We Experience The World We Infer, Not The World As It Is

Mar 20, 2015

If you want to understand the human mind, you have to reject the idea that we directly perceive and remember the world as it is. Our perceptual experience isn't simply a passive impression of the input received by our senses — and our memory isn't like a photobook or a video, comprehensively recording the details of our experience.

Take an example from visual perception. The image below seems to depict a woman behind a small girl, where the woman is much larger than the girl. But measure on the screen, and you'll discover that each figure is precisely the same height and width.

What's going on?

The key to understanding this illusion is recognizing that your experience of an object's size is not determined only by the amount of space it takes up on your retina, or (in this case) by the amount of space it takes up on a two-dimensional screen. Instead, your experience of size takes into account an object's inferred distance.

In the image above, the linear perspective leads you to infer that the "big" woman is farther away than the "little" girl. If an object that's farther away takes up as much space on your retina as an object that's much closer, then the farther object must be larger. And that inferred size, which is itself a function of inferred distance, determines your experience of size.

Our experience is so closely tied to what we infer — as opposed to what hits the retina — that it can be surprising to see the two figures when depth cues are removed, as in this version here, which first drops the linear perspective and then shifts the upper figure down to the same vertical position as the little girl:

This basic lesson about immediate visual experience is also true of visual memory. A dramatic illustration comes from the phenomenon of "change blindness": We have the experience of continuously and accurately monitoring our visual environment, but we can fail to notice surprisingly large changes. Consider the following classic demonstration here.

Lest you think you're immune to the phenomenon, take a close look at this video by Richard Wiseman:

It isn't just visual memory that's "gappy" — all of our memories are incomplete. But we're so good at filling in those gaps that we don't always appreciate when we're making an inference and not retrieving an original memory.

To illustrate this final point, think back to the way I described the first illusion (and don't look back at the text, that's cheating!). Did I describe the mother as following the small girl or as looking after the small girl?

In fact, I didn't describe a mother at all — I said a woman was behind a small girl. But it would have been reasonable to infer that the woman was the mother, and that she was following or looking after her daughter. If you incorrectly remembered the description in any of these ways, that could reflect an inference on your part — an inference about the relationship between the two figures involved, and why they were in the particular configuration observed.

The upshot is this: We experience and remember the world we infer. We don't have direct access to the world as it is, no matter how "direct" our perceptual experience and memories can sometimes appear to us.

Yet these illusions of perception and memory shouldn't be taken as evidence that we're out of touch with the world, that we're deluded or deceived. In fact, these illusions tell us something about how we get closer to the world than our limited sensory input might otherwise allow. We're not limited, for instance, by the retinal size a particular image happens to project at a particular distance, or by the way two characters happened to be described in one sentence. Instead, we combine this information with other things we know — about the likely distance of objects in our environment, about why a woman and a girl might be together on path. Sometimes the way we combine information and the inferences we draw can lead us to errors, but much of the time they take us closer to the world as it is, thanks to the world we've inferred.


Tania Lombrozo is a psychology professor at the University of California, Berkeley. She writes about psychology, cognitive science and philosophy, with occasional forays into parenting and veganism. You can keep up with more of what she is thinking on Twitter: @TaniaLombrozo.

Copyright 2017 NPR. To see more, visit http://www.npr.org/.