RT Journal Article
SR Electronic
T1 Early and Late Components of EEG Delay Activity Correlate Differently with Scene Working Memory Performance
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 143461
DO 10.1101/143461
A1 Timothy M. Ellmore
A1 Kenneth Ng
A1 Chelsea P. Reichert
YR 2017
UL http://biorxiv.org/content/early/2017/06/21/143461.abstract
AB Sustained and elevated activity during the working memory delay period has long been considered the primary neural correlate for maintaining information over short time intervals. This idea has recently been reinterpreted in light of findings generated from multiple neural recording modalities and levels of analysis. To further investigate the sustained or transient nature of activity, the temporal-spectral evolution (TSE) of delay period activity was examined in humans with high density EEG during performance of a Sternberg working memory paradigm with a relatively long six second delay and with novel scenes as stimuli. Sensor level analyses revealed transient rather than sustained activity during delay periods. Specifically, high amplitude activity encompassing the theta range was found early, during the first three seconds of the delay and correlated positively with subsequent ability to distinguish new from old probe scenes. Later, toward the end of the delay, lower amplitude activity encompassing the alpha/beta range was found that negatively correlated with subsequent ability to distinguish probes. Source level signal estimation implicated a right parietal region of transient early delay activity that correlated positively with working memory ability. Importantly, varying the trial temporal window duration and changing the baseline period for computation of TSE activity were found to influence the strength of the correlations obtained. This pattern of results adds to recent evidence that transient rather than sustained delay period activity supports visual working memory performance. The findings are discussed in relation to synchronous and desynchronous intra- and inter-regional neural transmission, and choosing an optimal baseline for expressing temporal-spectral delay activity change.