Abstract
High-level visual cortical areas that encode object identity have a coarse representation of visual space. Consequently, fine-grained representations of space from other cortical areas might have to be integrated in order to encode an object's identity and precise location. In the present study, we used an attentive tracking task to separate the processing of identity and location information.
Subjects viewed a circular array of eight ring placeholders displayed around a central fixation point. They covertly tracked a colored ring that rapidly stepped through the array. A random letter was briefly shown inside the ring at each location. In each trial the ring would change color at one random location. The task was to report the cued location as well as the letter displayed in that location.
Observers accurately reported the cued location, but often failed to report the correct letter. Subjects made systematic errors by choosing letters from adjacent locations, indicating that they successfully processed the identity of the letters without explicit knowledge of their location. These errors are notable because they involve misconjunctions across both space and time.
These results indicate a failure to integrate a letter's identity with its precise location, and suggest a spatiotemporal dissociation between the processing of identity and location.