When we look at an image, a set of interconnected neurons will fire. These, collectively, constitute what we are 'recognizing' about the image. But what does each individual neuron 'recognize' - that is, what sort of image is that particular neuron attuned to? The natural inclination is to say it's faces, or textures, or shapes - but no. Using a set of autogenerated images, researchers determined what an individual neuron reacts most to, and it's some sort of Gothic abstraction (pictured). "Experience, for one, must play a role in shaping the neuron's preference ... yet, since the neurons didn't evolve images of faces as we know them, perhaps their so-called vocabulary is looser than we thought, or perhaps it is not even fixed." The full paper is a pretty tough read, but this summary is a lot more accessible.
Today: 5 Total: 26 [Direct link] [Share]
