Multiple representations for situated agent-based learning

Designers interact with the world not as actors following preconceived plans but related to the situations encountered. Learning the situatedness of design knowledge is as important as learning design knowledge. Knowledge can be represented in many ways. Multiple representations combine the advantages of
different representational forms within one system. Situated agent-based learning discovers and acquires useful knowledge and recognises the situation from multiple representations of the knowledge. A model of this situated agent-based learning is described and the use of multiple representations in learning knowledge is presented.