ABSTRACT
This chapter outlines what one needs to understand before one attempts to model knowledge and learning in economics. It explains how we can go beyond both Inductivism in particular as well as its compromised form in Conventionalism and still address the need to recognize knowledge and learning in neoclassical economics. Just recognizing that a consumer might have a particular theory of his or her preferences is not enough. How a decision-maker deals with refutations also depends on the decision-maker’s theory of knowledge and learning. The Socratic view of knowledge and learning does not preclude the use of probability notions. It does, however, require that the model builder be explicit about how the decision-maker incorporates probability notions in his or her decision-making process. Once one recognizes the fallibility of all knowledge, particularly knowledge necessary for the process of decision-making, one must also recognize the necessity of addressing the possibility of decision errors.
