Fly Guy

“It seems ridiculously simple, but it’s an illustration of what you can do when you develop a new method for measuring something,” says Dickinson. “The instruments we make in the laboratory now are much more sophisticated. But it’s still the same challenge of measuring the unmeasurable—the thrill of engineering.”

These days Dickinson applies his engineering acumen to understanding how insects integrate myriad sensory inputs to coordinate the motor systems they use to navigate through space—in other words, how flies fly. Here, he talks about tethering flies, telling stories, and traveling with a ukulele.

DICKINSON TAKES OFF

A fortunate failure. After receiving a doctorate in zoology from the University of Washington in 1989, Dickinson did a brief stint as a postdoctoral fellow at the Roche Institute of Molecular Biology. “I was sort of testing the waters of going into more mainstream neuroscience, and for a number of reasons, it didn’t really work out. But in retrospect it was one of those cases where a failure ends up being a good thing.” Dickinson had already lined up a job at the University of Chicago, but instead of remaining at Roche to the bitter end, he headed to the Max Planck Institute for Biological Cybernetics in Tübingen, where he spent five and a half months working with a behavioral physiologist named Karl Götz. “That was probably the best decision I ever made in my career.” With no training in either physics or engineering, Dickinson somehow convinced Götz to collaborate on a series of experiments aimed at figuring out how flapping wings can produce aerodynamic forces. “Fluid mechanics is not a field you enter on whim. But it turned out to be a remarkably productive time. That’s when I decided to focus my energies on the physiological basis of flight. I probably would not have been nearly as successful as a scientist if my postdoc had worked out.”

“I think we’re still just scratching the surface of all the little modules of behavior that are necessary to make flight work.”

If you build it . . . “Karl was a real master at building these incredibly elaborate instruments for measuring and quantifying behavior in Drosophila. All of his equipment was custom-designed—you couldn’t get this stuff from the Fisher catalog. Karl was from this sort of central European, mostly German, tradition, where he believed, ‘You think of the experiment you want to do and then you build the instrument to do it.’ To this day, we do a lot of engineering of instruments that allow us to do experiments that we couldn’t do otherwise and to try to turn questions that are qualitative—like ‘Is the fly flying?’—into something more rigorous and quantitative.”

Winging it. In Tübingen, Dickinson and Götz tackled a fundamental puzzle. “At the time there was this paradox. If you started with mathematical models where you could say, ‘An insect is flapping its wings this fast, and it’s this big and this heavy, so the wings need to generate this much force to keep the animal in the air’—and then you took that wing and did standard aerodynamic analyses, you’d conclude that there was no way the insect could fly. Well, there had to be a solution to this paradox, because obviously flies can fly. We found that at least part of the solution is that insect wings are not acting like airplane wings. They flap them back and forth and there’s something about the process of flapping that enables the wings to make larger forces than you would normally predict. Karl and I made a simple model of the wing, one that flapped back and forth, and we were able to show that it created a leading-edge vortex that sits on top of the wing” to help generate lift.

Flight control. These days, Dickinson and his team spend more time studying live specimens as they soar and steer, climb and dive, take off and land. Some studies are carried out in an arena called the Fly-O-Rama, in which the insects’ flight paths are recorded on video. In other experiments, the flies are tethered to rods “and we fool them into flying while they’re stuck in place. We measure the motion of their wings with little electronic sensors and we can tell whether the fly is trying to steer left or steer right or fly forward. That information changes the electronic display around the insect—so it’s like a 360-degree virtual-reality flight simulator. There are zillions of experiments you can do with these instruments, and a couple of companies have even started to make them. You can actually go to a website and click ‘add to cart.’ So we’ve done a lot to develop technologies that the rest of the field can use.”

From the top. When Dickinson started out, he says, “there were debates raging about what the job of a neuron was and how it encodes information. Was there information in how fast it was firing or in the timing of each individual spike? But it seemed to me that a lot of that debate was uninformative, because the only way to know the role of a neuron is to ask the next cell down the line. So we can analyze all the sensory information coming into the nervous system, but what we really need to find out is what the motor neurons and muscles are doing with that information.” One of the key muscles involved in steering, for example, works like a variable spring whose stiffness is controlled by the arrival of nerve impulses sent by the brain. “So now we know that all the information coming from the eyes and the antennae and other sensors during flight has to get transformed into this ‘phase code’ that controls the stiffness of this spring. It gives this better idea of what is important to the system.”

Clever pest. “Everybody’s tried to swat a fly. And most people think of the fly’s escape reflexes as being just that—a simple reflex. But using high-speed video, one of my graduate students was able to show that in the fraction of a second prior to takeoff, the fly actually does this elegant little dance in which it positions its legs so that it will push away from an approaching swatter. So there’s this whole level of motor planning that’s going on that we never would have realized without these detailed quantitative observations. And I think we’re still just scratching the surface of all the little modules of behavior that are necessary to make flight work. There’s still a lot of mining to be done.”

