How to draw the blind

I met Betty in 1973 when I was looking for participants in a research project on tactile perception. The young girl from Toronto (Canada) was blind from the age of three. I was all the more astonished when she told me that she enjoyed portraying her relatives in profile - as a toddler she couldn't have practiced that.

Up until then I had assumed that drawing was about seeing, because noises, smells or tactile impressions are generally not painted without professional artistic ambitions, but the visible world is. But now I learned that blind people can also enjoy it and even have a talent for it, like Betty, who felt the volumes and outlines and converted them into lines on paper using her imagination (Fig. 1).

The encounter made me curious whether blind people in general can draw something recognizable for the sighted - and if so, to what extent such and ordinary images are similar. It was also of practical importance to find out whether they understood the many pictorial symbols and graphic signs of the seeing world, converted into tangible patterns, and how they would deal with them.

To do this, I used a wide variety of special drawing materials and tactile models, for example flat wire constructions and, above all, rubberized, plastic-coated special writing boards from the Swedish Blind Society, on which a raised line remains when you run a hard pen over them.

Much of what my colleagues and I have discovered in such experiments and investigations over the past 20 years have fundamentally changed our understanding of sensory perception. The fact that blind people implement and abstract what is represented in a similar way to sighted people, using the same pictorial shorthand, as it were, is probably the most important finding: They too draw, when they want to represent an object, its outline as a contour; and they reproduce spatial depth by letting lines converge or compressing surfaces. They also lay out the partial aspects of complex scenes in such a way that they all appear to be seen from a single location. They indicate movement with long drawn out or confused lines. They even represent symbolic content in a comparable way, even if the visual appearance is not quite as usual, for example a heart as a sign of love or a star as that of luck.

Recognize contours

From the beginning I was mainly interested in how many blind people can read from the pure outlines of objects - especially from profile drawings of faces - and also whether there are individual differences. For this purpose, over the years I have had many of my test subjects from North America and Europe draw the contours of a wide variety of objects and heads.

The good results encouraged me to try further together with Yvonne Eriksson, who works at Linköping University (Sweden) as well as at the Swedish hearing and braille library for the blind. In 1993, three men and six women from Stockholm were willing to do this, four of whom were blind from birth, three had lost sight as children after the age of three and two had only very little residual vision.

The test objects were four caricatures that psychology student Hans-Jørgen Andersen from the University of Århus (Denmark) had bent from thin wire and stuck on a blackboard (Fig. 3). It was important to recognize the most striking feature in each case. We told the participants about the four characteristics beforehand: grin, curly hair, beard, giant nose. Five of the subjects, including a man who had been completely blind since birth, assigned everything correctly. However, one of the participants couldn't do anything with the pictures either. The mean for the group was 2.8 correct assignments. The fact that the task can be difficult even for sighted people was shown when 18 students from Toronto tried the same blindfolded: They too got a group average of only 3.1 points.

Our experiments, as well as numerous others by colleagues in the United States, Japan, Norway, Sweden, Spain, and Great Britain leave little doubt that blind people can see the outlines of familiar objects. How can someone achieve this intuitively who has never seen in his life? Basically the answer is simple: A simple schematic drawing usually shows what the blind person feels on the object, namely corners and edges, furrows, ridges or other elevations as well as the boundaries of surfaces and volumes on a more complex structure.

However, there are individual differences. They are likely to be related to when someone went blind. Some people who could never see, or only in their very early childhood, actually find it difficult to depict images like the ones mentioned above. However, a team led by Yatuka Shimizu from the College of Technology in Tsukuba (Japan) found that 60 percent of the test subjects from this group of early blind people were able to make out outlines of everyday objects, such as a fish or a bottle. It is not surprising that sighted people who wore a blindfold in the test did a lot better.

Interestingly enough, however, at a later age blind people were superior to even blindfolded people with normal vision, as Morton Heller from the State University in Winston-Salem (North Carolina) discovered. Perhaps they will benefit from the fact that, on the one hand, they had visual experiences earlier and, on the other hand, because of their handicap, they trained their tactile skills specially.

Perspective drawing

The blind can do even more when abstracting spatial impressions. I was brought up by Kathy, a thirty year old woman from Ottawa, Canada. Because of a cancer of the retina that appeared in the first year of life, she had never been able to see clearly and at the age of three she had completely lost her eyesight. Nonetheless, she not only made pretty good outline drawings on the board, on which raised lines form, but was also remarkably able to depict spatial distances. For example, if I placed a cube and an angle piece next to each other or offset them in front of them, in the first case they both drew the same size, in the second the more distant building block was markedly smaller (picture 2 below).

