One of the most uncanny sights in nature is that of a school of fish, a seething, silvery mass of life thousands strong, darting and diving in perfect coordination. They skim the surface of a reef and take a hard right turn as one, only to divide and coalesce again when a predator plunges into their midst. To a human, it looks as if each fish has a tiny walkie-talkie and they are planning their next move en masse.
But what looks smart on the school-wide scale may actually be rather simple on the single-fish scale. Over the years, scientists who study collective behavior have found that computer-simulated fish programmed to obey just a few simple rules, like staying a specific distance from the nearest neighbor fish, show exactly the same kind of choreographed fluidity as living schools do. The beautiful coordination is what scientists call emergent: it’s a complex-looking behavior that arises naturally out of many individuals each doing something rather simple.
Now, in a new paper in Science, a team of researchers reports that another form of complex group behavior—keeping to the shadows in a pool with dappled light—can be achieved without the fish doing anything much more complicated than moving faster when the light falls on them and slower when they’re in the dark. What’s more, the greater the number of fish in the school, the better they are at staying in the dark, which the scientists suggest could mean that a large group has survival advantages over a small one. It’s the kind of observation that might not only contribute to our understanding of how swarms, schools, and flocks interact with their world, but also suggest new parameters for conservation.
To investigate the dappled-pool phenomenon, Iain Couzin, a professor of evolutionary biology at Princeton University and the lead author of the paper, oversaw a study that involved a dark room outfitted with a broad, shallow fish tank and a ceiling-mounted projector. The projector shone light down onto the tank in constantly morphing patterns, and a grad student, Andrew Berdahl, released ever-increasing numbers golden shiners, 3- to 5-in. (7.5 to 12.5 cm) bait fish common in North America, into the water. Berdahl began with 1 and worked slowly up to 256, filming their fluid movements with an infrared camera.
Going through their data with computer software, Berdahl, Couzin, and their collaborators saw that each fish’s tendency to stay within a fixed distance of its neighbors—an element critical to schooling—could explain most of the school’s coordinated movement away from light. “The direction of acceleration was almost always correlated with the direction of their nearest neighbors. They were acting mostly out of sociality. That was the dominant predictor of the direction of their acceleration,” says Berdahl. The other factor the scientists thought might be involved—that once a fish found itself in the light, it would look around for the nearest dark patch and then speed there with all haste, in essence sharing information about the light’s location with its schoolmates—didn’t seem to play a role.
But being in the light did change one thing about the fish: their speed. Though the fish in the light didn’t seem purposely to orient themselves back toward the dark, they moved faster than the shadow fish. Conducting a thought experiment, the team concluded that this would be enough to explain the school’s ability to swing away from the light. Berdahl uses the metaphor of a Segway in which the left wheel is turning slowly (the fish in the dark) while the right wheel is turning quickly (the fish in the light). Because of the difference in speeds, the whole vehicle turns to the left—or in the case of the school, into the darkness. The actions of the individuals could thus determine the behavior of the whole. Slow-moving fish, like slow-moving cars, also tend to bunch more tightly together. So not only do they spend more time in the dark, they also huddle closer there.
Intriguingly, the team also noticed that the larger the group, the better the individuals did at staying in darkness—sheer population size, in other words, improved performance. Exactly why this happens isn’t clear, but if it turns out that larger groups have an advantage in finding resources or safety in their environment, the finding could influence the human approach to conserving schooling or flocking species, Couzin says.
“By grouping together, animals may be much more capable of finding appropriate habitat within complex environments,” he says. “If the capacity of groups to search these environments and to find habitat is really an emergent property at the level of the group, then we have to be really extremely cautious about changes that will cause groups to fragment, like overfishing or population decline in migrating animals.”
Indeed, this sort of work is just the first step of many for researchers interested how animals use collective sensing in the wild. “For those of us working on collective behavior, figuring out the algorithm that these animals are using in one situation is a great first step,” says Deborah Gordon, a professor of biology at Stanford University who studies ants. “But after that we want to know how do they use it in the real world? What’s it for? How does it change in different conditions?” In animals, as in humans, there is clearly wisdom in crowds. Just how wise they get and just how big the crowd needs to be is a much more complicated tale.