The Secrets and Science Behind Starling Murmurations

In the 1930s, famed ornithologist Edmund Selous suggested that birds moving in murmurations were using some sort of telepathy to transmit their flying intentions. "They must think collectively, all at the same time... a flash out of so many brains," he wrote in his book, "Thought-Transference (or What?) in Birds."

As the years wore on, we found out that's not quite it. In the 1950s, scientists studying insects and fish and other collective animal behavior posited that group movement is more of a stunningly fast response to others in the flock (or the school, or the swarm) rather than some innate mind-reading ability or a command from the group leader.

It's "the rapid transmission of local behavioral response to neighbors" that enables such startling synchronicity, as the authors of a 2015 paper published in the journal Proceedings of the National Academy of Sciences wrote.

"There's two ways that you can elicit large group behavior. You can have the top-down control, where you have some kind of leadership, or some kind of top-down mechanism. Think of a rock show, you have the rock star in the front and he starts clapping his hands, and the whole stadium starts clapping," Pesendorfer says. "But these murmurations are actually self-organized, meaning that it's the individual's little behavioral rules that make it scale up to the large group. In order to understand this behavior, we have to go from the local scale — what is the individual doing, what are the rules that the individual is following? — to the global scale; what is the outcome?"

In 2013, a mechanical and aerospace engineer and her team from Princeton collaborated with physicists in Italy to study murmurations. "In a flock with 1,200 birds, it is clear that not every bird will be able to keep track of the other 1,199 birds," Naomi Leonard, the Princeton engineer, said back then, "so an important question is 'Who is keeping track of whom?'"

The Italian physicists used more than 400 photos from several videos to find out, plotting the position and speed of birds as they flocked. From that, they built a mathematical model that identified the optimal number of flock-mates for each bird to track.

Turns out the magic number is seven: Each bird keeps tabs on its seven closest neighbors and ignores all else. Considering all these little groups of seven touch on other individuals and groups of seven, twists and turns quickly spread. And from that, a whole murmuration moves. The scientists' findings were published in the journal PLOS Computational Biology in January 2013.

Though it looks coordinated on a large scale, the individual birds are concerned with only three aspects of their flight and the flight of those around them. These factors have been described in several ways, but they're all very similar. They are, from Pesendorfer:

"Depending on how you change those three parameters," Pesendorfer says, "you can get everything from those barrel-looking baseballs that you get in ocean fish, to loose-looking insect swarms, to highly, highly organized fish swarms and murmurations. All in those three little parameters."

Scientists believe these birds flock in the first place to confuse and discourage predators, through their sheer numbers, with the noise such a flock makes and, of course, its motion. Some communication between birds may be happening, too, in murmurations — say, pointing out good food sources — while some researchers believe simply keeping warm may be another reason for the murmurations

What may be most stunning to mere mortals is that these birds react so quickly and do so in such synchronization; if not immediately, within a couple of flaps of a bird's wing. They move almost as one, in a type of lock-step (or, as it were, lock-flap).

How?

"Birds have a much higher temporal resolution than we do," says Pesendorfer, meaning that birds take in certain information around them and process it much more quickly than humans. "They see much faster than we do."

Back in 1986, Craig Reynolds, an MIT-trained computer scientist, built computer models of bird flocking and fish schooling in something he called "Boids." These programs provided the basis for lifelike animation in movies, initially (and notably) a swarm of bats in the 1992 Tim Burton film "Batman Returns."

In applications to real life, the ability to understand the behavioral movements of large groups of starlings (or bats or bees or whatever) and to program swarms of robots into making similar movements has amazing possibilities. "We are trying to draw inspiration from biology," George Young, who was the lead author on the paper produced from Leonard's group, told Princeton University back in 2013, "to understand what measures of animal group performance can help us decide what measures we should use when we design responsive behaviors for robots."

An example: Las Cumbres Observatory has 22 robotic telescopes on seven sites around the world that coordinate with each other to function as one big telescope. From the LCO site:

Another example: The emerging field of swarm robotics uses information gleaned from the study of starlings that could, according to the Wyss Institute at Harvard, "enable new approaches for search and rescue missions, construction efforts, environmental remediation, and medical applications."

Swarm robotics also could have use in military applications, like these micro-drones released from fighter aircraft. A swarm of self-driving cars, working together, could help reduce or eliminate traffic jams. The possibilities — cancer-fighting? — are mind-boggling.

All from watching, studying, learning and building on the wondrous flocking of this simple bird.

"As humans who have very complicated decision-making processes, we're not used to looking at simple decision-making processes that scale up to what looks like complex behavior," Pesendorfer says. "These models help us understand these types of patterns."