I will compare my theory to non-visual schemes: Structure-Behavior-Function, PI, ACME, and SME. I will respond to all theoretically, and to Structure-Behavior-Function (SBF) experimentally and computationally by implementing an SBF model into my agent. The comparison of the non-visual to the visual will be a part of how the program will run- when the non-visual fails, the visual will attempt to help, showing that such representations are useful.
It's important for me to compare my theory to these other theories because it will show evidence for when visual representations give some benefit. If, for example, these non-visual schemes work just as well in all conditions, my argument for the usefulness of visual representations will not be supported.