Abstract
It is well established that visual working memory is capacity limited with retrieval performance declining as set size increases. The vast majority of studies investigating working memory capacity have focused on the maintenance phase of working memory tasks. In a recent study we found that neural resources allocated to individual items during encoding are an important factor in the overall capacity limitations of VWM. Here we expand upon those findings by investigating the effects of set size on encoding related processing. We use High Density Electroencephalography (HD-EEG) to analyze steady state visual evoked potentials (SSVEP) elicited by individual items in a standard change detection working memory recognition task. For each trial, participants viewed four shapes each flickering for 1s at distinct frequencies: 3Hz, 5Hz, 12Hz, 20Hz. Subjects were informed that squares presented in the array would never be probed. In this way a set size of two could be manipulated by having two of the shape be squares. After a blank delay period, a single shape appeared and participants had to respond whether the item was “old” or “new”. A Fast Fourier Transform (FFT) was performed on the data from which the fundamental frequency as well as the second harmonic of the probed stimulus were extracted. Behaviorally, we found participants were more accurate on set size 2 trials than set size 4 trials. Correspondingly, across a number of electrode locations, the amplitude of the second harmonic of the probed stimulus was found to be greater in the set size 2 condition compared to the set size 4 condition. This result suggests that working memory performance declines with increased set size in part as a consequence of encoding-related neural mechanisms.
Meeting abstract presented at VSS 2015