Abstract
Just as we can use attention to select relevant information in our environment, we can also selectively attend to information held in working memory. Here we consider whether shifts in attention differ between perception and memory. When shifting attention in visual perception, the duration of shift depends on distance (Tsal, 1983), consistent with the movement of an attentional “spotlight” (Posner, Snyder & Davidson, 1980). When attention moves within a representation held in the mind, will similar distance effects emerge? We had participants memorize the identity and location of 4 objects on a 4×4 grid. During memory maintenance, participants were instructed to update the spatial position of one of the objects, before being instructed to update a second. The updating process was self-paced, with the time between each step indicating the time required to shift to and access the relevant item within memory. Importantly, we manipulated displays such that the second target was either close to or far from, and either shared one or no features with the first target. Updating the second target was faster when the targets shared a feature, F(1,17) = 14.06, p = .002, but the distance between targets did not matter, F(1,17) = 1.33, p = .265. While participants showed clear evidence of remembering location, location had ceased to affect updating times, unlike featural overlap. In contrast, in a visual search task using the same displays, participants were faster at finding the second target when it shared a feature with, F(1,17) = 34.45, p < .000, and when it was closer to, F(1,17) = 39.28, p < .001, the first target. This suggests that accessing a new position in memory does not require traversing spatiotopic distance as appears to happen during visual attention and raises questions about how spatial information changes between perception and memory.