Bayesian views of generalized additive modelling

•Generalized additive models (GAMs) are a frequently used, flexible framework applied to many problems in statistical ecology. They are commonly used to incorporate smooth effects into models via splines, including spatial components in species distribution models. •GAMs are often considered to be a purely frequentist framework (‘generalized linear models with wiggly bits’), however links between frequentist and Bayesian approaches to these models were highlighted early‐on in the literature. From a practical perspective, Bayesian thinking underlies many parts of the implementation in the popular R package mgcv , so understanding these underpinnings can be informative during model building and assessment. •This article aims to highlight useful links (and differences) between Bayesian and frequentist approaches to smoothing, as detailed in the statistical literature, in accessible way, with a focus on the mgcv implementation. By harnessing these links we can expand the set of modelling tools we have at our disposal, as well as our understanding of how existing methods work. •Two important topics for quantitative ecologists are covered in detail: model term selection and uncertainty estimation. Taking Bayesian viewpoints for these problems makes them much more tractable in many applied settings. Examples are given using data from the NOAA Alaska Fisheries Science Center's groundfish assessment program.