Poster A6, Saturday, March 25, 5:00 – 7:00 pm, Pacific Concourse
Predicting attentional failures: the spatiotemporal neural dynamics of attention during sustained dual-task performance.
James Elliott1, Barry Giesbrecht1; 1University of California, Santa Barbara
Failures of attention are common both in everyday activities like listening to a lecture and in potentially life-threatening scenarios such as flying a jet. To investigate the temporal and spatial characteristics of neural activity that predicts failures of attention, 14 participants performed two versions of a continuous temporal expectancy task while both EEG and fMRI were recorded. In both versions participants monitored a stream of flickering faces and cars (15Hz) for 3.5 minutes. Standard images were presented for 800 ms while targets, requiring a face vs. car discrimination, were presented for 1100 ms.. Half of the auditory stimuli presented coincided with the visual target. During single task blocks participants only responded to the visual targets. Dual task blocks required responses to both auditory and visual targets. Single task visual detection (M=.51) was better than dual task detection (M=.39, F(1,12)=10.8, p < .01). There was a decline in performance as a function of time within block (F(8,96)=44.9, p< .001). Pre- and post-target classification of EEG alpha activity (8-14 Hz.) in occipital and parietal electrodes was significantly better than permutation tests in both conditions. Pre-target classification of fMRI BOLD activity was significant in error monitoring and whole cortex regions, but not in perceptual or default mode regions. Pre-target whole cortex classification was significantly better than an auditory vs. visual stimuli control classification (F(1,13) = 10.4, p <.001), confirming that the classification reflected pre-target activity. Overall, these results suggest that both fMRI and EEG can be used to predict subsequent performance.
Topic Area: ATTENTION: Nonspatial