Cuttings and Embankments
Visual Clutter in the Topographic Base of Geological Maps
Colour vs. Visual Texture for Area Symbols
These experiments set out to improve a specific pair of map symbols: the symbols for railway cuttings and railway embankments. Although the same symbols are found on topographic maps in almost any part of the world, there is clear evidence that they are easily confused with each other. For example, many geography undergraduates cannot correctly say which is which.
An experimental psychologist and five graphic designers at the Royal College of Art worked together to develop alternative pairs of symbols. Tests were conducted both on school children and experienced map readers. The illustration shows a detail from one of the test maps with one of many alternative pairs of symbols.
(Reproduced from the 1975 Ordnance Survey 1:50 000 map with the permission of the Controller of Her Majesty's Stationery Office, Crown copyright reserved.)
A range of tests were developed to evaluate alternative symbols. In this 'intervisibility' task, people were asked to imagine the four situations depicted: in two cases the railway workers can see each other, and in two they cannot. Pairs of railway workers were represented on test maps by pairs of crosses and people were asked whether or not they could see one another. One advantage of this type of question is that it does not make any use of the words 'cutting' and 'embankment' which may interfere with some people's understanding of the spatial situation.
For full details see -
Phillips, R.J., Coe, B., Kono, E., Knapp, J., Barrett, S., Wiseman, G., & Eveleigh, P. (1990). An Experimental Approach to the Design of Cartographic Symbols. Applied Cognitive Psychology 4, 485-497.
Abstract - Graphic designers and an experimental psychologist worked together to improve the design of two map symbols which are frequently confused: the symbols for cuttings and embankments on topographic maps. The problem was analysed in terms of the function of the symbols and their likely cognitive representations. Tests were developed to evaluate alternative designs, including an intervisibility task which required users to visualise the landform from the symbols viewed in the context of a map. Tests were given to schoolchildren and to experienced map users in order to compare the standard symbols with five alternative designs. Children's performance was strongly affected by the symbols they used, but experienced users were much less affected. After some refinement of the symbols a further experiment demonstrated the superiority of a number of alternative designs over the existing symbols on a range of tests: scores were almost double on the intervisibility task. The paper makes recommendations to cartographers and argues for greater consideration of the inexperienced map user in the design process.
ERRATUM IN THE PUBLISHED PAPER: in Figure 1 the words 'cutting' and 'embankment' should be reversed.
Full paper as pdf file (1.41MB).
Geological maps are arguably the most complicated visual displays in common use and so they were a good subject for an experiment to understand the nature of visual clutter. But this experiment also tackles the practical problem of how best to simplify the topographic base on geological maps.
Here is a small detail from one of the test maps used in the experiments and examples of the five types of topographic base which were compared: 'full', 'line', 'point', 'minimal' and 'design'. These bases were printed in grey and geological information was superimposed. The experiment compared the relative importance of line symbols and point symbols in causing visual clutter, but account was also taken of the importance placed on different topographic symbols by professional map users.
(Reproduced from the 1978 Institute of Geological Sciences and the 1975 Ordnance Survey 1:50 000 map with the permission of the Controller of Her Majesty's Stationery Office, Crown copyright reserved.)
For full details see -
Phillips, R.J. & Noyes, L. (1982). An Investigation of Visual Clutter in the Topographic Base of a Geological Map. Cartographic Journal 19 (2), 122-132.
Abstract - Visual clutter on maps is a familiar experience but its precise nature is only poorly understood. Clutter was investigated in an experiment using a 1:50 000 geological map. Twelve representative map reading tasks were used to compare map reading performance on maps which differed only in their topographic base. The aim was to assess the effect of removing topographic symbols which are of only minor importance to the map reader. This reduction in visual clutter significantly improved performance on a number of the questions. Some evidence was obtained to support the hypothesis that line symbols clutter other line symbols, and point symbols clutter other point symbols, but there is little effect between the two. In practical terms the removal of minor point symbols and type led to larger improvements than the removal of minor line symbols, even though more of the latter were deleted. The relevance of the experiment to other geological maps, and to maps in general, is discussed.
Full paper as pdf file (1924KB).
Thematic maps, such as geological maps and land-use maps, often use symbols to categorise areas of land. These experiments are concerned with the relative merits of using colour and visual texture as area symbols on thematic maps. Some experimental displays were created which always included 16 different area symbols, but employed area codes based on either colour alone, or visual texture alone, or a combination of the two.
These are small details from the displays used in the experiments. Each makes use of 16 types of area symbol. The first uses 16 colours, and the last uses 16 textures. The second combines eight colours with two visual textures. The third combines two colours with eight visual textures.
For full details see -
Phillips, R.J. & Noyes, L (1980). A comparison of colour and visual texture as codes for use as are symbols on thematic maps. Ergonomics, 23, 1117-1128.
Abstract - Three experiments compared colour and texture as methods of coding area symbols for thematic maps. Most previous research has been limited to displays with, at most, eight codes. This study employed displays containing 16 types of symbol coded either by colour, texture, or a non-redundant combination of the two. Symbols coded by colour or a combination of colour and texture were much easier to find than symbols coded by texture alone. Point symbols were easier to locate against a coloured background than a textured background. Texture codes may be slightly easier to remember than colour codes but the difference, if any, is small.
Full paper as pdf file (1.16MB)