The viability of many species depends on functional connectivity of their populations through dispersal across broad landscapes. This is particularly the case for the endangered brown bear in north-western Spain, with a total population of about 200 individuals in two subpopulations that are separated by a wide gap with low permeability. Our goal in this paper is to use state-of-the-art connectivity modeling approaches to provide detailed and quantitative guidance for conservation planning efforts aimed at improving landscape permeability for brown bears in Spain, with a particular focus on alleviating the barrier effect of transportation infrastructure. We predicted a regional connectivity network for brown bear by combining a multiscale habitat suitability model with factorial least-cost path density analysis. We found that the current composition and configuration of the landscape considerably constrain brown bear movements, creating a narrow bottleneck that limits flow of individuals between the two subpopulations. We identified key locations along the predicted corridors where efforts to increase road and railway permeability should be prioritized. The results provide a foundation for the development of spatially optimal management strategies to enhance connectivity within and between the subpopulations and to mitigate the impact of potential barriers to movement.