Attentional Blink Effect

Mental Processing Limitations

How much information the brain can handle at once has been a major question within the science of the mind. In 1980, John Duncan argued that the answer depends on the nature of the information. In the visual domain, information hitting our retina is first processed in parallel by the brain. The important distinction is between information that is relevant to an [Page 75]observer (information that needs to be remembered or acted on, referred to as target information) and information that is irrelevant (referred to as distractor information). In Duncan's view, targets are thought to be passed on from the initial parallel stage to a second, limited-capacity processing stage that enables awareness of the object. This stage works in a serial fashion, processing objects one by one. Distractors, however, are rejected within the first stage and do not make it to the second stage. As a consequence, the brain has problems when there are multiple relevant objects, since the second level can only deal comfortably with one object at a time. Some target information is therefore likely to be lost. It is not surprising that, when about a decade after Duncan scientists discovered the attentional blink, they soon explained the phenomenon in terms of limited-capacity resources. Specifically, these theories have stated that the attentional blink occurs because T1 occupies the second stage for up to 500 milliseconds. Indeed, there is evidence that the attentional blink varies with T1 processing difficulty, for instance, as manipulated by masking (the disruption of a stimulus by either an immediately preceding or a following stimulus), inter-item similarity, or cognitive load.

Limited Capacity or Selection?

The attentional blink is a profound and robust effect, but it is not universal. Several conditions have been found under which multiple targets can be processed within a short period of time. One important case is lag-1 sparing: When T2 is presented at the very first temporal position following T1, detection is often unimpaired. This is somewhat difficult to explain from standard two-stage theory, and several additional hypotheses have been developed (many assuming a combined episodic memory representation of the two targets). Performance on T2 also improves substantially when the distractors between T1 and T2 are replaced with a blank. Most remarkable is that the attentional blink virtually disappears when the intervening distractors are replaced by even more targets. The second stage should now be completely overwhelmed, yet performance improves. This has made scientists propose alternatives to two-stage limited-capacity theory. According to selection theories, the cause of the attentional blink lies not within the limited-capacity stage, but within the filtering or gating mechanism that performs the transition from the first to the second stage. The fact that performance on T2 is fine when there are no intervening distractors indicates a crucial role for those distractors. Selection theories, therefore, assume that distractors lead to active suppression (i.e., disruption or inhibition) of the perceptual input. The attentional blink is the direct consequence of this suppression. This means that as long as no distractor information is encountered, multiple targets can be processed at the same time. Thus, selection theory assumes a more important role for distractor rejection than originally conceived by Duncan and a larger second stage capacity than proposed by most attentional blink theories.

...