The #1 source for news about monkeys and apes

[Josh Greenman <josh.greenman AT gmail.com>]

http://news.yahoo.com/s/livescience/20090514/sc_livescience/monkeysponderwhatcouldhavebeen/print

Monkeys Ponder What Could Have Been

Rachael Rettner
LiveScience Staff

It's a good thing monkeys can't gamble. New research shows these
primates are capable of "woulda-coulda-shoulda" thoughts, like those
that keep gamblers at the tables.

Monkeys' brains respond to rewards that they observe but do not
experience - so-called fictive outcomes - and monkeys change their
behavior when they are shown the prizes that they could have had,
researchers from Duke University Medical Center say.

"Monkeys appear to be able to use fictive information to guide their
behavior, so they're not purely guided by their direct experience of
reward and punishment," said Michael Platt, the study's senior author.

This is one of the first studies to look at fictive thinking in
animals, he said. The results are detailed in the May 15 issue of the
journal Science.

Monkey thoughts

To study the monkey thought process, the scientists focused on the
anterior cingulated cortex (ACC), a region of the brain thought to be
involved in learning from experiences and adapting behavior. They
monitored the monkeys' brain neurons while the animals played a game
designed to study fictive thinking.

The monkeys were shown eight white cards arranged in a circle. Each
card had a color underneath that corresponded to a specific reward, in
this case, a certain amount of juice. One prize was larger than the
others - a high value reward. Over many trials, the monkey learned to
associate the color green with the biggest reward. After the monkey
picked a card, all the cards were turned over, and the monkey was
shown the prizes that it missed, and then given the reward.

The researchers found that the neural response was proportional to the
reward given - the bigger the reward, the more the neurons fired. And
the neurons responded the same way to fictive rewards. This result
shows that "nerve cells in the ACC, which we already knew carry
information about experienced rewards, also carry information about
fictive rewards, information about the rewards that got away," said
Platt.

Learning from mistakes

The monkeys also altered their behavior depending on the size of the
fictive reward.

If the monkey missed out on a big reward in one trial, it was more
likely to choose the card that had offered that large reward in the
next trial. This is similar to what people do when they gamble. For
example, if someone is playing roulette and bets on black, but red
wins and pays off big, the person is more likely to bet on red the
next time.

Fictive thinking may have helped the monkeys learn as well. During the
trials, the large reward was kept in the same spot 60 percent of the
time or moved clockwise by one position in order to see if the monkey
would pick up on the pattern. Indeed, they did. The monkeys chose
cards next to possible high value rewards in 38 percent of the trials,
while they chose cards adjacent to low value rewards in only 17
percent of trials.