Thus she had followed the principle of perspective painting, according to which an object occupies a smaller angle of the field of perception with increasing distance (which can be easily understood if one looks down a long street and the parallel edges seem to be tending to a point on the horizon). Correspondingly, when an object is touched close to the body, it takes up a large section of the pivoting range of the arm, and further away a smaller section.

But how much do the blind really understand about the optical perspective and the possibilities of translating it into the surface? Together with Paul Gabias from Okanagan University College in British Columbia (Canada), who was at the University of New York at the time, I tested 24 people who were blind from birth with the five tactile sketches in Figure 5.First, we presented them with the left one, a square with four symmetrical lines radiating out at the corners, and said that someone who was also blind had drawn such a table and wanted to express that it was built symmetrically; Another blind man had found an identical representation, but explained that it was the view from below, because otherwise - i.e. from the side or from above - the plate and the four legs could not be reproduced so well together.

Now we presented the other four pictures, each supposed to represent a cube, and asked which one was most likely to have been drawn by the second person (the first corresponds to a flat unfolded box; the second could see the front and top of the cuboid at the same time below 45 degrees, the third one from a flatter angle, the last one the view as with the second one, but with a shortened perspective).

Most of the test subjects were most likely to trust the second fictitious person to make the fourth drawing - that is, the one who supposedly understood better how to grasp a perspective projection. On the other hand, they were least able to imagine that this blind man had converted a cube into a cross of five squares, which did not even suggest perspective. They assigned this sketch primarily to the other allegedly blind draftsman, because he attached great importance to symmetry.

In addition, Heller and I used a test that was originally developed by the Swiss cognitive psychologist Jean Piaget (1896 to 1980) to examine the mental comprehension of children and called the three-mountain task: We placed three bodies on a table - a cuboid, a cone and a sphere - offset as in Figure 4 on the right. The subjects were placed on one side of the table; You should feel the arrangement and then draw from there as well as from the other sides and from above (the four side views are shown on the left in Figure 4). If you consider that this often overwhelms even adults with normal vision, it is all the more remarkable that the blind scored just as well with an average of 3.4 out of 5 possible points as a control group of sighted people.

In a second attempt the point was to indicate the point of view of the viewer for each of the five drawings; each picture appeared twice in a random sequence, so that a maximum of ten correct answers were possible. The blind test persons achieved an average of 6.7 points; the control group was only slightly better with 7.5.

Once again, people who became blind late even - albeit slightly in this case - performed better than the sighted: These nine people achieved a value of 4.2 in the drawing task and 8.3 in the second experiment. This supports the assumption that memory of optical impressions and a refined sense of touch worked together.

Image imagery

Even for the blind, a simple sketch may contain more than a mere image of an object. I suddenly became aware of this when one of my test subjects once drew a wheel on her own initiative and indicated its rotation with a sweeping line within the circle. This amazed me all the more because such techniques for giving the impression of movement came up very late. For example, Wilhelm Busch (1832 to 1908), who had a decisive influence on the development of humorous illustrations through to cartoons and animated films, only put movement on the paper after about 1877, according to an investigation by the art historian David Kunzle; At that time, many of his picture stories such as "Max and Moritz", "Hans Huckebein, the unlucky raven" or "The pious Helene" had already been completed.

I wanted to get to know the idea of ​​the blind better. When I then asked my test subjects to draw a wheel in swing, they repeatedly provided a circle with inner arc lines like curved spokes. When asked about this, they explained such graphic forms as expressions of dynamism - which one can easily understand. But why does it have to be such an arch? Wouldn't broken or other troubled lines do the same? This should prove another attempt that Gabias and I devised.

This time we presented five palpable sketches of wheels with different types of spoke patterns - made of curved and wavy lines, kinked or interrupted lines and straight lines that protruded from the circle (Fig. 6). They were assigned to idle, very fast, steady, bumpy or slowed rotation.

With only one exception, all 18 test persons associated each of the five drawings with a certain form of movement - and they decided predominantly like the majority of just as many normal-sighted students at the University of Toronto, and with hardly a lower degree of agreement. Since the blind were certainly hardly familiar with the visual implementation of such a phenomenon, it is astonishing that they solved this problem without further ado - and mostly in the same way as sighted people.

Kathy again gave cause for another investigation of visual imagery: once she drew a child's bed in a heart (picture 2). This means, she said, the love that surrounds the child.

Then I started exploring with the Chinese doctoral student Chang Hong Liu how much access the blind have to such pictorial symbolisms. A common psychological test is to assign opposing terms to two different geometric shapes, such as the properties of hard and soft, a square and a circle. We selected 20 pairs of terms which sighted test participants related more or less in agreement to the shapes (Fig. 7). Four blind test persons made very similar decisions, which certainly allows the conclusion that such shapes also mean the same thing to them in a figurative sense.

