Pre-cues increase capacity at the expense of precision in visual working memory.

Activity in a network of frontal, parietal, and occipital cortical areas has been associated with retention in visual working memory tasks. While parietal and occipital cortex are thought to encode and maintain information about objects and their spatial locations, the role of the frontal cortex in pure storage functions remains unclear. According to a recent biophysical network model of WM, the frontal cortex regulates capacity via excitatory feedback to posterior storage areas (Edin et al., 2009). Such feedback has been proposed to boost capacity at the cost of reduced precision. A recent study by Roggeman et al. (2013) confirmed the predicted trade-off in a task requiring the short-term retention of spatial locations. Moreover, a follow-up fMRI experiment revealed a correlation between the trade-off effect and increased BOLD responses in frontal and parietal cortex. In the present study, we used similar methods to determine whether the observed trade-off effect generalizes to color WM. Specifically, participants performed a cued color WM task requiring retention of either 2 or 4 items in WM. To elicit the postulated boost input, at the beginning of each trial a cue was presented (75% valid) indicating whether 2 or 4 items would be presented on that trial. Behavioral results confirmed the predicted trade-off between capacity and precision, and the cue-related change in capacity predicted changes in precision on a subject-by-subject basis. Additionally, analysis of event-related potentials (ERPs) revealed a significant cue-related modulation of the P1 ERP component in response to the memory display; the P1 was higher amplitude in Cue4/SS4 versus Cue2/SS4 condition. Additionally, we also observed a cue-related increase in theta-band oscillatory coupling between frontal and posterior electrodes. However, this difference was not predictive of cue-related changes in either capacity or precision. Meeting abstract presented at VSS 2015.