Human vision is profoundly non-uniform. Spatial resolution, contrast sensitivity, colour discrimination decrease, and the appearance of visual features become increasingly distorted with retinal eccentricity, yet visual experience appears remarkably rich and stable across the entire visual field. This apparent paradox has been a central challenge in vision science: how can a perceptually rich world emerge from a sensory system that samples only a small fraction of the environment with high precision? We argue that the apparent richness of peripheral vision cannot be explained solely by cognitive biases. Instead, it reflects genuine perceptual processes. Information sampled in central vision can be extrapolated to peripheral appearance, while prior knowledge and learned regular-ities allow the visual system to infer missing or distorted peripheral information. These processes operate within the constraints of active vision, where eye movements selectively sample task-relevant information rather than building a complete detailed representation of the scene. We propose that rich peripheral experience emerges from the interaction between centre to periphery extrapolation, and learned predictions, allowing the visual system to maintain a coherent and useful representation of the world.