Picture-Frames: Every slot of a property list is directly connected to the name of that frame, as in the list above for describing an Apple. However, other more complex kinds of frames may have more connections between their various slots. For example, to represent some view of a room, we could use what we call a “picture frame” to represent each wall of that room by in terms of a few large spatial regions as shown below. Then each such region can have some links that describe the objects close to that part of the wall, as well as some links to other nearby regions. This kind of structure would allow us to do a good deal of commonsense spatial reasoning. [See §§24 of SoM.]
Frames for Including Additional Slots: It makes sense to allow each Frame to include some additional slots for representing knowledge that is not already described by the networks contained inside that frame. For example, here is a Trans-frame for Joan’s trip to New York:
This frame includes two semantic networks that describe the situations Before and After that trip was taken. However, it also contains other slots that describe when, how and why Joan took that trip. Then the default assumptions included in those slots can supply additional knowledge for answering such questions as these.
Where did that action occur and when? Who or what caused it to happen?
Was it intentional or not? What purposes was it intended to serve?
What devices or tools were used? What were its other side effects?
Which resources did it engage? What was expected to happen next?
This suggests an explanation of how we quickly use our commonsense knowledge—without any sense that we’re doing it: it is an example of the “Immanence Illusion” that we described in §4-3.1. As soon as you activate such a frame, then many questions that you might otherwise ask will already be answered before you can ask them—because they are among the default values of that frame’s slots. For example, if you heard that Charles was holding a book, you would not stop to ask why he was holding it; you would simply assume that he has the most usual goal for which any person holds anything—namely, to keep it from falling to the floor.[174]
Connectionist and Statistical Representations.
Student: This book suggests some ideas about how high-level knowledge-based systems could come to achieve things like human commonsense reasoning. But why were no such systems built in the past?
Work on such systems almost came to a stop in the 1980’s because most researchers recognized that that this would need ways to acquire and to organize millions of fragments of commonsense knowledge. That prospect seemed so daunting that most researchers decided to search for simpler alternatives. This led to many attempts to design some single process that would somehow evolve whatever it needed—along with learning all the knowledge it would need by interacting with the external world. Some of these “baby machines” did learn to do some useful things (such as to recognize various kinds of patterns) but as we noted in Chapter §6, none of them went on to develop more higher-level reflective ways to think.
Why were none of those “Baby Machines” able to keep extending their abilities? It appears to me that this failure came mainly because most of their designers decided that their systems should represent the knowledge they were to acquire mainly in numerical terms. Consequently, most of those ‘baby machines’ were designed to use the techniques called Neural Networks, Perceptrons, Fuzzy Logic systems, and “Statistical Learning Programs.” All such systems represent knowledge about the relations between things by assigning numerical ‘weights’ or ‘strengths’ to connections inside a network of nodes. Such a representation might look like this:
Here we see only one kind of link, which reduces every type of relationship to a single numerical value or ‘strength.’ The trouble with this is that a single numbers is ‘opaque’ in the sense that it has so little expressiveness. For, whenever one computes an average or a probability, this conceals the knowledge or evidence that led to it.[175] For, consider that if you only see the number 12, you cannot tell if that number represents 5 plus 7, or 9 plus 3, or 27 minus 15! Did it come from counting the eggs in a nest, or from counting the years of your grandchild’s age? For example, if you represent the concept of ‘apple’ this way, your machine may be able to recognize an apple, but it won’t be able to reason about it. In short, numerical representations become obstacles to using more reflective ways to think—because it is difficult for other, higher-level processes to think about the knowledge that such systems contain. [We’ll discuss this more in §§§Opacity.]
Let’s contrast this with representing a concept of “apple” by using a semantic network like this:
This kind of representation can help you answer many questions about an apple, such as where you can find one and what you can use it for—because a semantic network can express all sorts of different relationships, whereas numerical representations ultimately limit a system’s mental growth, because they provides no parts that the rest of a mind can use to produce more elaborate explanations.
Micronemes for Contextual Knowledge. We always face ambiguities. The significance the things that you see depends on the rest of your mental context. This also applies to events in your mind, because what they mean depends on which mental resources are active then.[176] In other words, no symbol or object has meaning by itself, because your interpretation of it will depend on the mental context you’re in. For example, when you hear or read the word block, you might possible think that it means an obstacle to progress, a certain kind of rectangular object, a wooden board to chop things on, or a stand on which things in an auction are shown. Then which interpretation will you select?
Such choices will depend, of course, on the preferences that are active in your current mental context—which, somehow, this will dispose you to make selections from such sets of alternatives as these:
Conceptual or material.
Animal, mineral, or vegetable.
Well-established or speculative.
Common, rare, or irreplaceable.
Robust, fragile or reparability.
Indoors or outdoors.
Public or private.
Residence, office, theater, or car.
Urban, rural, forest, farm.
Color, texture, hardness, strength.
Irregular or symmetrical.
Hunting, gambling, entertainment.
Cooperation, conflict, etc.
Many contextual features like these have common names, but many others (such as aromas) have no such words. I have proposed to use the term “micronemes” for the myriad of nameless clues that color and shade how we think about things, and the diagram below suggests some machinery through which such contextual features could affect many mental processes. Imagine that the brain contains a bundle of thousands of wire-like fibers that pass through a great many of other structures inside that brain—so that the state of each of those ‘micronemes’ can influence many processes:
174
I should add that a frame can include some additional slots that activate other processes or sets of resources. This way, a frame could transiently activate ways to think—so that one almost instantly knows how to deal with some familiar object or situation.
175
I should add that numerical representations have many useful applications. However, even when those numbers have some practical use, one can only alter them by increasing or decreasing them, but cannot add other nuances. It is much the same ‘logical’ systems; each ‘proposition’ must be true or false, so the system still uses something like numbers, except that their values can only be 0 or 1. Also, see see SOM, section 5.3.