1972 Treisman and Davies 1973), mutual inhibition within the global workspace may depend on similarity between competing events. In line with well-known effects of task similarity in divided attention ( Allport et al. These results suggest that, if behavior requires access to a global neuronal workspace, this workspace must be less limited than usually supposed. 2010 Soto-Faraco and Spence 2002 Van der Burg et al. With unspeeded responses and similar T1/T2 tasks, however, many studies show the blink to be greatly reduced or eliminated with targets in different modalties ( Arnell and Jenkins 2004 Arnell and Larson 2002 Duncan et al. 2002), especially if T1 requires a speeded response (e.g., Jolicoeur 1999a Jolicoeur 1999b), or if tasks for the two targets are very different, calling for a large change of mental set between T1 and T2 (e.g., Arnell and Jolicoeur 1999 Dell'Acqua et al. Sometimes, in accord with the global workspace view, a blink occurs between targets in different modalities (e.g., Soto-Faraco et al. In this regard, an intriguing puzzle concerns targets in different sensory modalities.
On this view, all targets should show mutual interference, irrespective of their nature or origin. 2013 Watanabe and Funahashi 2014) data linking diverse aspects of restricted attentional capacity to competitive frontoparietal activity. This proposal is consistent with fMRI (e.g., Marois and Ivanoff 2005) and single-unit ( Kadohisa et al. 2014), explaining widespread limitations on divided attention and concurrent awareness including the attentional blink.
Mutual inhibition within this network severely restricts the number of simultaneous events that can be processed ( Sergent and Dehaene 2004) or operations that can be carried out ( Raffone et al. An influential theory proposes that this frontoparietal network constitutes a global workspace that broadcasts information widely between different brain systems and enables consciousness ( Dehaene et al. 2005), matching extensive frontoparietal activity for perceived targets of many kinds revealed by functional magnetic resonance imaging (fMRI) ( Beck et al. This N2-P3 complex likely arises in widespread neural generators, including multiple frontal and parietal sources ( Bledowski et al. The major difference occurs in a late N2-P3 complex, severely attenuated or eliminated in the unseen case ( Sergent et al. In vision, the neural basis for the attentional blink has been studied using event-related potentials (ERPs), comparing trials where T2 is correctly detected or identified (“seen” trials) to trials where T2 is missed (“unseen”). This phenomenon, called the attentional blink, is considered a key property of attentional resource limitation ( Chun and Potter 1995 Raymond et al. When targets occur very close in time, performance on T2 is impaired, but recovers gradually as temporal separation increases. Such limits are effectively tested by presenting two target events, T1 and T2, at varying time intervals and recording performance in target detection or identification. The results suggest parallel brain responses to target events in different sensory modalities.Īn enduring question is the neural basis for limits in divided attention, manifest in impaired performance when processing simultaneous or closely successive events ( Broadbent 1958 Kahneman 1973 Marois and Ivanoff 2005 Pashler 1994). In an attentional blink EEG/MEG design, however, an auditory T1 and visual T2 are identified without mutual interference, with largely preserved electromagnetic responses to T2.
Using functional magnetic resonance imaging, we confirm that visual and auditory targets produce similar, distributed patterns of frontoparietal activity. Often, however, the attentional blink is reduced or eliminated for targets in different sensory modalities, suggesting a limit to such global inhibition. An influential theory proposes that conscious perception requires access to a distributed, frontoparietal global workspace, explaining the attentional blink by strong mutual inhibition between concurrent workspace representations. Concurrently, the brain's electromagnetic response to the second target is suppressed, especially a late negative-positive EEG complex including the traditional P3 wave. In the attentional blink, a target event (T1) strongly interferes with perception of a second target (T2) presented within a few hundred milliseconds.