Introduction

We often talk about perception as a processing pipeline that only moves in one direction: photoreceptors to bipolar cells to retinal ganglion cells to LGN to V1 to V2, and so on. However, this simple “feed-forward” version of the story is not quite right.

Perception is iterative and information moves both up and down the processing hierarchy in waves of activation. Initial feed-forward sweeps of processing lead to subsequent “reentrant” feedback sweeps of further processing, and so on. Our perception of objects changes over time as our brain seeks out details and clarifies the initial information it received.

If it is true that perception involves reentrant processes, then there should be some way to see those processes in action. Vincent Di Lollo, James Enns, and Ronald Rensink discovered the phenomenon of object substitution masking, which provides just such a demonstration.

If you click the Easy button, you should not have much trouble perceiving the shape inside the four dots. If you click the Hard button, you should find it more difficult.

What’s the difference between the easy and the hard version of the display? One obvious difference is that the four dots stay on for longer in the hard version (in fact, that’s the only difference!). Why would having the four dots stay on longer make it harder to perceive the shape?

Instructions

For this activity, stare at the fixation point in the middle of the white box, and then click either the Easy or the Hard button. A number of shapes will flash on the screen, and one shape will be surrounded by four dots. The goal of the exercise is to identify which shape is surrounded by the dots.

Object Substitution Masking Explanation

In the easy version of the demo, the four dots disappeared at the same time as the rest of the objects. When your perceptual system’s reentrant processing took a second look at the cued location, everything was gone except for your rapidly fading visual memory of the display (i.e., the visual icon representation). Even though that representation is very unstable, it is enough for you to clearly see which shape had the four dots around it.

In the hard version of the demo, on the other hand, the four dots disappeared long after the rest of the objects. When your perceptual system’s reentrant processing took a second look at the cued location, everything was gone except for the four dots. Since four visible dots are a much stronger stimulus than the rapidly fading visual memory of the shape, you experience seeing the four dots rather than the shape you were looking for. In other words, the four dots acted like an object that was substituted for the object you were trying to see, thus masking the target object!