From: Catherine Woodgold (an588@FreeNet.Carleton.CA) Subject: Vision and learning to read Date: 1999/08/23 (0) introduction (1) How our eyes see (2) Optical illusions (3) Working with colour (0) introduction Some children have difficulty learning to read because the letters seem to be moving. In this article I describe how a person's eyes produce such an effect and discuss the Irlen method of finding the background colour which makes reading easiest. (1) How our eyes see Our eyes are not optimized for reading. Our eyes did not evolve to be excellent at looking at complicated little patterns of a deep, dark black colour on a bright, white background. Rather, our eyes evolved to be good at seeing things such as grey rocks on pink sand, pink rocks on pink sand, green leaves over brown earth, green leaves over other green leaves, and especially moving animals in a stationary forest. When looking at black letters on a white page, nobody's eyes see the white background as uniform. For example, the white area in the middle of the letter "o" might stand out more brightly as if it is a significant object. Such nonlinear effects are more intense for some people than for others, and can change with time; the letters may appear to be moving or getting brighter and darker. Higher-level processing in the brain may cause a person to perceive the background as a uniform white, but the information was not conveyed that way to their brain. Eyes do not simply report the brightness and colour of each point in the visual field to the brain. A lot of processing takes place in neurons in the eyes and on the way to the brain, and further processing happens in the brain before the information becomes available to the conscious mind. The eyes are tuned to note such things as contrast between nearby points; lines and curves; and motion. Light is sensed by four types of sensor on the human retina: rods, which respond to any colour of light; and three types of cones, each of which responds most strongly to one colour, either red, green or blue. Each of the cones also responds somewhat less strongly to other colours nearby in the rainbow to its main colour; thus we perceive yellow when the red and green cones are excited about equally and the blue cones are not. This can happen either when we see real yellow light such as in the yellow stripe in a rainbow, or when we see a mixture of red and green light, as in a yellow graphic on a computer screen. An (alien?) animal with different types of colour sensors might see the two types of yellow as very different colours. When a sensor in the retina reacts to light, it causes the neuron near it to become excited. Each neuron is connected to other neurons around it in a complicated pattern of excitatory and inhibitory synapses. An excitatory connection means that the more neuron A gets excited, then the more neuron B gets excited too. An inhibitory connection is the opposite: the more A gets excited, the more B becomes calm. An excited neuron fires frequently; a calm one fires less frequently. The complex web of neuronal connections involves an almost infinite number of feedback loops. A feedback loop is a system of neurons such that a neuron has an effect on itself, indirectly. For example, neuron A excites neuron B; neuron B excites neuron C; and neuron C excites neuron A again. There are three main types of feedback loops: synergistic, regulatory, and oscillatory. Each type can be composed of two or more neurons, each connected to the next in the sequence and the last connected back to the first in a ring or loop. In a synergistic feedback loop, the neurons cause each other to become more and more excited without bound -- except that each neuron is probably also involved in regulatory feedback loops with other neurons so that in practice there are limits to the level of excitement. Also, each neuron has a physical limit to how excited it can get. A synergistic feedback loop can also go into a negative spiral where each neuron becomes less and less excited. A regulatory feedback loop tends to bring a neuron back to a usual level of excitement. For example, suppose neuron A causes neuron B to become excited, but when B is excited it it sends inhibiting signals to A. Then both A and B will tend to stay at a middle level of excitement. Whenever A gets excited, it excites B, which sends calming signals to A. If A for any reason gets too calm, then B becomes calmer also, and the inhibiting signals from B slow down, so A naturally starts to become somewhat excited again. An oscillatory feedback loop is similar to a regulatory one, except that the neurons over-react to the signals they receive so that instead of settling down to a usual excitement level, they keep going up and down and up and down. For example, suppose A and B are connected with excitatory and inhibitory synapses as in the previous paragraph, but suppose the connections are very strong ones. Then when A gets a little excited, B becomes very excited. B then sends strong signals to A to calm down. A calms down completely. B reacts to this by calming down too. When A notices the lack of inhibitory signals from B, then A becomes excited again, and the whole cycle starts over again. Since each neuron is involved in many feedback loops simultaneously, it is difficult to predict how they will behave. In fact, since neuronal signals can pass from the brain to the eye as well as the other way around, a complete model of the workings of the eye would have to include a model of all the neurons in the brain -- in other words, to completely predict the workings of the eye one needs to know what thoughts are in the brain. However, simpler, incomplete models are easy to do, and effects of each type of feedback loop can be observed. Clearly, the three types of feedback loop are involved in vision. Suppose a person is looking at a white background with a few little black dots on it. At first, each neuron responds to the white with excitement because of the brightness. However, each neuron is wired to inhibit its neighbours, so they quickly calm each other down again, to respond appropriately to a plain field as just unimportant background. Some neurons are wired to respond to differences in the excitement levels of certain other neurons. These become exited near the black dots, since they contrast with the nearby background. This leads to a positive feedback loop where the neurons signalling