How is metacognition useful for orangutans?
Stromberg: Orangutans make lots of decisions every day, both in the wild and in human care. They decide where to move to, what to eat, whom to interact with and how closely to watch the world around them. These decisions require perceiving the environment and remembering things about the world around them. Sometimes, decisions require reflection—thinking about what they may be seeing or remembering.
For example, orangutans have to try to remember where food is located. But, decision-making can also be more complicated. Is an approaching individual aggressive or friendly, or even a combination of these? The last time that individual approached, did things go well or not? If enough information is available, they can behave confidently.
But, if their memory is weak, or if they are uncertain that they know what they need to know, they may instead try to adjust their behavior to reduce this uncertainty. For example, they may choose to avoid the social interaction until they are certain the other orangutan is a friend or foe. That is the function of metacognition.
How do you determine if an orangutan is confident?
Beran and Parrish: An orangutan’s confidence can be reflected in movement. The Zoo’s orangutans play memory computer games as part of their daily enrichment program. They tap a picture on the screen and have to remember it during a 1-, 2-, 4-, 7- or 10-second delay in which the screen goes blank. Then, they choose the original picture from a set of four options—similar to a multiple-choice test. The test may be easier if the delay is short, or harder if the delay is long.
Once they try to match the picture they saw, they have a few seconds before the computer sounds an auditory tone, telling them if they are correct or incorrect. If correct, they are rewarded with one of their favorite foods—grapes—but the treat is delivered somewhere else in the test room.
If they wait to learn from the computer that they correctly matched the image, they have to hustle to the reward area to get their treat or risk missing the treat altogether (it rolls outside of the testing area, out of the orangutan’s reach). However, if they answer and know they are correct, they can go over to the treat area early, take their time, and get their reward. If they are not sure if they are correct, they can wait for the “correct” tone and then hustle. If incorrect, a different tone will sound, indicating they do not get a grape and should remain in front of the screen and play the next matching game.
Thus, the best response to this task would be for the orangutan to move to the grape site as soon as they answer the question—but only if they think they are correct. This requires meta-memory. Their movements tell us their level of confidence in how they did on the matching game.
Lucy is in the training phase of the metacognition study. She does not yet have to move to receive the reward.