Abstract
We often need to process the spatial relationships between objects; e.g., when noticing that your mouse is to the right of your keyboard. One challenge of spatial relationship judgments is properly binding different identities with their respective positions, even when those identities can be extracted in parallel. One solution to this problem is to select one object at a time - if so, then even in the simplest spatial relationship judgments where we feel that we select two objects at once (keyboard and mouse), we must actually select single objects in a sequence over time (keyboard -> mouse). To demonstrate this effect, we primed participants with a Navon letter task to select either single (local attentional scope) or multiple (global attentional scope) objects, immediately before performing either a spatial relationship judgment between two colors (which should require proper binding) or a same-different identity judgment between the two colors (where binding should not be important). There was an interaction such that spatial relationships were judged faster under single object selection primes, and same-different identity was judged faster under multiple-object selection primes. These results provide evidence for a counter-intuitive possibility - that even when object identities are available, knowing where they are relative to each other requires that we select them sequentially over time.