Abstract
Prior studies of information transfer rates from brief iconic to durable post-iconic levels of processing differ significantly. Estimates range from roughly 10 ms/letter to 50 ms/letter consolidation rate, indicating that initial readout of information from iconic to post-iconic levels of processing may not be identical to subsequent processes of consolidation of information in visual working memory. Moreover, Woodman and Vogel (2005) report that concurrent WM maintenance does not affect the efficiency of information consolidation into WM. To clarify, we examined how maintaining variable-load color information in working-memory interacted with the information transfer from iconic to post-iconic levels. Using a dual-task paradigm, observers maintained color information, from a 300-ms display consisting of one, two or else three disks, in WM while concurrently reading out information from iconic storage of a briefly presented (10 ms) six-letter array. A random noise mask followed the letter array at interstimulus intervals from 0 to 100 ms. A probe display was then presented, in which on half the trials, the probe was a color disc, and on the other half, a single letter. Observers had to indicate whether the probe was or was not shown in the prior respective color or letter display. Results revealed concurrent readout of letters worsened WM performance, as load increased. Overall, iconic readout rose steeply over the early ISIs ranging from 0 to 60 ms, followed by a plateau from 60 to 100 ms. When concurrent WM maintenance was required the rapid rate of iconic readout rate from 0 to 60 ms decreased as WM load increases, indicating that the initial rate of readout from iconic to post-iconic levels of processing is slowed down as more information is maintained in WM. In contrast, the later, ISIs from 60 to 100 ms of iconic readout did not vary with WM load.
Meeting abstract presented at VSS 2014