A TOUCH TABLET ENHANCED WITH SYNTHETIC SPEECH AS A DISPLAY FOR VISUALLY IMPAIRED PEOPLE'S READING OF VIRTUAL MAPS

Emily Holmes (1) and Gunnar Jansson (2)

(1) Arlene R. Gordon Research Institute, The Lighthouse Inc., New York, NY, 11211, USA. emily@optica.lighthouse.org. (Reported research conducted while at the University of Uppsala.)

(2) Department of Psychology, Uppsala University, Box 1225,S- 75142 Uppsala, Sweden. gunnar.jansson@psyk.uu.se

Web Posted on: December 2, 1997

For sighted people, maps are common aids both for planning a visit to an unknown area and to guide travel when in the area. However, people with vision impairments are routinely denied access to the use of visual maps. Sighted people can also retrieve information from virtual maps from Geographic Information Systems (GIS), for instance, the street network of a town. In order to be accessible for people who are blind the map information must be presented via hearing and/or haptics. The aim of the reported study was to investigate the potentials of a touch tablet with a tactual overlay consisting of either a tactile grid or tactile map enhanced with synthetic speech.

An orientation aid for people with vision impairments

This investigation was part of an effort to develop an orientation aid for visually impaired people by the MOBIC consortium (see Acknowledgments). The aid consists of two parts, a planning system and a system to be used during travel. We will discuss the planning system. With this users can retrieve information from several sources, including GIS enhanced with information specific to the needs of visually impaired travelers. The system can also calculate a route when the users specify the journey start and destination.

Tactile Maps

Ideally, people with vision impairments should have access to virtual map information which is as versatile as a visual computer screen display, which allows zooming and rapid changes between maps. Unfortunately, there is no device for soft copies of tactile maps with sufficient spatial resolution and size(see Jansson, 1991a). Currently, hard copy displays have to be used, that is, embossed maps with variations in elevation over the background. There are several production methods (cf. Edman, 1992); here swellpaper (also called microcapsule paper) was used, that is a special paper producing an embossment of a black pattern when it is heated (Goto, 1983).

There are few tactile maps made for planning a route in an unfamiliar environment. However, even if available, they may be difficult to read. Typical problems for the user include difficulty in tactually discriminating symbols, clutter in the map due to the detail of information needed, and lack of training in how to interpret raised line graphics. Further, the majority of visually impaired people do not read Braille, which makes Braille labeling often useless. Enhancing the map with auditory information may solve these problems, at least partially.

Touch tablet

A touch tablet is a device consisting of a flat horizontal surface divided into a matrix of smaller units. It is connected to a computer that registers when one of these units is pressed, resulting for example in a message spoken with synthetic speech. In the context of graphic information for visually impaired people NOMAD (Parkes, 1988)and AUDIO-TOUCH (L'tzsch, 1994) have been developed. We used an Informatrix 4096 Touch Tablet (Semerc Co., UK), composed of 64 x 64 units that cover an A3 area (38.5 x 27 cm). Software, called Explore, see Holmes, Michel & Raab (1995), was designed so the digital map could be explored.

Using touch tablets to display route planning maps made from digital data can help solve many of the problems of tactile maps: these maps can "talk" by pressing down anywhere on the tablet. This verbal information deals with problems of symbol legibility and Braille labels, and can also provide navigational instructions. The user can rapidly obtain the precise map wanted by specifying the start and finish of their journey. The computer system can plan a route and provide verbal instruction. The map can optionally be printed out on swellpaper to make a tactile map overlay for the tablet.

In terms of the users' perception: firstly, information about spatial layout can be gained by the haptic sense, (with sensors in the skin, joints, muscles and tendons) when moving over the tactile display and when pressing information units. Secondly, auditory information is also gained on pressing units. Currently the tablet "beeps" indicating that a press was registered, and says verbal pieces of information about the spatial layout of the route and environment. Thirdly, according on what overlay is placed on top of the touch tablet, different amounts of tactile information is available.

Tactile overlays on the touch tablet - Touch tablet with tactile grid

A raised line, grid matrix overlay e.g. 8 x 8 cells can be placed on top of the touch tablet. The tactile grid is intended to facilitate route exploration compared to only having an empty touch tablet, by providing a spatial frame of reference within which to keep track of finger movements. The grid is not dependent on what map is used. It is therefore an economic and practical alternative to any system which requires the production of specific tactile map overlays. The map scale can be indicated by specifying what the dimensions of one grid cell represents. For example in the experiment reported below they correspond to approximately 70 x 50 meters.

Touch tablet with tactile map overlay.

A simple tactile map overlay can be created from the GIS data. It consists of just the street network of the route area and information points containing verbal directions along the route. Streets are displayed as double line symbols, their separation varying relative to the actual street. Information points are shown by circular, solid point symbols. The map overlay's advantage is that in addition to perceptually providing haptic and verbal information it allows direct overview of the spatial structure of the map tactually through the finger tips. The tactile presence of the map can also facilitate exploration of the interface as points may be more readily located or relocated than with the grid.

Experiment

Problem.

The aim of the experiment was to compare the two types of tactile overlay on a touch tablet, concerning their usefulness in providing map information for visually impaired readers planning a new route. We especially focused on the effect the users' spatial representation of the route and some performance measures.

The importance of having a good spatial representation for the guidance of moving in the environment may seem intuitively self- evident, but relatively little has been established in the context of guidance without vision (cf. Strelow, 1985; Jansson, 1991b). However, it can be expected that a spatial representation of a route is important for travelers' confidence and understanding of what they are doing, as well as allowing them to reorient if lost.