that there is a contrast cause nearby neurons to become yet more adamant that they are seeing either bright white or dark black. The effect of exaggeration of darkness level near contrasting boundaries can be seen in optical illusion number 1. Dots which are merely gray against the white background will be seen as darker than they are, because the eyes are used to a bright background and the gray contrasts with it; or to put it another way, because our eyes have evolved to respond to objects different from the background. Neurons responding to the edge of a black dot may begin by seeing gray, either because the dot is out-of-focus on the retina or because the neuron is connected to several rods only some of which record black; but the gray is exaggerated in contrast to the white background, so the gray edges are reported as black, and so the black dots appear larger than they are. Now suppose one looks at narrow, equally-spaced black and white stripes. Our eyes are not designed primarily to look at stripes; our eyes are designed to see objects in front of backgrounds. So at first, the stripes may look like black stripes on a white background. But if the white is interpreted as a background colour, then white is ignored and anything different from white is exaggerated. Neurons sensing the edges of the stripes report them as different from the background. The black stripes look a little wider than they actually are. But then the visual field has more black than white on it. All of a sudden it seems that the black is the background, with white stripes in front of it. Then white becomes the interesting colour, the black background is ignored, and neurons near the edges begin to report that they're seeing something different from the black background. The white stripes appear slightly wider. And so it repeats. The result is that the edges of the stripes appear to be moving slightly. They may oscillate back and forth, perhaps several times a second. At a given time, one part of the visual field might look like white stripes on a black background while another part looks the other way around, and then they may switch. The ultimate result may be that the person says "my eyes hurt", and looks away. (2) Optical Illusions Illusion number 1: look at a grey square on a white background and another grey square on a black background. Even if the two grey squares are identical, the one on the white background looks darker. This illustrates how our eyes exaggerate contrasts. Illusion number 2: Stare at pure red lettering on a pure blue background (or vice versa). The edges of the letters may seem to start to vibrate; or one's eyes may just hurt looking at this colour combination. This illusion doesn't work if there are narrow boundaries of white or black etc. around the edge of each letter. Actually, it's difficult to find lettering of those colours on such things as the covers of books, or packages in the grocery store. The combinations red-black, red-white and blue-white are far more common. I believe this is an attempt to avoid the unpleasant effects of the apparently wiggly letters. I don't know why the blue-red combination tends to produce this effect. However, the same effect can occur, at least for some people, with black-and-white. (3) Working with colour People who have trouble reading black-on-white letters can often benefit from a change in lighting or in the colour of the background or of the lettering. Pale green paper with black letters is easier for many people than white paper. However, this is an individual thing and the pale green is actually worse for some people than white. Pink, beige and many other colours can be tried out to see which works best for an individual. Many children tend to read in dim light. Certain children are often found by their parents reading with only the light from a window or dim light from another room, and the parent then turns the light on. Changing the brightness of the light or the colour of the light changes the input to the neurons of the eyes. For example, black letters on a red background are seen using primarily the red-sensitive cones in the retina, while anything read in dim light is seen primarily with the rods, not the colour-sensitive cones. Changing the input changes the response of the neurons. Thus a person's eyes which respond to red-and-blue patterns with oscillations in the excitement levels of neurons may happen to respond more calmly to green-and-black patterns. The educator Helen Irlen discovered that many children learning to read had difficulty perceiving printed letters, and worked out a system of tests to determine whether a child has such difficulties, and what are the best colours for that child to use. The child can then cover printed pages with transparent coloured sheets, or wear coloured eyeglasses. (See the book "Reading by the Colors" by Helen Irlen.) These effects have nothing to do with how well the image is focused on the retina, i.e. with whether the child needs ordinary glasses with lenses. They are the effects described in the first section of this article. Large print also helps, until the child reaches the level of reading skill where a quick glance at a word results in instant recognition of the word, so that it is not necessary for the eyes to be able to focus on each individual letter; the child may then be able to read small print, but may still have trouble reading nonsense, unusual names, etc., or learning to spell. Note that it is better to learn to read phonetically than to start by memorizing what each word looks like. The authors of "Teach Your Child to Read in 100 Easy Lessons" claim that of thousands of children taught by their method, practically none had dyslexia or similar problems. They use phonics heavily, and have a very, very gradual introduction of new skills; for example no more than about one new letter of the alphabet is taught per day. They also use large print and gradually introduce the child to successively smaller print. Once a person has learned to read, they can recognize words in printing so small that they cannot clearly see each individual letter. (That book doesn't discuss colour. See Helen Irlen's book "Reading by the Colors".) -- Cathy Woodgold TISSATAAFL Ottawa, Ontario, Canada http://www.ncf.carleton.ca/~an588/ an588@freenet.carleton.ca Inability is an abstract thing involving comparison with alternate universes; it cannot be experienced.