Ori Brafman - The Starfish and the Spider

It was like a game of Wheres Waldo. But instead of kids playing the game, the players were the worlds leading neuroscientists. And instead of Waldo, they were looking for a curly-haired, sweater-wearing grandma. Anyones grandma.

The neuroscientists were trying to answer what at first appeared to be a simple question. We all have memorieswhether of our first day at school or of our grandmother. But where, the scientists asked, do these memories reside? Little did they know that they were about to arrive at a conclusion that would have surprising implications, not only for biology but also for every industry in the world, for international terrorism, and for a host of far-flung communities.

Scientists had long assumed that our brains, like other complex machines, had a top-down structure. Surely, in order to store and manage a lifetime of memories, our brains needed a chain of command. The hippocampus is in charge, and neurons, which store specific memories, report up to it. When we recall a memory, our hippocampus, acting like a high-speed computer, retrieves it from a specific neuron. Want to access a memory of your first love? Go to neuron number 18,416. Want to access a memory of your fourth-grade teacher? Go to neuron number 46,124,394.

In order to prove this theory, the scientists needed to show that certain neurons are activated when we attempt to retrieve a particular memory. Beginning in the 1960s, scientists wired up subjects with electrodes and sensors and showed them pictures of familiar objects. The hope was that each time a subject was presented with a picture, a specific neuron would be triggered. Subjects spent hours staring at photos. The scientists watched and waited for specific neurons to fire. And they waited. And they waited.

Instead of a neat correlation between particular memories and particular neurons, they found a mess. Each time subjects were presented with a picture, many different neurons lit up. Whats more, sometimes the same group of neurons would light up in response to more than one picture.

At first, the scientists figured that it was a technological problemmaybe the sensors werent precise enough. For decades afterward, neuroscientists conducted variations on this experiment. Their equipment became more sensitive, but still they produced no meaningful results. What was going on? Surely, memories had to reside somewhere in the brain.

An MIT scientist by the name of Jerry Lettvin proposed a solution: the notion that a given memory lives within one cell was just plain wrong. As much as scientists wanted to find hierarchy in the brain, Lettvin argued, it just wasnt there. Lettvins theory was that rather than being housed in particular neurons that report to the hippocampus, memory is distributed across various parts of the brain. He coined the term grandmother cell to represent the mythical neuron that houses the memory of grandma. The picture Lettvin painted of the brain at first appears primitive and disorganized. Why would such a complex thinking machine evolve in such an odd way?