The "working" component of working memory predicts AB magnitude

Abstract

In rapid serial visual presentation (RSVP) second target (T2) accuracy decreases with temporal proximity to the first target (T1). This phenomenon is known as the attentional blink (AB). Colzato et al. (2007) used an individual differences approach to examine whether individual AB magnitude was predicted by working memory (WM) operation span, using the Operation Word Span paradigm (OSPAN), and general cognitive ability, measured with Raven's Standard Progressive Matrices (SPM). They found that OSPAN score inversely correlated with AB magnitude even with Raven's SPM partialled out. However, it is not clear from this study whether active “working” memory ability would predict AB magnitude better than a less dynamic measure of short-term memory capacity. Using a continuum of WM measures that reflect varying degrees of the active “working” process could better define the source of the relationship. The digit-forward task is a simple rote memory task that reflects static STM capacity, requiring individuals to repeat a string of digits in order of presentation. The digit-backward task requires an increasing degree of the active “working” component by having individuals repeat the string of digits in the opposite order of presentation. Our study used these measures plus a reliable measure of individual visual memory capacity (k) in addition to the OSPAN; as well as Raven's SPM and a reading comprehension task to look at as predictors of individual AB magnitude. Dramatically, the OSPAN still inversely correlates with AB magnitude with Raven's SPM, reading comprehension, and digit- forward and backward partialled out. This is strong support that a “working”, executive component of WM predicts temporal limitations of selective attention beyond static STM capacity and general cognitive ability.