Suzana Herculano-Houzel

Column_033 When playing loses its fun

Jonathan Schaeffer, a computer games expert at the University of Alberta in Canada, has just proved, after 18 years of mathematical calculations, that the game of checkers is doomed to a draw. From any of the 19 possible starting moves, no one wins if none of the opponents makes a mistake: winning at checkers is not a question of using a superior strategy, but of not making mistakes.

So it is with other childhood games, such as Tic-Tac-Toe, Mastermind, Rubik’s Cube and even the recent Sudoku: for all of them there is some infallible algorithm – a sequence of moves or steps – that solves the problem or at least guarantees a draw.

For some, the inevitable draw is enough for checkers and tic-tac-toe to be declared “dead”, solved. So why do children love these games, and why do they eventually lose their appeal? Games are extremely stimulating for the brain as long as they provide a challenge. A reasonable chance of success serves as motivation to tackle the problem, i.e. play, and occasional successes, by keeping the reward system interested and motivated to try again, increase the chances of learning the game and perhaps mastering it.

For some games, such as strategy games, complete mastery doesn’t happen: success is hardly ever 100% guaranteed. But when success becomes a certainty, the reward system loses interest and at most plays again from time to time, to see if it still remembers the algorithm for solving the problem.

Losing interest in the problem solved, by the way, is great for the brain, which abandons what has become trivial and moves on to more difficult problems that take advantage of the reasoning skills acquired in the easier game. And so we are promoted from checkers to chess, a game whose ending no computer has yet been able to predict since the opening.

It took Schaeffer’s computers 18 years to consider the paths between the 19 possible starting moves and the 39 trillion final positions for the game of checkers and conclude that they all lead to a draw, which “kills” the game. In perhaps the same amount of time, the human brain learns to evaluate the possible moves in order to avoid wrong moves and reach a draw – and then also concludes that the game has lost its fun. However, we know how to do something that computers don’t: complain that the game has lost its fun and ask to play something else.

Originally published in Folha de São Paulo on August, 2007

More Posts

pt_BRPortuguês do Brasil