Abstract
A number of studies demonstrate the important role of ensemble summary statistics in many aspects of visual perception and memory. However, it is still the subject of debate whether attentional resources are involved in the extraction of these statistics. In the present study, we tried to answer this question using the Rapid Serial Visual Presentation (RSVP), a demanding task which is known to produce a temporary processing lapse known as the attentional blink. We modified the paradigm to probe the precision of ensemble representation at different temporal points of AB. 13 participants viewed an RSVP stream of black letters in the center of the screen. One letter in stream was white (it was the first target to be remembered in the AB condition – T1) and followed by a set of white circles. Circle sizes were normally distributed in each trial and their mean size varied from trial to trial. After the stream, participants adjusted a probe circle to the mean size of the set and reported the white letter. The lag between T1 and the set could be 100, 200, 400, or 700 ms. In a control condition, participants were exposed to same stimulation but had to recall only the average ignoring T1. We found that averaging error was high (~30%) at 100- and 200-ms lags and decreased at 400- and 700-ms (~16-20%) in the AB condition. This suggests that attention engaged in processing T1 was diverted from ensemble processing. Surprisingly, the pattern was the same in the control condition. Our speculation is that, despite the instruction to ignore T1, a single white letter in a stream could capture attention involuntarily and produce an AB-like pattern. This speculation needs thorough examination in future studies. Overall, the results of the experiment suggest attentional involvement in ensemble processing.
Meeting abstract presented at VSS 2016