DICKINSON DELIBERATES

Now, that’s epigenetics. “For a lot of scientists, Drosophila are just these little gene holders that live in vials inside incubators. But they’re real animals. They can fly 13 kilometers in a desert, maintaining a compass direction. I hope that by illustrating the richness of the animal’s behavioral repertoire, we can help hone the questions regarding what the brain is capable of doing.”

Before MacArthur. In 2001, Dickinson was selected as a MacArthur Fellow. “It was a real honor, and I’m still a little embarrassed about it, quite frankly.” But even more critical was a fellowship he received from the Packard Foundation in 1992. “That was probably the most influential grant I ever got because it was right at the start of my career. That money allowed me to do exactly what I wanted to do—not what I would have needed to do to get an NIH grant. It really allowed the lab, at its very initial stage, to be curiosity- and hypothesis-driven. I’m incredibly thankful to the Packard Foundation for that.”

This, that, and The Other. “I’ve always gravitated towards invertebrates. I have this fascination with what I call ‘the other.’ A lot of neuroscience is narcissistic in the sense that we study animals so we can learn something about ourselves. But for me, what’s more fascinating are the differences. And leeches are really, really different. Sure, there are some fundamental mechanisms that are common. But it’s hard to picture them as little tiny humans. And I think science needs that. We learn a lot when we see how different creatures come up with different solutions to the same problems.”

Who you callin’ simple? “Flies and leeches are often described as simple systems. To me that’s backwards thinking. It’s one thing to describe them as ‘tractable’ systems because they’re small and amenable to experimentation. But in a fundamental way, they may actually be more complicated than we are. Think of the number of neurons in a leech or a fly—and the ratio of how many neurons they have to how many behaviors they generate. Flies are performing a ridiculous number of behaviors, some that are very, very complex, with a relatively small number of neurons. So they’re getting more out of every neuron than I think a mouse does or a human does. To me, there’s nothing simple about that.”

What’s the story? “I believe in telling stories. Not in the sense of making stuff up. But I really believe in narrative. You can’t just go listing the load of experiments you did, even if they’re nicely and accurately described. You really need to tie things together into a narrative that people will remember and understand. It does take effort. Yes, the 20 control experiments you did are important. But if you just talk about them in a didactic way, I don’t think you teach anybody anything.”

Pay no attention to that man behind the curtain. “When I was a graduate student, the Hox genes had just been cloned. So there was a lot of pressure to ‘go molecular.’ I think you definitely have to be very wary of that sort of thing. If you’re getting advice to do something, it probably means there are a lot of other people getting exactly the same advice. So I say: don’t pay too much attention to anybody’s advice. You’re better off finding your own way and working on something that really interests you.”

All work and no play. Dickinson wishes he had more time to tinker. “My happiest times were when I could be the guy who tries out crazy stuff—setting up preliminary experiments that are just too risky for a graduate student or a postdoc. I harbor fantasies about trying to get back to that place. Now my role is to attract clever people and to create an environment where I can nurture their advancement. Which is certainly important, but it sort of saddens me a little bit that I don’t get to play as much as I used to and as much as I’d like to.”

DISTINCTLY DICKINSON

In the public eye. “My father was, and still is, an Episcopalian minister. So I was a priest’s kid. When you grow up in a priest’s or rabbi’s family, you’re kind of always on display in the community. So I think that very early on you adopt or evolve this ability to perform. I almost never get nervous about giving talks, I think because I’m kind of used to being ‘on.’ At least that’s the story I tell myself.”

On the line. Dickinson paid his way through college working as a sauté cook. “When you’re turning out 200 or 300 meals in a few hours, it’s all about timing. As the table orders come in, you almost instantaneously perform all this mental calculus of how long each dish takes. I think it’s perhaps an often overused term, but the notion of ‘zone’ is very apt. I don’t think I’ve ever done anything else in my life where time just warps in that way. The orders start coming in at 7:30 and you just get into this state where it seems like you blink and it’s already 11 o’clock.” And when the last meal is served, the day is done—which is much different from the way things work in the lab. “Science is never finished,” says Dickinson. “There’s always something to worry about at the end of the day.”

Strummin’ the four-string. Dickinson plays jazz—on the ukulele. “I know it sounds crazy, but the voicings of a ukulele, with its four strings and small neck, make it a fun and easy instrument for playing jazz chords.” And its size makes it perfect for travel. “It fits in the overhead bin, so it’s a convenient instrument because you can carry it wherever you go. If I know I’m going to have a lot of downtime in a hotel room, I usually take my uke along. Definitely beats watching bad TV.”