A Thousand Brains: A New Theory of Intelligence: Summary & Key Insights

Imagine closing your eyes and reaching for your coffee mug. You know where it is, how it feels, even what temperature to expect. That knowing is not stored as a static image but as a dynamic, predictive model of reality—an internal representation built from experience. In the cortex, perception is not passive. Every sensory input is interpreted through a model the brain has already built. Seeing, hearing, and touching all are acts of prediction.

I explain that the neocortex constructs models layer by layer, using sensory data and movement to refine understanding. This modeling is how we make sense of a complex world. When light hits your eyes or sound reaches your ears, your brain does not just record those signals—it integrates them into a spatial and temporal understanding of the world around you. The world you perceive is, in essence, your brain’s best guess at reality, continuously updated.

This process is essential for survival. You don’t experience the world as raw data; you experience meaning, and meaning is the result of internal models. These models aren’t static—they are constantly evolving as your sensory experiences change. The brain’s job is to minimize uncertainty by predicting accurately. Every time a prediction fails, the model adapts. Through this process, your brain becomes a master at understanding the environment—not by memorizing, but by modeling.

In neuroscience we know the cortex is organized into repeating structures called columns. What I discovered is that each of these columns learns complete models of objects independently. This means that your brain doesn’t have one unified model stored in a single place; instead, thousands of columns each hold their own version of understanding.

When you touch a coffee mug with your fingers, each fingertip activates different columns that learn the shape, texture, and spatial layout of the mug. Over time, these columns combine their local models into a coherent whole. This is revolutionary because it suggests that intelligence is distributed. Every column is a little scientist, observing, learning, and hypothesizing about what it is sensing.

In this framework, intelligence is not the property of a specific part of the brain—it emerges from the collaboration of these countless small learning units. The Thousand Brains Theory flips the traditional view: instead of a hierarchical top-down structure, the cortex is a democratic system of many local experts voting on reality.