Across three experiments, we found that implicit distractor-location learning is viewer dependent when embedded in active behavior. That is, spatial inhibition cannot be abstracted from the agent moving through a 3D world and how they can suppress sampling the world from their perspective. [4/5]
