Working memory and the attentional blink

After successful identification of a target item (T1) in a Rapid Serial Visual Presentation (RSVP) task, identification of a second target (T2) is typically impaired for approximately 500ms (the “attentional blink”, AB). Although several models of the AB exist, a common element is an attention demanding stage, where a durable representation of T2 in working memory (WM) fails to occur due to ongoing processing of T1. So what effect do ongoing working memory processes have on the AB? Whilst previous work (e.g. Akyurek & Hommel, to appear) has shown that concurrent WM tasks impair overall performance, no modulation of the AB effect has been found. This study investigates the effect of WM encoding processes on the AB using a hybrid WM/AB task. Subjects performed an RSVP task (T1-faces /T2-scenes, identity judgement) whilst maintaining two objects from a WM array; one was coloured red and the other green (e.g. a car in red, flower in green). The WM task involved matching a greyscale image (test image) to one of the items from the array. The test image was always one of the array items (e.g. the car). The item in the array to be matched against the test image was determined by a cue colour (red or green) that occurred either prior to the RSVP stream, or coincident with T1. For example, an array with a car in red and an apple in green followed by a green cue means the apple (green item) is to be matched to the test image. Results show that compared with presentation of cue colour outside the RSVP stream, concurrent presentation of the colour information with T1 increased the magnitude of the attentional blink. Follow up experiments show that this effect is not simply due to a low-level perceptual change, e.g. altered target contrast. These results inform the attention critical components of working memory processes.