RT Journal Article
SR Electronic
T1 Overlapping Neural Representations for Dynamic Visual Imagery and Stationary Storage in Spatial Working Memory
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 2022.09.24.509255
DO 10.1101/2022.09.24.509255
A1 Eren Günseli
A1 Joshua J. Foster
A1 David W. Sutterer
A1 Lara Todorova
A1 Edward K. Vogel
A1 Edward Awh
YR 2022
UL http://biorxiv.org/content/early/2022/09/26/2022.09.24.509255.abstract
AB Representations in working memory need to be flexibly transformed to adapt to our dynamic environment and variable task demands. Recent work has demonstrated that activity in the alpha frequency band enables precise decoding of visual information during both perception and sustained storage in working memory. Extant work, however, has focused exclusively on the representation of static visual images. Here we used EEG recordings to examine whether alpha-band activity supports the dynamic transformation of representations in spatial working memory using an imagery task that required the active shifting of a stored position to a new position. In line with recent findings, a common format of alpha-band activity precisely tracked both the initial position stored in working memory as well as the transformed position, with the latter emerging approximately 800-1200 ms following an auditory cue to “rotate” to a new position. Moreover, the time course of this transformation of alpha activity predicted between-subject differences in manual reaction time to indicate the new position (Experiment 1), as well as within-subject variations in saccade latency in a speeded version of the task (Experiment 2). Finally, cross-training analyses revealed robust generalization of alpha-band reconstruction of working memory contents before and after mental transformation. These findings demonstrate that alpha activity tracks dynamic transformations of representations in spatial working memory, and that the format of this activity is preserved across the initial and transformed memory representations. These findings highlight a new approach for measuring voluntary shifts in online memory representations and show common representational formats during dynamic mental imagery and the maintenance of static representations in working memory.Competing Interest StatementThe authors have declared no competing interest.