Tactile maps and diagrams need to be carefully designed in order to be readable by the visually impaired user. In the Tactile Inkjet Mapping Project, we attempt to provide a new orientation for tactile map design through a rigorous programme of experimental research on design issues (cognitive tactualization approach). In this article we discuss three of our studies. The first study evaluates the use of different background materials for tactile graphics. The second study looks at the minimal perceptible distance between two lines. The third study investigates the optimal elevation (height off the paper) at which graphics are produced. We argue that experimental studies like these represent a necessary basis for the design of tactile maps. Introduction Tactile maps and diagrams have been available in some form for centuries, and now have increasing significance in the lives of blind and visually impaired people for education, work and leisure, as graphical information becomes ever more prevalent and central in our daily experience. Methods for the production of tactile materials continue to be devised, and a range of materials and processes is currently available, the most recent of which (e.g. microcapsule paper, Tiger printer) allow computer-derived images to be printed more or less directly in tactile form. Nevertheless, however accurate and nuanced the printing of the relief image may be, a map or diagram is only as good as its original design. A poorly designed map is an illegible map, no matter how well it is printed. A direct translation of a visual image into a tactile image generally produces inadequate results. This is because the tactile modality imposes very different requirements with regard to spatial layout and information content of a map or diagram. In recognition of this fact, a number of guidelines on tactile map design have been published (Amick and Corcoran, 1997; Bentzen, 1997; Edman, 1992; Eriksson, Jansson and Strucel, 2003; Gardiner and Perkins, 2003). However, these largely reflect experience-based assumptions and aesthetic judgements rather than empirical research. Although these may well result in good ideas and practical solutions (especially when a feedback loop involving users of the materials is included), this trial-and-error approach is generally also quite hit-and-miss. This state of affairs is not unique to tactile images: cartographic design and the design of diagrams have generally been based on aesthetic decisions of cartographers and designers, some unique design insights (e.g. Harry Becks’ London Underground diagram) and, occasionally, on feedback from users of the materials. The last 20 years have seen a shift within visual cartography from a practice-led approach, based on aesthetic judgements and experience-based assumptions, to a more scientific approach, which has attempted to reassess the design of maps by drawing on findings, and employing research methodologies, from cognitive psychology (Keates, 1982; MacEachren, 1995). For example, MacEachren discusses, in the context of cartographic legibility, research on how components of a visual scene are detected, discriminated and identified. This research, which can be termed ‘cognitive visualization’, recognizes that reading a map or diagram involves a set of perceptual and cognitive processes. An empirical approach on the design of tactile maps and diagrams: The cognitive tactualization approach by Sandra Jehoel, Don McCallum, Jonathan Rowell and Simon Ungar © 2006 SAGE Publications The British Journal of Visual Impairment • Volume 24 • Number 2 • 2006 • ISSN 0264-6196 67 DOI:10.1177/0264619606063402