For a hungry predator, choosing which prey to dine on can be tricky. Is it better to go after a nourishing but well-camouflaged target; or to focus on more obvious, but less tasty, morsels?

The same question faces Homo economicus, the reward-conscious human with a visual system wired up to fix its attention on to perceptually striking features of its environment.

At the Vision Sciences Society’s annual meeting in Naples, Florida, Caltech researcher Vidhya Navalpakkam unveiled research showing that humans are surprisingly quick at calculating the value of different targets and deciding how much attention to give them.

In a clever set of experiments, Navalpakkam recorded the eye movements of study participants sent on visual treasure hunts. The participants would earn points by fixing their gaze onto certain targets, such as horizontal or vertical bars, that were surrounded by worthless distractions – bars oriented at an angle, say.

In only half a second, participants had to move their eyes around a bar-filled screen to collect as many points as they could. Sometimes those valuable vertical and horizontal bars were worth the same number of points, but in other games one target was worth twice or four times as many points as the other.

In some games the distractor bars weren’t very distracting at all (it is easy to find a vertical or horizontal bar amid a clutter of 45-degree diagonal bars), but in others the distractor bars were nearly vertical or nearly horizontal, making it hard to find one of the treasure targets but easy to find the other.

Navalpakkam predicted an outcome of these games, based on three strategies her human participants might use: saliency information only, value information only, or both. She found that the third model made nearly perfect predictions of the outcome of the games, suggesting that even in a split second, humans are able to incorporate information about value to guide the way their eyes hunt for treasure.

Posted on behalf of Chaz Firestone