Abstract
When two target letters are presented among distracters in a rapid serial visual presentation paradigm (RSVP), correct identification of the first target often results in a deficit for identifying the second target if it appears within 200–500ms after the first target. This ‘attentional blink’ shows that the time-course of encoding successive elements in the stream is interrupted by the presence of the first target. However, less is known about how the presence of a foveal target affects the spatial encoding of information around the target location.
To test this, observers participated in a dual-task RSVP paradigm in which subjects were instructed to identify a white target letter embedded in a stream of black distracter letters. A randomly chosen full-field outdoor scene was placed behind each letter in each frame of the stream. Following each stream, subjects were presented with a test scene and asked if they recognized this scene as one from the RSVP display. On half of the trials the test scene was one of the scenes presented during the presentation stream. Correct recognition of the test scene was at chance performance (50%) when it had occurred behind one of the distracter letters. Surprisingly, we found that observers performed significantly above chance on trials when the test scene coincided in time with the target letter. We also report a deficit in recognition performance of the test scene when it occurred immediately after the target letter - an effect resembling the ‘attentional blink.’ These results suggest that the entire visual scene is encoded when the target letter is identified, and that the visual system performs something analogous to a ‘screen capture’ at behaviorally relevant moments in time in order to store this information for further processing.
This research was funded by National Institutes of Health Grant NIH EY12925.