Currently my lab is interested in two important pieces of this puzzle. First, peripheral vision efficiently encodes its inputs using a scheme that preserves a great deal of useful information, while losing the information necessary to perform certain tasks used to probe the details of visual experience. We have developed the state-of-the-art model of peripheral encoding, and to date have tested this model on over 70 visual tasks. Because behavioral studies have often confounded peripheral vision with visual attention, our new understanding of peripheral vision necessitates rethinking what we know about attention from the ground up. Second, many tasks used to demonstrate impoverished vision are arguably inherently difficult; poor performance on these tasks may indicate general-purpose limits on the complexity of the tasks one can perform at a given moment. It seems promising that together these two components can make sense of a wide variety of phenomena, including vision’s marvelous successes, its quirky failures, and our rich percept of the visual world.