ABSTRACT
This chapter examines a psychological evidence for the model. It begins by providing evidence for the reliability and validity of the representational system. The chapter presents a study that collected goodness-of-answer ratings in order to assess the plausibility of the arc search procedures, distance limitations, and node restrictions of the model. It briefly describes a computer simulation of question-answering (Q/A) procedures. This computer simulation demonstrated that the arc search procedures are computationally sufficient. Q/A procedures with high overlap have corresponding subject-generated answers with a high degree of overlap. The chapter summarizes the nature of the Q/A procedures. A representational system can be evaluated in terms of its reliability and validity. A reliable system generates similar representational structures for a knowledge domain across various scorers and subjects. In the goodness-of-answer rating task, subjects first read a story and then rated the adequacy of possible answers to questions from the story.
