Bio Mirror / Regional
Cortical Architecture
The neocortex is strikingly uniform: six layers, the same cell types, the same local connectivity motif, across visual, auditory, motor, and prefrontal areas. Regions differ mainly in their inputs and outputs. That uniformity is the strongest structural argument that a single learning principle might underlie very different capabilities — and it is why a stacked, homogeneous architecture like a transformer is a less absurd model of cortex than it first appears.
The cortex repeats one circuit motif everywhere. What differs is what it is wired to.
The Cortical Column
A repeated six-layer motif with a stereotyped input, processing, and output division.
Biology
The column is the cortex's repeated computational unit. Its layered structure implements a consistent division: receive, transform locally, send onward, and send a signal back down the hierarchy. That descending pathway is substantial — feedback …
In AI
A transformer block is also a repeated, uniform unit stacked many times, with a fixed internal division of labor: route between positions, then transform in place, then pass on. The uniformity argument is the same …
In Robotics
A layered control architecture where each level operates on a different timescale and abstraction — reflexes at the bottom, trajectory planning above, task planning above that — with information flowing in both directions. The bidirectional …
The Visual Hierarchy
Edges to textures to objects, across a chain of areas — the closest thing to a confirmed prediction in this field.
Biology
Each stage builds more abstract and more invariant representations from the previous one. Early stages encode local oriented structure; middle stages encode texture and contour combinations; late stages respond to objects and faces largely independent …
In AI
Convolutional networks reproduce this progression without being told to. First-layer filters in a trained CNN are oriented edge and color-opponent detectors closely resembling V1 measurements; deeper layers become texture-selective and then object-selective. Trained CNN activations …
In Robotics
A perception stack layered from features to objects to scene graph to affordances. The engineering motivation is identical — invariance built in stages, so each stage solves a tractable piece — and modern stacks have …
Attention Networks
Biological attention is a resource allocator with a physical bottleneck. Machine attention is a routing mechanism. The shared name hides a real difference.
Biology
Attention allocates limited processing capacity. Attending to a location or feature increases the gain of neurons representing it and suppresses competitors, improving the effective signal-to-noise ratio for the attended item at the direct cost of …
In AI
Self-attention shares the name and part of the concept — content-based selection of what to read — and the query-key-value formulation is a reasonable abstraction of "search for what matches my current need". The correspondence …
In Robotics
Active perception: a robot with a steerable camera and a compute budget must decide where to look and what to process, and it faces the biological version of the problem rather than the transformer version …