A Cognitive Theory of Information

Ronaldo Vigo (Ohio University)

Abstract: In this talk, I will introduce and discuss a new notion and measure of information referred to as representational information. The measure was derived from Generalized Invariance Structure Theory (GIST; Vigo 2013, 2016), a mathematical theory of human categorization and concept learning (i.e., of human generalization) that is based on invariance and abstract symmetry principles. In the first part of the talk, I will give a basic introduction to GIST, followed by a discussion of the inadequacy of Shannon information theory in cognitive science. The second part of the talk will consist of an empirical and theoretical discussion of the notion of informativeness and how representational information effectively accounts for human informativeness judgments and for the key constructs of prototypicality and representation in cognitive science.