from the basic-motivations dept.
An article in Technology Research News ("Software agents evolve purpose", by Kimberly Patch, 2 January 2002) describes work by researchers from the Keldysh Institute of Applied Mathematics at the Russian Academy of Science have shown that purposeful behavior can emerge naturally in a software simulation that has simple software beings, or agents, evolving over many generations. The researchers described these evolved behaviors as purposeful motivation. The researchers say the simulation showed that a system that uses motivations to control simple reflexes can emerge in an evolutionary process. Having motivation was an advantage likely to be passed on to subsequent generations of the agents, said Mikhail Burtsev, one of the researchers. "The population of agents with motivations had obvious selective advantages compared with the population of agents without motivations," he said.
The researchers began with a small population of simple, identical neural-net based agents that could move, eat (gain energy from the environment), and mate with other agents to reproduce. The agent population as a whole had one goal — survival. This goal required individuals to push toward two basic subgoals — to replenish energy, and to reproduce, said Burtsev. The agents evolved to seek out [food] and other agents. "The most important thing here is that we didn't force agents to follow these needs. The needs were prescribed explicitly by [the] environment, and only agents that had these two needs could successfully undergo selection pressure," said Burtsev.
The article also contains comments from another artificial life researcher, who expressed some skepticism at the interpretation that the agents had evolved motivated behavior; rather, he said, it may simply be the result of the neural net having better access to information about the environment and acting on it more effectively.
The Russian researchersí technical paper ("A Life Model of Evolutionary Emergence of Purposeful Adaptive Behavior") is available online at the Lawrence National Laboratory archive, as an Adobe Acrobat PDF file.