[Paleopsych] Discovery: Robot Demonstrates Self Awareness
checker at panix.com
Fri Jan 6 03:44:09 UTC 2006
Robot Demonstrates Self Awareness
[Uh oh...too late for AI friendliness, time to buy How to Survive a Robot
By Tracy Staedter, Discovery News
Dec. 21, 2005- A new robot can recognize the difference between a mirror image
of itself and another robot that looks just like it.
This so-called mirror image cognition is based on artificial nerve cell groups
built into the robot's computer brain that give it the ability to recognize
itself and acknowledge others.
The ground-breaking technology could eventually lead to robots able to express
Under development by Junichi Takeno and a team of researchers at Meiji
University in Japan, the robot represents a big step toward developing
self-aware robots and in understanding and modeling human self-consciousness.
"In humans, consciousness is basically a state in which the behavior of the
self and another is understood," said Takeno.
Humans learn behavior during cognition and conversely learn to think while
behaving, said Takeno.
To mimic this dynamic, a robot needs a common area in its neural network that
is able to process information on both cognition and behavior.
Takeno and his colleagues built the robot with blue, red or green LEDs
connected to artificial neurons in the region that light up when different
information is being processed, based on the robot's behavior.
"The innovative part is the independent nodes in the hierarchical levels that
can be linked and activated," said Thomas Bock of the Technical University of
Munich in Germany.
For example, two red diodes illuminate when the robot is performing behavior it
considers its own, two green bulbs light up when the robot acknowledges
behavior being performed by the other.
One blue LED flashes when the robot is both recognizing behavior in another
robot and imitating it.
Imitation, said Takeno, is an act that requires both seeing a behavior in
another and instantly transferring it to oneself and is the best evidence of
In one experiment, a robot representing the "self" was paired with an identical
robot representing the "other."
When the self robot moved forward, stopped or backed up, the other robot did
the same. The pattern of neurons firing and the subsequent flashes of blue
light indicated that the self robot understood that the other robot was
imitating its behavior.
In another experiment, the researchers placed the self robot in front of a
In this case, the self robot and the reflection (something it could interpret
as another robot) moved forward and back at the same time. Although the blue
lights fired, they did so less frequently than in other experiments.
In fact, 70 percent of the time, the robot understood that the mirror image was
itself. Takeno's goal is to reach 100 percent in the coming year.
More information about the paleopsych