Computer-based studies examining how adults learn name-object mappings typically present objects from a single view across trials (e.g., from above). In the real-world, however, we encounter objects from a variety of views. In the current study we explored the real-time dynamics of object rotation on word learning. Each participant received three word learning blocks. Within blocks objects were either presented always from the same view, from one of two views or from one of four views. Overall, adults quickly learned the 24 name-object mappings. Initial accuracy was best when objects were presented from only one or two views, however, by the sixth encounter there were no differences between conditions. Reaction times decreased within blocks at similar rates across conditions, indicating that object rotation does not impede skilled (adult) word learners. Trial-by-trial analyses provide additional insight into the dynamics and cognitive processes underlying learning names for rotated objects.