Ample research has suggested that visual attention is biased towards threat and it was argued that this bias is an essential component of survival and implicated in anxiety. However, it is less clear how this bias is translated into memory, and specifically into the memory of items presented near a threatening stimulus. Here, we investigated this issue by testing how well people remember neutral and threatening images presented under various task demands. On each trial, observers saw two images before performing a dot-probe task (Experiment 1), a colour discrimination task (Experiment 2), a global or local attention task (Experiment 3), or no task at all (Experiment 4). A recognition memory test was performed at the end of each experiment to assess how the presence of a threatening image influences the memory of both images presented in the display. In all experiments, overall memory was enhanced as more threatening images were presented in the display. However, this enhancement did not occur at the expense of the processing of the surroundings. That is, with the exception of the dot-probe task, memory performance was not affected by an adjacent threatening image. Together, these findings challenge trade-off accounts, which predict that the processing of a threatening stimulus should take place at the expense of the processing of nearby items. Instead, these findings suggest that any effect of threat on the visual processing of the display is short-lived and more limited than previously thought.
Multimodal Integration and Expertise