I selected a hard-to-use method, but realized that I knew a simpler yet better one.
I now see that the action I took had a bad, irreversible side effect.
I regarded that as an obstacle, but now I see that it was valuable.
Although that method caused some trouble, I learned a lot from using it.
To recognize those kinds of events would require Critics that work at higher levels—and all this suggests that our model of mind should include Selectors and Critics at every level.[133]
The following sections will discuss some of our many Ways to Think, and some of the Critics we use to recognize various ways in which we get stuck.
§7-4. What are some useful “Ways to Think?”
“When you want people to think you are brilliant, just imagine the worst thing that you could do and then do precisely the opposite.”
It is mainly when we get into trouble that we engage the processes that we call thinking or reasoning. However, ‘thinking’ is no single, definite thing; instead, we use different ‘Ways to Think’ for dealing with different types of obstacles. It ought to be one of our central goals—both for AI and for Psychology—to classify our Ways to Think. However, we don’t yet have systematic ways to classify those abilities—so I’ll just list some examples of them.
Knowing How: The best way to solve a problem is to know how to solve it and use that solution. However, we may not know how to retrieve what we know, or even know that we know it.
Extensive Search. When one knows no better alternative, one could search through all possible chains of actions—but this is usually impractical because that search grows exponentially.
Reasoning by Analogy: When a problem reminds you of one that you solved in the past, you may be able to adapt that case to the present situation—if you have good ways to tell which similarities are most relevant.
Divide and Conquer. If you can’t solve a problem all at once, then break it down into smaller parts. For example, every difference we recognize may suggest a separate sub-problem to solve.
Planning. Consider the set of sub-goals you want to achieve and examine how they affect each other. Then, with those constraints in mind, propose an efficient sequence for achieving them.
Simplification.Sometimes, a good way to make a plan is to make a simplified problem by ignoring some aspects of the original one. Then any solution to the simplified one may serve as a sequence of stepping-stones for solving the initial problem.
Elevation. If you are bogged down in too many details, describe the situation in more general terms. But if your description seems too vague, switch to one that is more concrete.
Reformulation. Find a different representation that highlights more relevant information. We often do this by making a verbal description—and then ‘understanding’ it in some different way!
Self-reflection. Instead of pursuing a problem itself, ask what makes that problem seem hard, or what you might be doing wrong. This can lead to better ways to represent the problem.
Contradiction. Try to prove that your problem cannot be solved, and then look for a flaw in that argument.
Use external representations: If you find that you’re losing track of details, you can resort to keeping records and notes, or drawing suitable diagrams.
Simulation. One can avoid taking physical risks if one can predict “what would happen if” by imagining possible actions inside the mental models that one has built.
Correlation. When certain events seem to happen together, try to find ways in which they may be connected.
Logical Reasoning.We sometimes make ‘logical chains of deductions,’ but those conclusions may be wrong because of exceptions to our assumptions.[134]
Wishful thinking. Imagine having unlimited time and all the resources that you might want. If you still can’t envision solving the problem, then you should reformulate it.
Impersonation. When your own ideas seem inadequate, imagine someone better at this, and try to do what that person would do.
Cry for help.You can always resort to other techniques that most people would call “emotional.”
Resignation.Whenever you find yourself totally stuck, you can shut down the resources you’re using now and relax, lay back, drop out, and stop. Then the ‘Rest of Your Mind’ may find an alternative—or conclude that you don’t have to do this at all.
How do we choose which of these to use? The Critic-Selector model suggests that each person can recognize particular ways in which one gets stuck—and can use each such diagnosis to select one or more particular ways to deal with that kind of predicament. We each do this in different ways, and the Critics that we each develop must be among our most precious resources.
§7-5. What are some useful Critics?
“Don’t pay any attention to the critics. Don’t even ignore them.”
We are always developing new ways to think—so we also need to make Critics to help to select when to use each of those techniques—by recognizing which kinds of problem we face. This means that our Critics must serve as ways to classify all the barriers, obstacles, impasses, or snags that make our problems hard to solve. Indeed, it would be an important goal, both for people and for computing machines, to have a systematic catalog of the types of problems we most frequently face.[135] However, we do not yet have adequate, orderly ways to do this—so here we’ll merely try to describe a few types of Critics that people seem to use.
Innate Reactions and built-in Alarms. Many types of external events arouse detectors that make us quickly react, as when an object is quickly approaching you, a light is too bright, you touch something hot, or hear a loud sound. We’re also born with ways to detect abnormal conditions inside our skins—such as wrong levels of chemicals in the blood. Many of these have built-in connections that make us react to correct those conditions, in ways that work so automatically that we react to them without any thought.
However, an unexpected touch, sight, or smell—or a sense of hunger, fatigue, or pain—does interrupt the flow of our thoughts. Indeed we’d never survive through our infancies unless such emergencies (or opportunities) could pull us away from our reveries. We can sometimes suppress some of those alarms; for example, when we suppress a sneeze, or stop ourselves from scratching an itch. But if you try to hold your breath, you can’t resist the alarm of asphyxia—and it is hard to ignore a baby’s cry, a constantly ringing telephone, or an amorous opportunity.
133
At the lowest levels, the Critics and Selectors become the same as the
134
Logic can be useful after a problem is solved, for making credit assignments [§8.5] and for solving simplified versions of problems. See §§Logic.
135
There is an excellent survey of attempts to classify Problem-Types on Manuela Viezzer’s webpage at