Behavioral Biases in the Uncertainty Quantification Process

Humans play important roles in the process of quantifying uncertainty. The participation of humans in this important exercise opens the process to behavioral biases. In this paper, we examine the different types of biases that may occur when quantifying uncertainty using a process-oriented framework. In particular, we examine the potential biases in each of the major phases of the uncertainty quantification process. Traditionally, most of the literature has focused on biases that occur in the elicitation phase, but we discuss how these can also permeate in the pre- and post-elicitation phases. The recognition of biases in these different phases allows practitioners to potentially avoid, minimize, and correct behavioral biases that result in the multiple stages of the uncertainty quantification process.