Abstract
The mixed-category advantage in visual working memory refers to improved memory for an image in a display that includes four images taken from two different categories relative to the same number of images taken from a single category (Cohen et al., 2014). Jiang et al. (2016) found that this advantage characterizes mainly faces and suggested that it is a category-specific effect, due to within-category interference related to the visual characteristics of faces (e.g., their shared spatial configuration). Faces, however, possess evolutionary and social significance that may bias attention toward them, potentially reducing the overall processing resources directed toward their counterpart category in a mixed display. Consequently, any category paired with faces may suffer from little/no mixed-category advantage, or even an inversed effect (i.e., a disadvantage when appearing in a mixed-category, relative to a single-category display). Here, using a change-detection task, we showed that a category (highways) that demonstrated a mixed-category disadvantage when paired with faces, demonstrated a mixed-category advantage when paired with a different, non-facial category (clocks). Furthermore, manipulating the relative significance of a category in mixed-category display trials (by increasing/decreasing the probability of a memory test on that category) altered the category’s advantaged/disadvantaged status, suggesting that attention may play an important role in the mixed-category effect. Taken together, the asymmetric mixed-category pattern appears to rely on an attentional bias, rather than on a fixed, category-dependent mechanism. These results suggest that domain-general, not domain-specific factors underlie the asymmetric pattern of the mixed-category advantage in working memory.
Acknowledgement: This research was funded by the Israel Science Foundation (ISF) #1622/15 to NG.