Refreshing refers to the use of attention to reactivate items in working memory (WM). In the present study, we aimed to test the hypothesis that refreshing is closely related to memory search. The assumption is that refreshing and memory search both rely on a basic covert memory process that quickly retrieves the memory items into the focus of attention, thereby reactivating the information (Cowan, 1992; Vergauwe & Cowan, 2014). Consistent with the idea that people use their attention to prevent loss from WM, previous research has shown that increasing the proportion of time during which attention is occupied by concurrent processing, thereby preventing refreshing, results in poorer recall performance in complex span tasks (Barrouillet, Portrat, & Camos, Psychological Review, 118, 175-192, 2011). Here, we tested whether recall performance is differentially affected by prolonged attentional capture caused by memory search. If memory search and refreshing both rely on retrieval from WM, then prolonged attentional capture caused by memory search should not lead to forgetting, because memory items are assumed to be reactivated during memory search, in the same way that they would be if that period of time were used for refreshing. Consistent with this idea, prolonged attentional capture had a disruptive effect when it was caused by the need to retrieve knowledge from long-term memory, but not when it was caused by the need to search through the content of WM. The present results support the idea that refreshing operates through a process of retrieval of information into the focus of attention.