It is also far from self-evident how the spatial understanding should be measured (cf. Siegel, 1981). However, in this study the experimental measures used covered aspects of

• (1) the participants' appreciation of distances in the mapped area from either a path following or a birds eye view perspective,
• (2) the participants' appreciation of directions in the mapped area using a pointing task,
• (3) their ability to reorient on the map when positioned in a place off the route,
• (4) the time they spent studying the route before confident to walk it,
• (5) their explicit knowledge of the route through describing it verbally,
• (6) their ability to represent the route by drawing it, and finally
• (7) their ability to walk the route.

Method.

Two route maps were used of approximate scale 1:1500. Each participant explored one route with the tactile grid and the other route with the tactile map overlay on different days, in a counterbalanced order of presentation. Three congenitally, totally blind and one adventiously totally blind participant took part. Their age ranged between 21 and 43 years. None had used a touch tablet previously and only one had traveled independently on routes they had not learnt with a mobility instructor. Participants explored routes to two cafes just less than a kilometer from a starting point. When they felt confident to navigate the route they were tested on the seven experimental tasks mentioned above.

Results.

To summarise, participants were successfully able to explore the maps with both interfaces and navigate the route. However, there were advantages of using a touch tablet with a tactile map overlay rather than a grid overlay: Less exploration time is needed until the map explorer feels confident to navigate. Their ability to reorient between two points on the route on the map is also facilitated. This measure indicates both that the usability of the interface and the usability of the mental representation developed of the route is more efficient. Participants appreciation of relative distances in the route was also more accurate, as was their judgment of the directions between points.

Conclusion.

Overall this indicates that the map explorer can build up a richer mental representation of the route, which is more similar to a "bird's eye view", after using a touch tablet with a tactile map overlay. Use of a grid overlay may limit the explorer to a mental representation from a "path following" perspective. This may allow them to walk along a route but may not be as effective for making corrections if the traveler became disorientated. In other words, it is more limited to just following the path in a given direction, and the explorers overall understanding of the area is not as rich as when using a tactile map overlay.

Acknowledgments

MoBIC is an acronym for Mobility of Blind and Elderly People Interacting with Computers. The consortium consists of Universität Magdeburg (Institut für Simulation und Graphik), Magdeburg, Germany; Freie Universität Berlin (Institut für Informatik), Berlin, Germany; F H Papenmeier Gmbh & Co K G, Schwerte, Germany; Royal National Institute for the Blind, London, UK; University of Hertfordshire (Psychology Division), Hatfield, UK; University of Birmingham (Research Centre for the Education of the Visually Handicapped), Birmingham, UK; British Telecom Labs, Ipswich, UK; and Uppsala University (Department of Psychology), Uppsala, Sweden. The project was supported by the European Commission under the TIDE program (Technology Initiative for Disabled and Elderly persons). The Swedish participation was funded by Uppsala University and the Swedish Work Environment Fund. For further background of the MoBIC project, see Jansson (1995). The authors are indebted to the other participants in the consortium for many discussions. The Uppsala co-workers Cecilia Bringhammar and Anders Jansson have made important contributions.

References

Edman, P.K. (1992). Tactile graphics. New York: American Foundation for the Blind.

Goto, R. (1983). New techniques for tactual graphics. In J.W. Wiedel (Ed.), Proceedings of the First International Symposium on Maps and Graphics for the Visually Handicapped (pp. 113-125). Washington, DC: Association of American Geographers.

Holmes, E., Michel, R. & Raab, A. (1995). Computerunterstützte Erkundung digitaler Karten durch Sehbehinderte (Computer supported exploration of digital maps for visually impaired people). In W. Laufenberg & J. L'tzsch (Eds.), Taktile medien. Kolloquium über tastbare Abbildungen für Blinde (pp.81-87). Marburg, Germany: Deutsche Blindenstudienanstalt e. V., Carl- Strehl-Schule.

Jansson, G. (1991a). Possibilities and limitations of tactile computer graphics for the blind. In Y.Queinnec & F.Daniellou (Eds.), Designing for everyone (Vol.2,pp.1616-1618). London: Taylor & Francis.

Jansson, G. (1991b). The control of locomotion when vision is reduced or missing. In A.E. Patla (Ed.), Adaptability of human gait. Implications for the control of locomotion (pp. 333-357). Amsterdam: North-Holland.

Jansson, G. (1995) (Ed.). Requirement for effective orientation and mobility by blind people. London: Royal National Institute for the Blind.

L"tzsch, J. (1994). Computer aided access to tactile graphics for the blind. In W.Zagler, G.Busby, & R.Wagner (Eds.), Computers for handicapped persons. Proceedings, 4th International Conference on Computers for Handicapped Persons,Vienna,1994 (Lecture Notes in Computer Science 860, pp. 575-581). Berlin: Springer.

Parkes, D. (1988). Nomad, an audio-tactile tool for the acquisition, use and management of spatially distributed information by visually impaired people. In A.F. Tatham & A.G. Dodds (Eds.), Proceedings of the Second International Symposium on Maps and Graphics for Visually Impaired People, London, 1988.

Siegel, A.W. (1981). The externalisation of cognitive maps by children and adults. In E.B. Liben, A.H. Paterson, & N. Newcomb (Eds.), Spatial representation and behavior across the life-span. New York: Academic Press.

Strelow, E.R. (1985). What is needed for a theory of mobility: Direct perception and cognitive maps Lessons from the blind. Psychological Review, 92, 226-248.