients and performance on everyday tasks important to mobility, reading, and successful use of optical aids. We will also develop procedures for scoring ordinary visual field maps that are specific to these tasks. The study will concentrate on four visual abilities that are expected to mediate performance on the specific tasks: 1) detecting, 2) finding, 3) scanning stationery objects, and 4) tracking moving objects. Our performance measures are 1) a self-report questionnaire designed to probe visual function in everyday circumstances, 2) direct observation of the subject's visual function both indoors and outdoors, 3) mobility on an indoor test course, 4) reading ability, and 5) psychophysical measures of scanning and tracking performance. The long-term objectives are to 1) provide a better understanding of the relationships between field defects and visual performance, 2) describe the factors that determine successful use and acceptance of low vision aids, and 3) provide clinicians and rehabilitation professionals with tools that can better predict the client's functioning in everyday life.

This application seeks further support for the development of a wayfinding system specifically for building interiors of the facilities of public accommodations (e.g. office buildings, transportation facilities, hospitals, etc.) that will increase accessibility of those facilities to visually impaired users and serve as an environmental rehabilitative intervention for vision loss. This computerized system, called Pathfinder, interacts with users using artificial speech, voice recognition, electronic tactile touch tablet, and a high contrast visual CRT display. Embodying principles of universal design and intended to be useful for all users, including fully-sighted, the system will also increase accessibility for wheelchair users and hearing-impaired persons. Phase I results demonstrated feasibility with blind users unfamiliar with one floor of a Manhattan office building, and showed that the system is capable of training users with no experience with visual or tactile maps on its use without human intervention. Phase II research will address further development of the physical design of the product, the user training module, the tactile and voice recognition interfaces, and the visual display for low vision. Additionally, computer programming tools for building map and signage design professionals will be developed and tested. The product, along with an instructional package will be commercialized in Phase III.

DESCRIPTION (provided by applicant): This proposal has the long-term goal of understanding how people with low vision analyze and process two-dimensional layout of images and scenes at multiple scales. The overarching theme of the proposed studies is perception of spatial layout in low vision, particularly on layout in two dimensions such as, for example, perception of a magnified or unmagnified computer screen, or of a monocular image of a visual scene through an optical or video distance telescopic aid. Using subjects with a highly prevalent form of low vision, central visual field loss (CFL), and normally-sighted subjects under both normal viewing conditions and simulated CFL, two main issues will be addressed using psychophysical methods: Global layout perception: Image element identification and accuracy of element localization will be measured with three kinds of (unmagnified) images: web pages, natural urban scenes and colored drawings composed of basic shapes. Exposure duration will also be varied in order to assess the degree to which spatial layout can be retained pre-attentively from glimpses of such images. Linking global and local views: The strategies people use in attempting to link objects in global view to objects they re-image within a magnified image view will be examined, by 1) measuring subjects' accuracy in assessing the amount of magnification required to perform focal visual tasks based on information that is sub threshold with respect to those tasks, by 2) assessing ability to learn to assess magnification needs accurately based on feedback, and by 3) assessing subjects' abilities to gauge direction and size of pointing errors based on feedback from a magnified view, as a function of magnification.

[unreadable] DESCRIPTION (provided by applicant): This application seeks support for exploratory research to develop a user interface (UI) specifically designed for moderate to severe low vision, that will allow improved access to structured documents such as those written in hypertext markup language (HTML) or extensible markup language (XML). The UI will incorporate intelligent, well-accepted principles and empirical knowledge from vision and rehabilitation sciences, and will be initially embodied in a web browser for low vision. With an emphasis on simplicity and ease of use, the configurable features of the UI will be limited to those that have the highest impact on legibility and reading performance, and will be designed so that once configured, no adjustments will be needed, regardless of what document the user is viewing. The UI will allow users simultaneous access to both the global features of web documents as the author intended, along with enlarged text tailored to the user's needs; this will allow access to all web pages without requiring that sites have special markup or authoring. The design will be motivated by the corpus of existing of health sciences based psychophysical research in low vision, and will utilize recent software engineering advances including XML-based User Interface Language (XUL) and eXtensible Application Markup Language (XAML), eXstensible Style Sheet Language (XSL) and XSL-Transform, the Document Object Model (DOM), and the general trend toward separation of content and structure in electronic documents. The technology and research developed in this project is expected to lead to general user interfaces for low vision that might be incorporated into operating system design, as well as electronic books and web documents. The broader impact of this project is that it can help close the digital divide for millions of users with moderate and severe vision loss who are at a relative disadvantage in access to information. In addition, the technology can, if widely adopted, alleviate the failure of most authors of structured documents to comply with best practices for authoring accessible content, because document pages with reduced visual accessibility can now be more easily appreciated. This project is intended to enhance web and document accessibility to the growing number of older visually impaired people who are reluctant to learn to use large, feature-laden software. Since the technology can be easily extended to any language, such solutions can, assuming computers, electronic documents and web access are available, enhance the lives of tens of millions of users worldwide. [unreadable] [unreadable] [unreadable]