A book by William H. Calvin UNIVERSITY OF WASHINGTON SEATTLE, WASHINGTON 98195-1800 USA |
HOW BRAINS THINK
A Science Masters book, to be available in 12 translations (BasicBooks in the US) copyright ©1996 by William H. Calvin |
What to Do Next
Piaget used to say that intelligence is what you use when you dont know what to do (an apt description of my present predicament as I attempt to write about intelligence). If youre good at finding the one right answer to lifes multiple-choice questions, youre smart. But theres more to being intelligent a creative aspect, whereby you invent something new on the fly. Indeed, various answers occur to your brain, some better than others. Every time we contemplate the leftovers in the refrigerator, trying to figure out what else needs to be fetched from the grocery store before fixing dinner, were exercising an aspect of intelligence not seen in even the smartest ape. The best chefs surprise us with interesting combinations of ingredients, things we would ordinarily never think went together. Poets are particularly good at arranging words in ways that overwhelm us with intense meaning. Yet were all constructing brand-new utterances hundreds of times every day, recombining words and gestures to get across a novel message. Whenever you set out to speak a sentence that youve never spoken before, you have the same creativity problem as the chefs and poets furthermore, you do all your trial-and-error inside your brain, in the last second before speaking aloud. Weve lately made a lot of progress in locating some aspects of semantics in the brain. Frequently we find verbs in the frontal lobe. Proper names, for some reason, seem to prefer the temporal lobe (its front end; color and tool concepts tend to be found toward the rear of the left temporal lobe). But intelligence is a process, not a place. Its a way, involving many brain regions, by which we grope for new meanings, often consciously. The more experienced writers about intelligence, such as IQ researchers, steer clear of the C word. Many of my fellow neuroscientists avoid consciousness as well (some physicists, alas, have been all too happy to fill the vacuum with beginners mistakes). Some clinicians unintentionally trivialize consciousness by redefining it as mere arousability (though to talk of the brain stem as the seat of consciousness is to thereby confuse the light switch with the light!). Or we redefine consciousness as mere awareness, or the searchlight of selective attention. Theyre all useful lines of inquiry but they leave out that activism of your mental life by which you create and edit and re-create yourself. Your intelligent mental life is a fluctuating view of your inner and outer worlds. Its partly under your control, partly hidden from your introspection, even capricious (every night, during your four or five episodes of dreaming sleep, it is almost totally out of control). This book tries to fathom how this inner life evolves from one second to the next, as you steer yourself from one topic to another, as you create and reject alternatives. It draws from studies of intelligence by psychologists, but even more from ethology, evolutionary biology, linguistics, and the neurosciences. |
There used to be some good reasons for avoiding a comprehensive discussion of
consciousness and the intellect. A good tactic in science, especially when mechanistic-level
explanations dont help structure your approach to a fuzzy subject, is to fragment the problem
into bite-sized pieces and that is, in some sense, whats been going on. A second reason was to avoid trouble by camouflaging the real issues to all but insiders (maintaining deniability, in the modern idiom). Whenever I see words that have everyday meanings but also far more specific connotations used only by insider groups, I am reminded of code names. Several centuries ago, an uncamouflaged mechanistic analogy to mind could get you into big trouble, even in relatively tolerant western Europe. Admittedly, Julien Offroy de La Mettrie didnt merely say the wrong thing in casual conversation: this French physician (1709-1751) published a pamphlet in which he wrote of human motivations as if they were analogous to energy-releasing springs inside machines. That was in 1747; the year before, La Mettrie had fled to Amsterdam from his native France. He had written a book, it seems, entitled The Natural History of the Soul. The Paris Parliament had disliked it to the point of ordering all copies burned. This time, La Mettrie took the precaution of publishing his pamphlet, entitled Man a Machine, anonymously. The Dutch, considered the most tolerant people in Europe, were scandalized and tried with a vengeance to discover who the pamphlets author was. They nearly found out, and so La Mettrie was forced to flee once more this time to Berlin, where he died four years later, at the age of forty-two. Though he was clearly ahead of his time, La Mettrie didnt invent the machine metaphor. Thats usually ascribed to René Descartes (1596-1650), writing a century earlier, in his De Homine. He too had moved to Amsterdam from his native France, at about the same time that Galileo was getting into trouble with the Vatican over the scientific method itself. Descartes didnt have to flee Holland, as did La Mettrie; he took the precaution, one might say, of publishing his book a dozen years after he was safely dead. Descartes and his followers werent trying to banish all talk of spirits; indeed, one of their characteristic concerns was to identify exactly where in the brain lay the seat of the soul. This endeavor was a continuation of a scholastic tradition which focused on the big reservoirs of cerebrospinal fluid inside the brain called the ventricles. Religious scholars of 500 years ago thought that the subdivisions of the soul were housed in these cavities: memory in one; fantasy, common sense and imagination in another; rational thought and judgment in a third. Like the bottle with the genie inside, the ventricles were supposedly containers for spirits. Descartes thought that the pineal gland was a better locale for the seat of government, on the grounds that it was one of the few brain structures that didnt come in pairs. Here at the fin de millennium, though there are theocratic countries where using code names would still be a good idea, we are generally more at ease when it comes to machine metaphors for mind. We can even discuss principled grounds for disputing any analogy of mind to machine. Minds, the argument goes, are creative and unpredictable; the machines we know are unimaginative but reliable so machines such as digital computers initially seem like an unreasonable analogy. Fair enough. But what Descartes established was that it was useful to talk of the brain as if it were a machine. You tend to make progress that way, peeling away the layers of the onion. Even if there is something else hidden beneath the obscuring layers, the scientist tentatively assumes that there isnt anything fundamentally unknowable, in order to test the alternative explanations. This scientific tactic not to be confused with a scientific conclusion has produced a revolution in how we see ourselves.
|
Mechanistic approaches to mind were, for a long time, missing an essential ingredient: a
bootstrap mechanism. Were used to the idea that a fancy artifact such as a watch requires an
even fancier watch designer. Its common sense just as Aristotles physics still is (despite
being wrong).
