Qualitative and quantitative evaluations of mathematical models for animal learning and behaviors

Abstract

Mathematical or computational models have played important roles in investigating animal behaviors and underlying processes. A traditional approach in psychology to evaluate such models is “qualitative” one, which examines whether the model can explain observed phenomena or not. In contrast, “quantitative” approaches, which evaluate how well the model can account for the observed data compared to alternative models, have been proved useful especially in neuroscience. This article reviews the quantitative approaches to model evaluation. The advantages and limitations of the quantitative approaches are discussed. The author suggest that the quantitative and qualitative approaches are complimentary and jointly provide a powerful framework for investigating psychological processes underlying animal learning and behaviors. Key