... representations.

This problem has been identified by Yarlett and Ramscar (2000), whose system takes two different symbols and evaluates their similarity using Latent Semantic Analysis (Landaur, 1998). The analogical mapper treats as identical any symbols that are evaluated as sufficiently similar.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... respectively.

The terms ``entity,'' ``element,'' ``object,'' ``manipulation,'' ``action,'' and ``transformation'' are used merely to differentiate the visual, non-visual, and super-ordinate counterparts of things and operators. Their common sense meanings in English do not have anything to do with which term gets used with which meaning.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...locations

Relative locations are classified under primitive visual relations.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...map.

A map is called a match hypothesis in the SME literature.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... it.

It may well be that you cannot understand how to solve a problem in a physical system without some perceptual representation of it. The visual representations I'm dealing with in this dissertation are more abstract than a full-blown, pictorial image. You can imagine how to decompose a ray of radiation quite realistically, and solve a problem with this in mind, but the visual level I'm talking about is more akin to a sketched diagram than an instructional video. They are so abstract that they are often ambiguous as to what they represent (e.g. a circle representing a person).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... propositional.

Barsalou argues against the existence of purely non-perceptual symbols, but even his theory is propositional: ``Because perceptual symbol systems have the same potential to implement propositions, they too are propositional systems'' (Barsalou 1999, p. 26).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.