As for which strategy worked best, there was really no contest: copying was far and away the most successful approach. The winning entry exclusively copied others—it never innovated. By comparison, a player-bot whose strategy relied almost entirely on innovation finished ninety-fifth out of the one hundred contestants. The result came as a surprise to Laland and to one of his collaborators, Luke Rendell. “We were expecting someone to come up with a really clever way to say, ‘In these conditions you should copy, and in these conditions you should learn stuff for yourself,’ ” says Rendell. “But
...more
