Why Animals Don’t Get Lost

By Kathryn Schulz, a staff writer at The New Yorker, won the 2016 Pulitzer Prize for feature writing.

More generally, the astonishment is that any physiology can contain a navigational system capable of such journeys. A bird that migrates over long distances must maintain its trajectory by day and by night, in every kind of weather, often with no landmarks in sight. If its travels take more than a few days, it must compensate for the fact that virtually everything it could use to stay oriented will change, from the elevation of the sun to the length of the day and the constellations overhead at night. Most bewildering of all, it must know where it is going—even the first time, when it has never been there before—and it must know where that destination lies compared with its current position. Other species making other journeys face additional difficulties: how to navigate entirely underground, or how to navigate beneath the waters of a vast and seemingly undifferentiated ocean.

How might an animal accomplish such things? The Goulds, in “Nature’s Compass,” outline several common strategies for staying on course. These include taxis (instinctively moving directly toward or directly away from a given cue, such as light, in the case of phototaxis, or sound, in the case of phonotaxis); piloting (heading toward landmarks); compass orientation (maintaining a constant bearing in one direction); vector navigation (stringing together a sequence of compass orientations—say, heading south and then south-southwest and then due west, each for a specified distance); and dead reckoning (calculating a location based on bearing, speed, and how much time has elapsed since leaving a prior location). Each of these strategies requires one or more biological mechanisms, which is where the science of animal navigation gets interesting—because, to have a sense of direction, a given species might also need to have, among other faculties, something like a compass, something like a map, a decent memory, the ability to keep track of time, and an information-rich awareness of its environment.

The easiest of these mechanisms to understand are those that most closely resemble our own. Most humans, for instance, routinely navigate based on a combination of vision and memory, and we are not alone. One scientist, puzzled to find that his well-trained rats no longer knew their way around a maze after he moved it across his lab, eventually determined that they had been navigating via landmarks on the ceiling. (That was a blow to the notion, much beloved by behaviorists, that such rats were just learning motor sequences: ten steps forward, turn right, three steps forward, there’s the food.) Other animals use senses that we possess but aren’t very adept at deploying. Some rely on smell; those migrating salmon can detect a single drop of water from their natal stream in two hundred and fifty gallons of seawater. Others use sound—not in the simple, toward-or-away mode of phonotaxis but as something like an auditory landmark, useful for maintaining any bearing. Thus, a bird in flight might focus on a chorus of frogs in a pond far below in order to orient itself and correct for drift.

Many animals, however, navigate using senses alien to us. Pigeons, whales, and giraffes, among others, can detect infrasound—low-frequency sound waves that travel hundreds of miles in air and even farther in water. Eels and sharks can sense electric fields and find their way around underwater via electric signatures. And many animals, from mayflies and mantis shrimp to lizards and bats, can perceive the polarization of light, a helpful navigation cue that, among other things, can be used to determine the position of the sun on overcast days.

Other navigational tools are simultaneously more prosaic and more astounding. If you trap Cataglyphis ants at a food source, build little stilts for some of them, give others partial amputations, and set them all loose again, they will each head back to their nest—but the longer-legged ones will overshoot it, while the stubby-legged ones will fall short. That’s because they navigate by counting their steps, as if their pin-size brains contained a tiny Fitbit. (On the next journey, they’ll all get it right, because they recalibrate each time.) Similarly, honeybees adjust their airspeed in response to headwinds and tailwinds in order to maintain a constant ground speed of fifteen miles per hour—which means, the Goulds suggest, that by tracking their wing beats the bees can determine how far they have travelled.

I have presented these navigation mechanisms serially, but most creatures possess more than one of them, because different conditions call for different tools. What works at noon might not work at night, what works close to home might not work far away, and what works on a sunny day might not work in a storm. Yet even all these tools in combination cannot account for the last of the way-finding strategies described by the Goulds, which is by far the most arresting and confounding: true navigation.

True navigation is the ability to reach a distant destination without the aid of landmarks. If you were kidnapped, taken in pitch darkness thousands of miles away, and abandoned somewhere uninhabited, true navigation would be your only option for finding your way home.

To do so, you would need a compass, along with the know-how to use it—for instance, an awareness that magnetic north and geographic north are not identical. Failing that, you would need to be able to orient based on the movement of the sun—a tricky business, especially if your kidnappers weren’t kind enough to inform you of your latitude. If you plan to travel after dark, you’d better hope that you aren’t in the Southern Hemisphere, which has no equivalent of the North Star, or you’d better be able to rival Galileo with your knowledge of the nightly and seasonal course of the constellations. But, even if all this applied, you would still be in trouble if you did not also have a map. Being able to maintain a given bearing with perfect precision isn’t much help if you have no idea where you are vis-à-vis your destination.

Some animals plainly do have such a map, or, as scientists call it, a “map sense”—an awareness, mysterious in origin, of where they are compared with where they’re going. For some of those animals, certain geographic coördinates are simply part of their evolutionary inheritance. Sand hoppers, those tiny, excitable crustaceans that leap out of the way when you stroll along a beach, are born knowing how to find the ocean. When threatened, those from the Atlantic coast of Spain flee west, while those from its Mediterranean coast flee south—even if their mothers were previously translocated and they hatched somewhere else entirely. Likewise, all those birds that embark on their first migrations alone must somehow know instinctively where they are going.

