Academic journal article Memory & Cognition

Memory Search for the First Target Modulates the Magnitude of the Attentional Blink

Academic journal article Memory & Cognition

Memory Search for the First Target Modulates the Magnitude of the Attentional Blink

Article excerpt

Published online: 25 June 2014

(Q> Psychonomic Society, Inc. 2014

Abstract The resolution of temporal attention is limited in a manner that makes it difficult to identify two targets in short succession. This limitation produces the phenomenon known as the attentional blink (AB), in which processing of a first target (T1) impairs identification of a second target (T2). In the AB literature, there is broad agreement that increasing the time it takes to process T1 leads to a larger AB. One might, therefore, predict that increasing the number of possible T1 identities, or target set, from 1 to 16 would lead to a larger AB. We were surprised to find that this manipulation of T1 difficulty had no influence on AB magnitude. In subsequent experiments, we found that AB magnitude interacts with T1 processing time only under certain circumstances. Specifically, when the T1 task was either well masked or had to be completed online, we found a reliable interaction between AB magnitude and the target set size. When neither of these conditions was fulfilled, there was no interaction between target set size and the AB. Previous research found that when the target set changes from trial to trial, trials with more possible targets elicited a larger AB. In the present study, the target set is held constant, reducing the demands on working memory. Nevertheless, AB magnitude still interacts with target set size, as long as the T1 task cannot be processed offline. Thus, the act of searching memory delays subsequent processing, even when the role of working memory has been minimized.

Keywords Attention * Working memory * Memory


The limits of visual attention have been central topics in cognitive psychology over the past 30 years. However, while most undergraduate psychology students know that people can only process a finite set of items or locations in a scene at any one moment, temporal limitations on processing are not as well understood. The attentional blink (AB) has been an important paradigm in understanding the limits of visual at- tention in a temporal domain (Dux & Marois, 2009; Shapiro, 1994). The AB refers to the impaired processing of the second of two targets (T1 and T2) within a stream of items. Single items are presented in quick succession centrally in a rapid serial visual presentation (RSVP). Typically, processing of T2 is impaired when it appears within 200 and 500 ms of T1 (Raymond, Shapiro, & Amell, 1992). One common measure of blink magnitude is percent correct of T2 given that T1 was correcT1y identified. Larger deficits are evidence of a larger "blink."

The current literature supports the notion that processing of T1 occupies some aspect of a limited-capacity central re- source, thereby preventing the subsequent processing of T2. This botT1eneck theory of the AB posits two stages of process- ing. During the first stage, items are flagged on the basis of how well they fit the target definition. Once an item is selected as a potential target, it moves to the second stage, where it receives additional processing. ImportanT1y, this second stage is thought to be capacity limited and time consuming. There- fore, while T1 is occupying these resources, processing of T2 is delayed, thereby making it vulnerable to masking by sub- sequent items in the stream (Chun & Potter, 1995).

As a result of this series of events, the difficulty in correcT1y identifying T1 plays a central role in determining the magni- tude of the AB (Bowman & Wyble, 2007; Dux & Marois, 2009; Olivers, van der Stigchel, & Hulleman, 2007). The central interference theory (Jolicoeur & Dell'Acqua, 1998) predicts that increasing the difficulty of T1 should lead to a larger AB. Chun and Potter (1995) found that increasing T1- distractor similarity led to a larger AB, and looking across several early AB studies, Sieffert and Di Lollo (1997) found a negative correlation between AB magnitude and mean per- centage correct on T1. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.