A Hamkari¹, D Melcher^1,2, X Liu¹; ¹Division of Science, Department of Psychology, New York University Abu Dhabi, ²Center for Brain & Health, NYUAD Research Institute, New York University Abu Dhabi

The sensory recruitment hypothesis posits that the same cortical areas used for visual perception are also used for maintaining visual working memory (VWM). Support for this idea comes from evidence that information stored in VWM can influence perception. To test whether there is competition for sensory processing resources between perception and VWM, we measured the sensitivity (d') of detecting a near-threshold flash during the delay period of an oculomotor delayed response task. We compared detection performance for participants (N=30) under three conditions: (1) an oculomotor delayed matched-to-sample task, in which participants prepared to make a saccade to a remembered target location, (2) an oculomotor delayed non-matched-to-sample task, in which they made a saccade to a new target that was not part of the memory set, and (3) a detection-only condition in which participants ignored the VWM display and only focused on detecting the flash. The visual input was matched across conditions up to the point of the oculomotor response. Based on previous studies, we expected that the matched-to-sample task would only require maintenance of a motor plan, while the non-matched-to-sample version would require maintenance of a sensory code. In a separate session, participants also repeated a simplified version of the study in the MRI. We found that detection varied significantly between the three conditions. Participants had a significantly lower sensitivity (d') for the nonmatched condition, when a sensory code was maintained, compared to the matched condition where a motor code is maintained. Highest sensitivity was found when the VWM stimuli were ignored (condition 3). This pattern of results suggest that maintaining a sensory code in VWM influences basic visual processing, potentially due to the recruitment of sensory processing areas in the active maintenance of visual memory representations.

Acknowledgements: This work was supported by the NYUAD Center for Brain and Health, funded by Tamkeen under NYU Abu Dhabi Research Institute grant CG012. This research was partially conducted using the Brain Imaging Core technology platforms resources at NYUAD.