Abstract
When monitoring two RSVP streams of letters for one or two targets indicated by one or two green annulus cues, is attention to two cues independent? Letter streams were presented above and below a fixation cross for 80ms/letter. On a given trial there was one cue, two simultaneous cues, or two cues presented sequentially at 80 ms SOA in the two streams. Trials were intermixed.
Single-cue trials showed bias toward the top stream, in report of the true target (the modal report) and in reports of letters in neighboring positions. The modal report of single bottom cues was the +1 letter (following the annulus). In the simultaneous condition reports from the top stream were highly similar to top stream reports with a single top cue. In contrast, report from the bottom stream was much worse in the simultaneous-cue than the single-cue case. Sequential trials showed that report from the first cued stream (top or bottom) was similar to a single cue condition, especially for the top stream, whereas there was an overall decrease in report of letters from the second cued stream, as well as a displacement towards the +1 letter.
The evidence suggests that attention was not evenly divided between the two streams but could be focused rapidly on one stream for single cues or first sequential cues. The probability of reporting cued letters from both streams (including letters in the neighborhood) was lower than predicted by multiplying the probabilities of getting top and bottom singletons, showing that processing of the two streams was not independent. Moreover, the reported letters showed more temporal displacement in the simultaneous condition than the single condition. We will report results of a second experiment in which the streams were placed to the left and right of fixation.