Observers adopt attentional control settings (ACSs) based on their current attentional goals that specify which external stimuli should capture attention. The capture of visual-spatial attention is contingent (i.e., dependent) on the current goal: When searching for a friend wearing a blue shirt, blue stimuli will capture attention, and red stimuli will not. While ACSs play a crucial role in determining which stimuli should capture attention, our understanding of them is still quite limited. The goal of the present thesis was to expand our understanding of ACSs by investigating the memory systems that maintain ACS representations, the internal representations of searched-for external stimuli (e.g., “blue”). To accomplish this goal, all experiments in this thesis used an attention cueing task where participants searched for one or more target stimuli, thereby inducing the ACS representation(s). Just before the target appeared, a cue appeared, which should be ignored. Importantly, if the cue is a searched-for stimulus, it will capture attention, otherwise it will not (Folk et al., 1992). Chapter 1 investigated whether visual working memory (VWM) can maintain one or more ACS representations. Results indicated that VWM can maintain an ACS representation of one searched-for stimulus. Chapter 2 investigated whether external stimuli that match ACS representations in long-term memory (LTM) truly capture visual-spatial attention. Electroencephalography results indicate that this is the case, as searched-for stimuli elicited shifts in attention within 170 ms. Chapter 3 investigated whether observers can simultaneously maintain a VWM ACS representation and multiple LTM ACS representations. Results suggest that this is indeed possible. Thus, the results of the present thesis indicate that multiple memory systems can maintain ACS representations. Additionally, these findings have implications for current models of attentional capture, which are described in the General Discussion. This thesis enhances our understanding of ACSs and the memory systems that support ACS representations, thereby expanding our understanding of how we successfully navigate our complex visual environments. This work was supported by the National Sciences and Engineering Research Council of Canada under Discovery Grant number 418507-201 and Canadian Foundation for Innovation under Grant number 30374 to Dr. Naseem Al-Aidroos.