When we look around in a scene, visual information comes to us only through a series of discrete “snapshots” (i.e., fixations). Why do we not perceive this? Why does it appear that we can “see” the scene in its entirety, smoothly and continuously?

To answer this (and other) questions, experimental psychologists usually begin with one or more theories that explain the phenomena at hand. So what we will do is first specify a theory, generate some hypotheses or predictions that the theory makes, and then we design an experiment to test the hypotheses. This is called the deductive approach, i.e., starting with a theory and then predicting data. (We will illustrate another type of scientific reasoning, an inductive approach, later in this unit).

Integrative Visual Buffer Theory

You have seen what happens on the retina as our eyes move. Now, we need an explanation of why we are unaware of these events during normal perception. Fortunately we do not have to come up with a theory from scratch. Such a theory already exists that explains why we seem to "see" the world in its entirety, smoothly and continuously. Moreover, it also explains why the world does not appear to 'jump around' despite that the fact the retinal image shifts its position on the retina from one fixation to the next.

This theory, the integrative visual buffer theory (Breitmeyer, 1984) assumes that with each fixation, all the information on our retinae is encoded and stored in a visual buffer. Moreover, all the objects and features represented in this buffer have codes that specify where they are located in the world. Thus we know not only what objects are in the world but also where those objects are in the world. After the saccade, a new part of the world is foveated and information contained in the new retinal image is “integrated” in the visual buffer.