Most of us are familiar with the unpleasant feeling of being kept in suspense about an uncertain outcome. In fact, psychological experiments have shown that people will typically avoid uncertainty by obtaining advance knowledge of the future if they can.
Now researchers have found evidence that this aversion to uncertainty is wired into our brains at a basic level. The same dopamine-based brain system that responds to rewards such as food and drink also pushes us to seek out “cognitive rewards” that give us advance information about future outcomes.
“Our data show that advance information is treated as a valuable goal in its own right and that both reward-seeking and information-seeking could be controlled by the same dopaminergic reward system,” says Ethan Bromberg-Martin, a postdoctoral researcher who conducted the study with principal investigator Okihide Hikosaka at the National Eye Institute, part of the National Institutes of Health in Maryland.
For the study, which appears in the July 16 issue of Neuron, two rhesus macaque monkeys were trained to recognize and choose between two symbols on a video screen. One symbol was followed by an image indicating the size of an upcoming reward of water. The other symbol led to information that wasn’t correlated at all to the size of the upcoming reward. Neither choice affected the actual size or the delivery of the actual reward, yet the monkeys quickly came to show a near absolute preference for the informative cue.
The researchers considered the possibility that the advance information was preferred because it conferred some subtle physical benefit to the monkeys, perhaps by cueing them to tense “their cheek muscles to swish water around in their mouths in a more pleasurable fashion.” So they set up a similar experiment in which an additional cue about the reward size was always given just before the actual reward. Thus the monkeys would always have advance knowledge of the reward size. Yet even in this case the monkeys strongly preferred to have the earlier information as well. Bromberg-Martin and Hikosaka concluded that “the monkeys treated information about rewards as if it was a reward in itself, preferring information to its absence and preferring to receive it as soon as possible.”
Information is power
Read Montague, a neuroscientist at the Baylor College of Medicine who has done prominent work on the brain’s reward system, finds the result impressive. In the experiment, he notes, each “animal is making conscious, willful choices to collect information early to diminish his uncertainty about the future delivery.”
While the monkeys made their choices and received their rewards, Bromberg-Martin and Hikosaka also used electrodes to monitor neurons in the midbrain. Electrical characteristics and responses to actual water rewards or their image cues verified these cells as reward-sensitive dopamine neurons. But these neurons also had the same “reward” response to the symbol that merely promised advance knowledge, in keeping with the monkeys’ strong preference for that symbol.
Why would the brain treat the informative symbol as rewarding, compared with the uninformative symbol, even when the information available via the two symbols has the same “expected value” in terms of water-reward prediction? Bromberg-Martin suggests that the brain has evolved this way simply because in real life such information would tend to have real value by enabling an animal to secure good outcomes and avoid bad ones. It could also allow animals to get quicker feedback for learning purposes. “You can never be sure which information will be useful in the future, so it may be best to have an internal drive to seek information for its own sake,” he says.
Hikosaka and Bromberg-Martin are now trying to understand how the brain converts such cognitive rewards, as well as other complex abstract rewards, into the “common currency” of the relatively primitive dopamine-based reward system.