In 1921, American naturalist William Beebe was exploring the rain forest in Guyana when he came across an astonishing sight: a vast column of army ants, millions of them, marching relentlessly around the jungle. It wasn’t their number that gobsmacked him, but where they were going: around and around in a huge circle. The circuit was so big—1200 feet across—that it took each ant two and a half hours to complete. And there seemed to be no escape. The ants marched on and one over the course of the next two days, Beebe reported, “with ever increasing numbers of dead bodies littering the route as exhaustion took its toll.”
In the years to come, other naturalists would report witnessing the phenomenon, which came to be called by a variety of names, including “circular mills,” “death circles,” and “ant mills.” (You can see videos here.)
Back in Beebe’s day, such behavior presented a baffling mystery. What, short of some kind of illness or collective madness or illness, could drive the ants to circle and circle until they died? Today, having figured out how ants navigate, we understand how ant mills form, and the phenomenon turns out to be far more interesting than anyone back then could have guessed.
We now know that as ants move around through the ground litter of the forest, they follow a trail of pheromone molecules laid down by those who have gone before. They also leave a trail of their own, to recruit other ants to follow. (If they didn’t, the first ant’s marker would gradually fade, and the trail would die out.) Such a system facilitates rather remarkably complex collective behavior, including sophisticated decision-making, by groups of individuals who are themselves all but brainless. But it can go wrong. When an ant trail by accident crosses itself, the ants following it can become stuck in an endless loop, laying down a stronger and stronger trail that sucks in any other nestmate who happens to come across it.
What does any of this have to do with human psychology? Well, the ant mill is a graphic illustration of a complex system that goes off the rails without any damage or trauma to any of its components. The ants aren’t sick; they aren’t insane. They’re doing exactly what millions of years of evolution have programmed them to do. The problem is that an ant colony at a system-wide level has an error mode which, once entered, cannot be escaped.
As Thomas D. Seeley describes in his wonderful book “Honeybee Democracy,” the complex societies of eusocial insects like ants and bees can be a useful model for the workings of the human brain. In both cases, an intercommunicating population of dumb agents (insects/neurons) spontaneously gives rise to an entity capable of complex behavior (the hive/the brain). And, as it turns out, the human brain has its own version of the ant mill. It can get caught in a loop, so to speak, without anything being wrong with it organically.
Anxiety is one example, as I can attest to from personal experience. A few years ago, after a particularly vigorous few hours of partying at a crowded shindig in Hollywood, I passed out cold and woke up to find myself staring up at a circle of strangers’ faces. I splashed some cold water on my face and went home. I assumed I was fine. But the next time I went to a social function, I realized that my heart was racing. My palms were sweating, I felt lightheaded, and a wave of nausea swept over me. I realized, to my horror, that something was happening to me that I had no control over—and the more I thought about it, the worse the feeling got. I was in the grips of an anxiety attack. I headed for the door.
Fear is, in a sense, like an ant trail. It’s an emotional response to perceived danger that guides our future behavior. When we experience something bad – in my case, the unpleasant feeling of passing out at a party – we lay down emotional memories that in future will prompt fear in the event that a similar prospect appears to be at hand. This helps us to avoid unpleasant consequences. But when we feel fear, we’re not just following a trail; we’re leaving one too. That is to say, fear is not just a response to aversive stimuli, but an aversive stimulus itself. And so, like an ant mill, the path of fear can double back on itself. In the grip of an anxiety attack, I wasn’t fearing the party anymore, but the physical sensation of my unease.
Like an ant in an ant mill, I was in a feedback loop that I could not escape. Yet I wasn’t sick. I wasn’t insane. My brain was perfectly healthy. All the same I felt myself slipping into a madness whose dimensions I could not fathom. It was like a perfectly good microphone feeding back into a perfectly good amplifier: everything was working as it should, but the result was horrible.
People who’ve never had an anxiety attack can be rather unsympathetic to those who suffer them. The problem is, truly, all in the head. So buck up! Get a grip on yourself! If only it were so easy. Populations of neurons cannot simply be ordered to change their tack, any more than a million marching ants can be told to knock it off.
That doesn’t mean that an anxiety sufferer’s plight is hopeless—simply that the biomechanics must be respected. In my case, my anxiety was relatively mild and its cause was easy to identify. To fall back on the ant-mill metaphor, I knew that what I had to let the pheromones of the endless loop dissipate and in their place gradually lay down new, healthier ones.
To do this, I learned to avoid big crowds in favor of small events where I always had a clear line of sight to the door. I positioned myself far enough into the group that I would feel a tingle of anxiety, but not so far that I risked letting the fear double back on itself. In time, I learned to trust my ability to handle my fear. (Some people find that anti-anxiety medicines such as benzodiazepines help the process along, though psychologists today tend to believe that that benefit does not outweigh the drugs’ high addictiveness—another manifestation, incidentally, of feedback loops gone awry.) With practice, I was able to gradually attend larger and larger social settings. After a few months, the fear went away entirely.
Ants stuck marching in a circle have no Xanax to pop or side door to slip out of. But they do, occasionally, manage to break away and find a path to freedom. After Beebe spent two days watching the ants go round and round, he wrote, “eventually a few workers straggled from the trail thus breaking the cycle, and the raid marched off into the forest.”