02/13/2003
Hanging Out Together
In what has become our annual February ritual, my wife and I have fled New Jersey’s cold for the warmth of Marco Island in Florida. Followers of this column will know that for me a highlight of Marco is the early predawn walk on the beach. In the past, I’ve found topics for my column based on such things as the effects of red tide on the beach environment, black water off the coast of Florida, etc. So far this year the beach has been devoid of any striking features. Especially interesting is a virtually complete lack of any shells larger than an inch in size. This is most unusual but I haven’t found an explanation yet. Fortunately, there are still the birds.
The past couple of mornings, I have walked through gatherings of about a hundred birds, a mix of common seagulls and my favorites, birds with longer beaks and peaked heads. When I walk through this crowd of birds, virtually all the seagulls just walk a few feet out of my way. The other birds, however, will fly off together in a swarm, typically circling in formation back to land again in the same area after I’ve passed through. It’s as though one member of the group gives a signal and they all follow that bird’s lead. Actually, this “swarming” behavior is a subject of great interest and generally the swarming group appears to behave in a more intelligent manner than is the case.
Swarming is the theme of a rather scary book I received for Christmas and just finished reading here on Marco. The book is “Prey” by Michael Crichton, author of “Jurassic Park” and creator of the TV show ER. “Prey” examines what might happen if nano-, bio- and computer technologies were combined to create a swarm of nanoparticles. Crichton’s miniscule nanoparticles aren’t simply bits of dust but are imbued with sensors and other capabilities. The swarms turn out to reproduce and quickly evolve into a menace that threatens mankind. I won’t go into details so as not to spoil the plot for potential readers of the book.
Normally, I would dismiss the possibility of such a scenario as just a clever sci-fi story. However, a marriage of computer and nanotechnologies with a biological component gives me pause. We’ve already seen HIV, drug-resistant TB and other diseases evolve with great speed in a matter of years. Adding swarming behavior could indeed be frightening. Spurred by the Crichton novel, I went online to find out more about swarms and was quite impressed with the amount of work on the subject. These studies are closely allied with studies on artificial intelligence. My surfing of the Web turned up a truly startling proposal attributing our human self-awareness and our other human attributes to swarming characteristics and behavior.
But first let’s consider the nature of a swarm. A typical swarm consists of a number of mobile species, insects being a typical example. In an earlier column, we talked about how ants communicate the location of a food source, leading to a directed swarming of the ants to harvest the food. Communication among the members of a swarm is another feature of a swarm. At times, the behavior of a swarm is such that one is tempted to attribute a considerable degree of intelligence to the insects composing the swarm. Yet, the brain capacity of the individual insect is miniscule.
Scientists using computer models and simulations have shown that complex behavior can result in swarms in which the individual members are programmed to follow only one or less than a handful of simple rules. This following of simple rules by the individual members is another characteristic of a swarm. By obeying these simple rules, the swarm becomes a problem- solving group that, without any superior intelligence, can solve surprisingly complex problems. No single member of the group has anywhere near the intellect to solve the problem.
Suppose you wanted to construct a skyscraper. How would you solve the problem of how to build it? I may have discussed the African termite earlier but, if so, it’s worth repeating here. The African termite has just this problem - constructing mounds several feet high that tower over the termite. Some mounds may even be connected by arch-like structures. Sounds like you need a pretty smart termite. However, back in 1959, a French biologist showed that the achievement was the result of obeying a few simple rules. First, hundreds of termites mill around in a random fashion, depositing regurgitated pellets of masticated earth on spots that are slightly elevated from the ground. With hundreds of termites, just by chance, little mounds of these pellets are formed. So, Rule 1 - chew up dirt and spit it out!
Once these mounds are formed, Rule 2 kicks in – if you’re a termite and come across a mound, start working hard to build up pellets of dirt on top of it. With a bunch of hard-working termites carrying out Rule 2, you soon have a pretty tall column. Depending on what kind of termite you are, when the column gets to a certain height, it’s time to quit, at least if there’s no other column nearby. However, if you do spot a column that’s close enough, Rule 3 kicks in – start piling up dirt on the slopes of the adjacent columns. Keep obeying Rule 3 until you have an inverted arch connecting the columns.
The termite swarm solves a major construction problem with only a few simple rules that have evolved over the course of time and been programmed into the individual termite. With today’s powerful computers, scientists can create swarms of virtual objects. By programming simple rules (e.g., if the next square is occupied, move away two squares), swarms of these virtual entities can be set loose to evolve on their own to achieve an objective. The programmer may have no idea what the final solution will look like. The swarm is set free to optimize itself in its own way. This is where someone like Michael Crichton can come in and frighten us to death by suggesting that harmful results can result if a swarm of real, not virtual objects is released to optimize itself and ends up going down the wrong path.
The dire result becomes even more frightening if the swarm has the capability to reproduce. We already have virtual swarms of viruses that reproduce. I got an e-ticket for our flight to Florida on Continental. When I changed the flight several days before departure, the Continental agent said she would send an e-mail confirmation that would arrive in 15 minutes. I never received the e-mail. The reason was a successful viral attack on the Continental computer system. Crichton may not be too far off the track!
How do all the millions or billions of cells that make up our bodies manage to cooperate to allow us to function? And how do we manage to think and act without some kind of chief honcho in our brain to keep things running smoothly? Could the answer be swarming? A fellow named Jesper Hoffmeyer thinks so. Hoffmeyer proposes that in our body the cells are the units that make up swarms. These swarms of cells of the same type are in turn members of higher order swarms. Our amazing capabilities of mind and physical function are the result of swarm interactions, in Hoffmeyer’s view.
Hoffmeyer emphasizes that these swarms of cells differ from insect swarms. The cells aren’t mobile but are stuck together. In a swarm the members communicate. The communication among the cells and among the swarms is not simple. Each cell has a multitude of receptors and the response of a given cell depends strongly on its surrounding cells and the particular chemical or electrical stimulus or signal. The rules are vastly more complex than with the termites, to say the least.
Hoffmeyer doesn’t presume to understand the complexities of the swarm interactions he proposes. However, he stresses that the swarms of immune system cells and the swarms of nerve center cells interact closely. A consequence of his thesis is that the brain, with its swarms, is connected closely with the rest of the body. With the swarms and swarms of swarms all doing their individual problem-solving tasks, the individual not-too-bright cells end up in combination with their colleagues and swarms of other cells to make us the brilliant creatures we are. At least that’s my impression of Hoffmeyer’s proposal.
Hoffmeyer concludes a 1995 paper with the statement that “thoughts and feelings are not localized entities. They swarm out of our body collective.” If the swarm idea is true it seems to me this is indeed a profound suggestion that would solve, conceptually at least, a lot of problems relating to our human nature and capabilities. I have no idea if Hoffmeyer’s ideas are on target but to my simple mind they have a distinct appeal, much like black holes. That is, I can convince myself that I understand the broad concepts, but my simple mind can only handle so much! I’ll try to follow up on swarms and see if I can find critiques of Hoffmeyer’s ideas. Meanwhile, if you’re up to a somewhat disturbing read, try “Prey” and you’ll get a better and much more exciting feeling for the nature of swarms.
Allen F. Bortrum
|