But, ever since Darwin, weve known that fancy things can also emerge (indeed, self organize)
from simpler beginnings. Even highly educated people, as the philosopher Daniel Dennett notes
in the preface to Darwins Dangerous Idea, can be uncomfortable with such bootstrapping
notions:
This shaping-up-the-improbable version of darwinism involves generating lots of copies of certain cerebral firing patterns, letting them vary somewhat, and then letting variants compete for dominance over a workspace (rather as those variants called bluegrass and crabgrass compete for my back yard). The competition is biased by how well those spatiotemporal firing patterns resonate with the bumps and ruts in the road the memorized patterns stored in the synaptic strengths. Such Darwin Machines are a favorite topic of mine, as youll see, but let us first get some idea of what intelligence is and isnt.
|
A useful tactic for exploring intelligence, one that avoids premature definitions, is the
journalists who-what-where-when-why-how checklist. Ill start with what constitutes intelligence
and when intelligence is needed, simply because the term is used in so many ways that it is easy
to talk at cross purposes, just as in the case of consciousness. Narrowing intelligence down a
little, without throwing out the baby with the bathwater, is the task of the next chapter, after
which Ill tackle levels of explanation and the consciousness confusions. A little ice-age perspective turns out to be important when exploring the evolutionary why aspects of intelligence, particularly in discussing our hominid ancestors. Alaskas coastline is the best place to see the ice age still in action Glacier Bay, some fifty miles long, was totally filled with ice only two hundred years ago. Now its populated with enough harbor seals, kayaks, and cruise ships to cause traffic jams. In the context of Glacier Bay, Ill raise the question of how jack-of-all-trades abilities could possibly evolve when efficiency arguments tell us that a streamlined specialist (the lean, mean machine beloved of economists) always does better in any one climate. The short answer? Just keep changing the climate, abruptly and unpredictably, so that efficiency doesnt remain the name of the game. In the fifth chapter, Ill discuss the mental machinery needed for parsing sentences that are complicated enough to require syntax. Many observers, myself included, suspect that the big boost in intelligence during hominid evolution was provided by those logical structures needed for a grammatical language (and also useful for other tasks). The chimpanzees and bonobos (the chimpanzees of the pygmies are a distinctly different ape, now called by the name that the natives were once said to use) provide some essential perspective for judging the role of language in intelligence and consciousness. Stones and bones are all that is left of our actual ancestors but our distant cousins show us what ancestral behaviors might have been like. The sixth chapter takes up the problems of convergent and divergent thinking in the darwinian context. Small neurobiology meetings, such as one I recently attended down on Monterey Bay, certainly illustrate convergent thinking all those specialists trying to find the one right answer, as the search for memory mechanisms narrows down. But divergent thinking is what creative people need to discover a scientific theory or write a poem or (at a more mundane level) need to make up all those wrong answers to use in multiple-choice exams for testing convergent thinking. Whenever a neuroscientist proposes one explanation for a memory storage mechanism, questioners from the audience promptly suggest several alternative explanationsones theyve dreamed up on the spot with divergent thinking. So, how do we shape up a novel thought into something of quality, without the equivalent of the guiding hand that shapes up a lump of clay into a pot? The answer may be in the title of chapter six: Evolution in the Brain. The same darwinian process that shapes up a new species in millennia, or a new antibody during the several weeks of an immune response, may also shape up ideas on the timescale of thought-and-action. In the penultimate chapter, Im going to venture past the mere analogy of mental processes to other known darwinian processes and propose how (the mechanistic how of the physiologist) our brains can manipulate representations in such a way as to cause a copying competition, one that can be darwinian and so shape up randomness into a good guess. This descent into cerebral codes (which, like the bar codes in supermarkets, are abstract patterns that stand in for the real thing) and cerebral circuitry (particularly the circuitry of the superficial cortical layers responsible for the brains interoffice mail) has provided me with my best glimpse so far of mechanisms for higher intellectual function: how we can guess, speak sentences weve never spoken before, and even operate on a metaphorical plane.. This cerebral version of a Darwin Machine is what, in my opinion, will most fundamentally change our concept of what a person is. Like the Dodo in Alice in Wonderland, who said it was better to demonstrate the game than to explain it, I will walk you through the darwinian process in some detail as it shapes up a thought and makes a decision. Trying to describe intelligence is not, I am happy to report, as difficult as describing how to ride a bicycle; still, you will understand it a lot better if you develop a feel for the process rather than being satisfied with an abstract appreciation (what youll get from chapters six and eight if you skip over my favorite chapter). In the final chapter, I will come back up for air and summarize the crucial elements of higher intelligence described in earlier chapters, focusing on those mechanisms that an exotic or an artificial intelligence would require in order to operate in the range spanning clever chimps to human musical genius. I will conclude with some cautions about any transition to superhuman intelligence, those aspects of arms races that the Red Queen cautioned Alice about why you have to keep running to stay in the same place.
|
You are reading HOW BRAINS THINK.
The paperback US edition |