Essay 7.2 Boundary Extension

Studying the nature of conscious experience is much harder than one might think. You can’t just ask people what they see. Scientists have long realized that the act of measuring something changes what you are measuring. In physics, this only becomes a problem when you start to worry about very small things like the path of an electron. The effects on large scale phenomena will be very small. In perception, however, the effects are massive. If you are asked “Do you see that bird in the tree?” you will direct your attention to the tree, search for a bird, and these deployments of attention will fundamentally change your perception.

One way to avoid changing your perception is to ask what you remember about a scene once it is gone. However, this approach is not without its own pitfalls. Again, how you ask the question can change the answer. Consider picture memory experiments. You get very different answers if you ask observers to recall material (What did you see?) versus asking them to recognize material (Did you see this picture?). Still, you can learn a lot by asking what someone remembers. Look, for example, at the photograph below. Study it well. Now click here to hide the picture, find a piece of paper, and draw what you saw from memory. Then click here to see the photo again, along with a drawing that an experimental participant produced after studying the photo.

A picture from the files of Helene Intraub. Study it for a moment, then click here to hide the picture. Then get a piece of paper and draw what you remember. Click here to see the photo again.

What you should notice is that this observer drew more than was in the original picture. The tips of the fence are not visible nor are the outer sides of the two trash cans. The observer has drawn what is probably there in the world but not what was actually there in the image. Helene Intraub (1989) has called this phenomenon boundary extension. The basic idea is that observers treat the picture as a window onto the “real” scene and, consequently, they infer parts of the scene that are hidden beyond the edges of the picture. Boundary extension tells you something about the constructed nature of scene perception, or at least about the reconstructed nature of scene memory.

In one sense, this behavior is perfectly logical. If you see a friend sitting behind a desk, you automatically infer the rest of the hidden body. You also infer whole books from the visible spines on the bookshelf and, in the case of the picture here, you infer the spiky tips of fences that were almost certainly just outside the camera’s viewfinder when the photo was taken. Thus the surprising thing about boundary extension is not that you make an educated guess about what is hidden, but that you seem quite unsure about the boundaries between the stimulus and the inference—at least when asked to reconstruct the stimulus after the fact.

Reference

Intraub, H., and Richardson, M. (1989). Wide-angle memories of close-up scenes. J Exp Psychol Learn 15: 179–187.