Bayesian networks are graphical models that have been developed in the field of artificial intelligence as a framework to help researchers and practitioners apply probability theory to inference problems of substantive size as encountered in real-world applications. Influence diagrams (Bayesian decision networks) extend Bayesian networks to a modeling environment for coherent decision analysis under uncertainty. This article provides an overview of these methods and explains their contribution to the body of formal methods for the study, development and implementation of probabilistic procedures for assessing the probative value of scientific evidence and the coherent analysis of related questions of decision-making.