We use images in a way that psychologists would describe as Stimulus-Response. If I tell you that the Big Dipper points to Polaris which is to the north, then I have imparted some knowledge to you. But there are several hurdles to jump before you have gained anything of value: you have to remember the facts; you have to be able to decode them (what part of the Big Dipper points to Polaris?); you have to recognize a situation in which you can use them (looking at the night sky); and you have to apply them successfully in that situation.

Our research and anecdotal experience suggests that very few people will successfully clear all of these hurdles, with the result that a large proportion of what people 'learn' from books is lost.

Suppose instead that I simulate the environment where you will have a chance to use the knowledge, in this case a realistic image of the sky, and have you find the Big Dipper, find Polaris and find north. The next time you look at the night sky you may be reminded of the image I showed you. Whether you can apply your skill in the real sky will depend on how similar what I showed you was to the real thing. (Though there is some evidence that the simulation does not need to be perfect: software designers often use 'paper prototypes' to gauge use reactions without actually building the software, see Snyder, 2003).

Microworlds
The images in this example are filling at least three roles. First they form a 'microworld'--a simplified learning environment where you can concentrate on getting one this right at a time. Modern methods to teach people to ski create a microworld where everything is the same as real skiing except the skis, which are a lot shorter and so far easier to handle. Beginner skiers get to practice correct technique without falling over all the time (Burton, Brown, Fischer, 1984).

Second, each image is a stimulus, like the smell of meat powder that caused Pavlov's dog to salivate. When you next see the stimulus (or something similar enough to it to have the same effect, in this case the real night sky), your earlier response it triggered. In other words, you are reminded of the image I showed you but you would not necessarily be reminded of a written or verbal description of the same thing. If this is what happens, you have cleared all the hurdles to getting real value from the new knowledge.

Third, the images perform the more usual role in educational material: they help explain the text. Research has shown that illustrations significantly aid learning (Levin, Anglin, Carney, 1987).

Images in Stikky Weight Management
As another, less obvious, example, take the images of opportunities to exercise in Stikky Weight Management such as a view of stairs and an escalator next to each other after a description of the benefits of even light exercise. People know that walking up stairs is better exercise than taking the escalator, but that doesn't often change their behavior. Our aim with Weight Management was to imprint the situation in your mind so that you would recall it automatically when faced with the stairs/escalator view in the real world. We hope you will be reminded of the image in the book and generate the response, "maybe I should take the stairs." Of course, you may still choose the escalator but at least you considered the alternative, in contrast to people who read about the benefit but never link that to the action when given the choice.

Similar are the images of burgers and fries in Stikky Weight Management. Some test readers of an early draft suggested they should be removed since they made the reader feel hungry which was, they said, surely the opposite of the intended effect.

But that suggests that these readers had been 'conditioned' to think "mmm" whenever they see a picture of a Big Mac. If that is the case, they will either have to avoid pictures of Big Macs for the rest of their life or they need to be unconditioned so that their response becomes "600 calories" or "sat fat in a bun" or whatever. That is what we tried to do. (So, incidentally, we did not use manufacturer's images, which tend to make food look better than it is in reality. We had the food re-shot in a realistic way.)

The real question, of course, is whether it worked for you. Let us know by email at theory@stikky.com.

Links to more articles for educators and learning theorists here.

References

Richard R Burton, John Seely Brown, Gerhard Fischer (1984), 'Skiing as a Model of Instruction', in Barbara Rogoff and Jean Lave, Everyday Cognition, toExcel

J R Levin, G J Anglin and R N Carney (1987), 'On Empirically Validating Functions of Pictures in Prose', in D A Willows and H A Houghton (Ed) The Psychology of Illustration, Volume 1, Springer-Verlag

Carolyn Snyder (2003), Paper Prototyping, The Fast and Easy Way to Design and Refine User Interfaces, Morgan Kaufmann