Abstract
The spatial distribution of alpha-band activity across the scalp measured by electroencephalography (EEG) can be used to track specific spatial representations of stimuli held in working memory (WM; Foster et al. 2016; MacLean et al. 2019). Here, we tested the extent to which spatially specific information is maintained in WM when it is no longer task-relevant. Participants (n=7) performed a simple delayed recall task in which a circular stimulus was briefly presented (250ms) at locations on an imaginary circle around fixation. After a retention interval (1750 ms), participants recalled the task relevant feature of the initial memorandum. In different blocks, participants recalled either the stimulus location (requiring maintenance of spatial information during the retention period) or color (maintenance of spatial information not required) while EEG was recorded. Precision was lower when recalling color compared to spatial location (p<.05) and guess -rates were not different (p>.05). We used an inverted encoding analysis using EEG alpha power to estimate the spatially selective response throughout the trial. The spatially selective response was quantified using the slope of the response profile. We observed a robust stimulus representation (greater positive slope) during the stimulus presentation phase of the trial in both the spatial and color WM conditions. This was followed by a sustained spatially specific response in the spatial memory condition throughout the entire 1750 ms retention period, whereas in the color memory condition the spatially specific representation was maintained for ~750 ms post-stimulus. Together, these effects demonstrate that spatial information is encoded in the alpha band in combination with other task-relevant features, but this information degrades if spatial location is task-irrelevant.