But instinct alone does not explain what such birds can do. In 2006, scientists in Washington State trapped a group of white-crowned sparrows that had begun their annual migration from Canada to Mexico and transported them in a windowless compartment to New Jersey—the avian equivalent of the kidnapping thought experiment. Upon release, the juvenile birds—those making their first trip—headed south along the same bearing that they had been using back in Washington. But the adult birds flew west-southwest, correcting for a displacement that nothing in their evolutionary history could have anticipated. That finding is consistent with many others showing that birds become better navigators during their first long flight, in many cases learning entirely new and more efficient strategies. Subsequent experiments found that mature birds can be taken at least six thousand miles from their normal trajectory and still accurately reorient to their destination.

How do they do it? At present, the most compelling theory is that they make use of the earth’s magnetic field. We know about this ability because it is easy to interfere with it: if you release homing pigeons on top of an iron mine, they will be terribly disoriented until they fly clear of it. When scientists went looking for an explanation for this and similar findings, they found small deposits of magnetite, the most magnetic of earth’s naturally occurring minerals, in the beaks of many birds, as well as in dolphins, turtles, bacteria, and other creatures. This was a thrilling discovery, quickly popularized as the notion that some animals have built-in compass needles.

As with many thrilling and popular scientific ideas, however, this one started to look a little strange on closer inquiry. For one thing, it turned out that birds with magnetite in their beaks weren’t navigating based on north-south alignment, as we humans do when using a compass. Instead, they were relying on the inclination of the earth’s magnetic field—the changing angle at which it intersects the planet’s surface as you move from the poles to the equator. But inclination provides no clues about polarity; if you could sense it, you would know where you were relative to the nearest pole, but you wouldn’t know which pole was nearest. Whatever the magnetite in birds is doing, then, it does not seem to function like the needle in a compass. Even more curiously, experiments showed that birds with magnetite grew temporarily disoriented when exposed to red light, even though light has no known effect on the workings of magnets.

One possible explanation for this strange phenomenon lies in a protein called cryptochrome, which is found in the retina of certain animals. Some scientists theorize that, when a molecule of cryptochrome is struck by a photon of light (as from the sun or stars), an electron within it is jolted out of position, generating what is known as a radical pair: two parts of the same molecule, one containing the electron that moved and the other containing an electron left unpaired by the shift. The subsequent spin state of those two electrons depends on the orientation of the molecule relative to the earth’s magnetic field. For the animal, the theory goes, a series of such reactions somehow translates into a constant awareness of how that field is shifting around it.

If you did not quite grasp all that, take heart: even researchers who study the relationship between cryptochrome and navigation do not yet know exactly how it works—and some of their colleagues question whether it works at all. We do know, though, that the earth’s magnetic field is almost certainly crucial to the navigational aptitude of countless species—so crucial that evolution may well have produced many different mechanisms for sensing the field’s polarity, intensity, and inclination. Taken together, those mechanisms would constitute the beginnings of a solution to the problem of true navigation. And it would be an elegant one, capable of explaining the phenomenon across a range of creatures and conditions, because the magnetic field is omnipresent on this planet. Given some means of detecting it, you could rely on it by day and by night, in clear weather and in foul, in the air and over land and underground and underwater.

That kind of sweeping explanation would be convenient, because true navigation, which was once thought to require the kind of advanced reasoning and sophisticated toolmaking exclusive to humans, seems increasingly likely to be a widely shared capacity. Countless bird species can do it, as can salmon. Those conga-line rock lobsters are so good at it that they appear to be impossible to disorient, which we know because scientists have gone to outlandish lengths to try to do so. As Barrie describes in “Supernavigators,” you can cover a rock lobster’s eyes, put it in an opaque container filled with seawater from its native environment, line the container with magnets suspended from strings so they swing in all directions, put the container in a truck, drive the truck in circles on the way to a boat, steer the boat in circles on the way to a distant location, drop the lobster back in the water, and—voilà—it will strike off confidently in the direction of home.

Needless to say, you and I cannot do this. If you blindfold human subjects, take them on a disorienting bus ride, let them off in a field, remove the blindfolds, and ask them to head back toward where they started, they will promptly wander off in all directions. If you forgo the bus and the blindfolds, ask them to walk across a field toward a target, and then conceal the target after they start moving, they will stray off course in approximately eight seconds.

The problem isn’t that humans don’t have any innate way-finding tools. We, too, can steer by landmark, and we can locate the source of sounds or other environmental cues and make our way toward them. (With sounds, we do this much like frogs: by unconsciously assessing either the intensity differential or the time delay between a noise in our right ear and in our left one.) We also have a host of specialized neurons to help keep us oriented: head-direction cells, which fire when we face a certain way (relative to a given landscape, not to cardinal directions); place cells, which fire when we are in a familiar location; grid cells, which fire at regular intervals when we navigate through open areas, helping us update our own position; and boundary cells, which fire in response to an edge or obstacle in our field of vision.