Climate variability and change occurs largely as a result of the interaction between its two dynamic components; the atmosphere and ocean. This coupled interaction is a two-way interaction whereby the atmosphere drives the ocean and the ocean drives the atmosphere, such that climate variability and change include components that are atmosphere- and ocean-driven. The atmosphere- and ocean-driven components of the climate interaction are often isolated with stand-alone model simulations (atmosphere-only, ocean-only or slab-ocean), where one climate model is used and the interaction with the other climate component is prescribed as boundary conditions. Here, we describe a partial coupling approach that can be used to disentangle the coupled interaction between atmosphere and ocean while both climate components are coupled. In a partial coupled experiment, the impact of the ocean dynamical change is removed from coupling with the atmosphere, such that only changes of atmospheric origin can drive the ocean. The partially coupled experiment thereby simulates an atmosphere-driven one-way climate interaction, as well as, the atmosphere and ocean dynamical responses to this interaction. When compared to the atmosphere and ocean responses in a fully coupled experiment, the partial coupling method provides insight into the roles of the atmosphere and ocean and the coupled feedbacks between them in the fully coupled climate interaction. We discuss the applications of the partial coupling method for separating out the atmosphere- and ocean-driven components of North Atlantic SST variability and the climate response to CO2 quadrupling, and show that the partial coupling method provides a more accurate attribution of roles of the atmosphere and ocean compared to stand alone studies