There has been much recent debate within the field of working memory (WM) on the storage mechanism (persistent delay-period activity vs. activity-silent) and storage location (frontal vs. sensory cortex) of latent WM items. This thesis addresses these debates using a novel combination of transcranial magnetic stimulation (TMS) and electroencephalography (EEG) with in a double retrocue working memory experiment. We found that participants had higher memory performance for unattended items when those items received TMS during a delay period as opposed to when those items were not directly targeted by TMS. This increase in behavioral performance was mirrored by a recovery of decoding ability for the item's feature (i.e., orientation) as measured with the application of inverted encoding models to the EEG data. Specifically, unattended items that received TMS to contralateral, retinotopic visual cortex showed significant reconstruction of orientation whereas those items that received TMS to ipsilateral visual cortex did not. While concerns about data quality and TMS protocol (particularly for Experiment 2) limit the conclusions that can be drawn at present, the results from this thesis converge with other recent studies to provide support for the sensorimotor recruitment hypothesis and "activity-silent" retention mechanisms in WM.