In a later paper, this author says more about such massive mental cascades:
It seems, then, that both the quantity and quality of thoughts build during hypomania. This speed increase may range from a very mild quickening to complete psychotic incoherence. It is not yet clear what causes this qualitative change in mental processing. Nevertheless, this altered cognitive state may well facilitate the formation of unique ideas and associations. … Where depression questions, ruminates and hesitates, mania answers with vigor and certainty. The constant transitions in and out of constricted and then expansive thoughts, subdued and then violent responses, grim and then ebullient moods, withdrawn and then outgoing stances, cold and then fiery states—and the rapidity and fluidity of moves through such contrasting experiences—can be painful and confusing.[42]
It is easy to recognize such extremes in the mental illnesses called ‘bipolar’ disorders, but Chapter §7 will conjecture that we also use such processes in the course of everyday commonsense thinking. Thus, you might use a procedure like this whever you face a new problem:
First, shut most of your Critics off. This helps you to think of some things you could do—without concern about how well they might work—as though you were in a brief ‘manic’ state.
Then, you could turn many Critics on, to examine these options more skeptically—as though you were having a mild depression.
Finally, choose one approach that seems promising, and then proceed to pursue it, until one of your Critics starts to complain that you have stopped making progress.
Sometimes you may go through such phases deliberately. However, my conjecture is that we frequently do this on time-scales so brief that we have no sense that it’s happening.
Learning from Failure
“Never interrupt your enemy when he is making a mistake.”
Many things we regard as positive (such as beauty, humor, and pleasure itself) may be partly based on censorship—hence, to that extent, could be considered negative. Thus pleasure can seem ‘positive’ to the processes that now are presently “in control”—no matter that other processes (whose expressions are currently being suppressed) might otherwise see this as ‘negative.’ (See §9-2 of SoM.) For, “I’m enjoying this” could mean, both at once, “I want to stay in my present state,” and “I want to prevent any changes in it.”
Student: But I thought that it was widely believed that learning works by ‘reinforcing’ connections that have led to success, and by weakening those that contribute to failure. Many educators say that we should always make it pleasant to learn, because pleasure is our reward for success—whereas failure deters and discourages us.
That popular view is mainly based on research (mostly done with pigeons and rats) that also shows that quicker rewards make learning more rapid. This has many teachers toward the idea that learning should be a pleasant experience. However, we should not be too quick to apply this idea to beings like us, who also can learn by reflecting on the things they have done!
I’m not saying that ‘reinforcement theory’ is wrong—but that, for humans, it’s just part of the story; in §8-5 I’ll argue that what we can learn from how we have failed could be more important than ‘reinforcement’ can be—at least, for our highest levels of thinking.[43] For, while pleasure may help us learn easy things, section §9-4 will argue that we may need to endure some suffering to make larger-scale changes in how we think. If so, as an ancient Stoic might say, rewarding success can lead you to celebrate more than to investigate. Here are a few other reasons why to ‘learn from success’ is not always wise—especially when that success was expected.
Reinforcement can lead to Rigidity. If a system already works, additional ‘reinforcement’ could make some its internal connections become stronger than they need to be, which could make it harder for that system to adapt to later new situations.
Dependency leads to Side Effects. If a certain resource R has worked so well that other resources have come to depend on it, then any change you make in R will now be more likely to damage those others. In other words, as the saying goes “Don’t fix it, unless it is broken.”[44]
Negative Expertise. One way to avoid such side effects is to leave an established resource unchanged, but to add Critics and Censors to intervene in conditions where it has failed to work. In other words, treat them as exceptions to rules.
Radical Learning: You can “tune up” a skill by many small steps, but eventually no more small changes will help, because you have reached a local peak.[45] Then further improvement may require you to endure some discomfort and disappointment. See §9-4.
Papert’s Principle: When two or more of your methods conflict, then instead of seeking a compromise, abandon the lot and then try something else. Many steps in mental growth are less based on acquiring new skills, but more on learning better ways to choose which older ones to use. [See §10-4 of SoM.]
For all of those reasons, we need to learn, not only methods that worked in the past, but also which methods have failed—and why—so that one can avoid the most common mistakes.
Student: Yes, but why can’t we do that by breaking connections—so that once you’ve made a bad mistake, your brain won’t ever do it again?
One reason why this is a bad idea is that you’ll lose the opportunity to understand just what went wrong (so that you can later avoid related mistakes). A second problem with this tactic is that whenever you change some of a system’s connections, this may also affect some other behaviors that are partly based on those same connections. If you don’t know quite how that system works, then you’re in danger of making it worse by ‘correcting’ any remaining mistakes.
Programmer: I know exactly what you mean. Every attempt to improve a program is likely to introduce new bugs. That’s why new programs so often contain very big sections of ancient code: no one remembers quite how they work, and hence they’re afraid to change them.
Student: But what if you have no alternative, because something is wrong that you need to fix.
Perhaps our most important ways to improve ourselves come from learning to think about thinking itself—that is, to ‘reflect’ on what our minds have been doing. However, to do this one must first learn to enjoy the distress that results when one’s forced to inspect oneself. See §8-5 and §9-4.
Varieties of Negative Expertise
Creativity: Why do some people get more good ideas? I did not specify ‘new’ ideas—because it is easy to build a machine that spouts endless streams of things that have never been seen; what distinguishes thinkers that we call ‘creative’ is not how many new things they produce, but how useful are the few they produce. This means that those artists have ways to suppress—or not even generate—products that have too much novelty, leaving only the ones that are just different enough to be useful.
42
Kay Redfield Jamison, “Manic-Depressive Illness and Creativity,” Sci. Amer., Feb. 1995 V. 272 No. 2 Pp. 62-67
43
Most animals simply do not have the high-level resources that people have, and this makes it risky to apply to ourselves what we learn from laboratory animals.
45
Thus, to ascend from the top of Kilimanjaro to the summit of, say, Mt. Everest, you would have to climb down and then up again.