Mechanisms of perception processing

Even if one usually thinks that the mind opens up the shape of things primarily through the eye, everyone knows that it grasps many things through literal comprehension. In general, there is a close relationship between the two sensory systems: You can often imagine how something would feel when you see it - and vice versa. But it is still astonishing that a line felt with the finger and a line seen are interpreted in the same way. You get closer to an explanation if you consider what information these sensory impressions actually contain.

For example, what does the line of a contour drawing convey? According to the common understanding, it stands for the outline of an object and is understood as such by the perception system - at least a thin line. But it's not that simple, because every line actually has two edges; if they are far enough apart, i.e. if the line is thick enough, they are also perceived separately. The increasingly thicker contour of a profile can ultimately appear as a sketch of two faces - or like that of a head with its shadow (Fig. 8).

What do the blind perceive? I investigated this with Sanne, a student at Aarhus University. She scanned several profile representations of different widths, the inner and outer edges of which were raised. At a distance of one millimeter between the two lines, she perceived one face, at eight millimeters, however, already two.

According to a further thesis, outlines generally stand for perceptible borders, including those within a depicted object that can be seen but not touched, for example for the transition from light to shadow. Many examples of this can be found in commercial graphics and art. But that this is generally not true, is illustrated in Figure 9. There are two ways of presenting a photo of me: The drawing on the left shows hard flat contrasts, the drawing on the right only the border lines between the white and black fields. This makes the portrait as such completely unrecognizable. A line as a substitute for the transition from light to shadow is therefore not always sufficient for perception to recognize an object - not even here, where the actual contours of the head are largely missing.

One could even say that the shadow areas in the left portrait are practically particularly thick lines, i.e. those with irregular and widely spaced edges, while the line edges in the second are closely parallel. Here, too, it becomes clear that a graphic element that can be perceived as two boundaries is interpreted differently than when it is understood as just one. And this obviously applies to the tactile as well as the visual system. Accordingly, there should be a processing unit that is generally responsible for taking contour impressions from complex scenarios and interpreting them independently of the sensory modality and physical reality. It is not based solely on visual aspects such as brightness and color, but takes a black line, for example, which is set off on both sides against the white background, as the edge of a structure or surface. And what such an optical limit is for the sighted is for the blind person, for example, a raised line that can be traced with the fingertip.

Insofar as the principles that are effective here are not solely of a visual nature, the brain region involved could also be called multi- or - as is common practice - amodal. In my book on the subject, which appeared in 1993, I discussed this aspect in more detail. Visual and tactile impressions seem, as I explain there, to be combined into such spatial parameters as foreground and background, overlaps, straight and uneven surfaces or the location of the viewer.

While sighted people receive information about the environment from both sensory systems and are calculated together, the brain area apparently also functions when visual impressions are completely absent. How else can it be explained that blind people often understand the pictorial content of contours and other graphic elements quite well and, conversely, are able to represent spatial structures adequately in two dimensions, and can even record and reproduce symbolic pictorial content.

All of this encourages them to make suitable material available to them, for example to experience works of painting, which is already happening.For example, an organization in the USA dedicated to the education of the blind and associated with the Whitney Museum of American Art and the Museum of Modern Art in New York has tactile versions of graphics by the French Henri Matisse (1869-1954) and of prehistoric ones Cave drawings made. Perhaps soon, images converted into raised lines will be just as common as Braille text.


- Picture and Pattern Perception in the Sighted and the Blind. By M. A. Heller in: Perception, Volume 18, Issue 3, pages 379 to 389, 1989.

- Drawing and the Blind: Pictures to Touch. By J. M. Kennedy. Yale University Press, 1993.

- Symbolic Forms and Cognition. By C. H. Liu and J. M. Kennedy in: Psyke og Logos [Denmark], Volume 14, Issue 2, pages 441 to 456, 1993.

- Tactile Pattern Recognition by Graphic Display: Importance of 3-D Information for Haptic Perception of Familiar Objects. By Y. Shimizu and others in: Perception and Psychophysics, Volume 53, Issue 1, pages 43 to 48, January 1993.

- Meaning, Presence and Absence in Pictures. By J. M. Kennedy and others in: Ecological Perception Research, Visual Communication, and Aesthetics. Edited by K. Landwehr. Springer, Berlin, Heidelberg 1990.

- Buttons and shapes. Art and art lessons with the blind. Edited by K. von Spitzer and M. Lange. Association for the Promotion of the Blind Education, Hanover 1988.

From: Spektrum der Wissenschaft 3/1997, page 84
© Spektrum der Wissenschaft Verlagsgesellschaft mbH

This article is contained in Spectrum of Science 3/1997