The very first robot "brains" were made of old alarm clocks

Elmer and Elsie, two electronic 'tortoises' built in the late 1940s by neurobiologist Grey Walter, were the first robots in history programmed to "think" the way biological brains do. What did they do with their freedom? What everyone tries to do: stay alive.


Walter was interested in brains. He studied electroencephalographs and collaborated with Pavlov's students and brain-body maps. He was also interested in processes that mimicked the brain. He leaned towards anarchism, because it had no top-down structure. He wanted to see how individuals and societies self-organized. And he built a couple of mechanical 'tortoises' that were meant to show exactly that in microcosm. Elmer and Elsie were cobbled together from war surplus materials and old alarm clocks. They had a single light or touch sensor hooked up to two different paths that engaged two different motors. Essentially, they were two-neuron brains.

The Els were the first robots built like brains and meant to have free will. They were allowed to wander around the floor with no fixed pattern. Walter watched them and saw what he could learn. One of the first things he learned was that Buridan's Ass, the dilemma that said an animal with no high brain function would be unable to choose between two equally-attractive points, was not correct. When the tortoises were presented with two light sources equidistant from the sensor, the sensor own pattern of scanning would decide the way to go. They'd head towards whichever light they saw as a consistent part of the scanning process.

Illustration for article titled The very first robot brains were made of old alarm clocks

The tortoises, especially Elsie, also had a few preferences when it came to people. It was noticed that they loved, above all things, women's legs. This was discovered to be because women wore nylons, and nylons reflected the light. Or, at least, they liked mild lights. Too-bright lights made them veer away. They would, like animals, be drawn to mirrors. The machines had a pilot light that was reflected, and would draw them close. Eventually Walter built Elsie and Elmer a 'hutch' where they could recharge their batteries, and in their wanderings they made their way back to it regularly enough that Walter speculated on their life versus the life of actual animals.

When given enough of a chance to stay alive, they did. When presented with certain stimuli, even outside of their programmed range of experience, they responded consistently, as if with a personality. They had their quirks and odd behaviors the way living beings do. And if they were pre-programmed in certain ways, well, weren't we programmed, biologically, far more than they were? On two neurons they had many of the behaviors and weirdnesses that biological entities seemed to have. Imagine the nuanced ways they would have behaved if they had had a full complement of human brain cells. Or maybe, looking at ourselves, we don't have to imagine.

Images: Eric Long and
Rutherford Journal

Via Forget Mori, Cerebromente, and Rutherford Journal.


Share This Story

Get our newsletter



animated gif if image doesn't show []

it's really quite amazing how far we've progressed...