Abstract
When a visual target (T1) is masked it can produce an interference on the identification of a second masked target (T2) for at least half a second (the attentional blink, AB). The exact role of target masking in the attentional blink is still unclear. Previous evidence suggests that the AB is enhanced when T2 is masked by interruption and reduced when T2 and its mask have a common onset. Four studies contrasted the effects on the attentional blink of masks which temporally overlap targets (common-onset masking) and those that follow the targets (delayed-onset masking). Study 1 showed that common-onset masking can be obtained with centrally presented stimuli when the mask is spatially close to the target and lasts longer than the target. Study 2 showed that the AB was much reduced when a common-onset mask was used at T2 compared to a delayed-onset mask. Study 3 showed that the same effect can be obtained when a common-onset mask is used at T1, i.e, an AB was produced when a delayed-onset mask is used at T2 but a much reduced AB with common-onset masking of T2. Study 4 showed that the delayed-onset mask must be a new visual event to produce an AB. These results show the importance of mask timing on the AB and suggest that the attentional blink depends critically on automatic orienting of attention to a new visual event.