INAPPROPRIATE PARAMETERIZATION OF FOSSILIZED BIRTH-DEATH MODELS CAUSES INCORRECT ESTIMATION OF NODE AGES

Recent methodological advances allow researchers to make more complete use of the fossil record in divergence dating analyses using combined morphological and molecular datasets. Fossilized birth-death processes, in particular, present a coherent model of speciation, extinction and preservation dynamics that can be used to date phylogenetic trees. In this talk, I look at the effects of inappropriate parameterization of these models. A corollary to the fossilized birth-death process is that there is a possibility of sampling an ancestor and some or all of its descendents. I will use simulated and empirical data, and Bayesian implementations of fossilized birth-death processes, to examine the effect of not accounting for the possibility of sampling the ancestors of lineages present on the tree.