INDIVIDUAL-DIFFERENCES IN SELF-REPORTED COGNITIVE FAILURES - THE ATTENTION HYPOTHESIS REVISITED

It unclear which attention functions are related to self reported cogn itive failures [measured with the CFQ (Broadbent, Cooper, FitzGerald a nd Parks, 1982)]. Experiment 1 showed no significant correlations betw een CFQ and two perceptual speed tests. In Experiment 2 we identified two attention/control factors. Shifting between action-schemas was not significantly correlated with CFQ. However, frequent cognitive failur es were associated with slow performance on focused attention tasks (r = 0.61). In Experiment 3 subjects named tachistoscopically presented letters, appearing in one of eight locations along a circular display. In 75% of the trials targets appeared in one of two pre-cued location s. When the cues were adjacent and the Stimulus-Onset-Asynchrony was l ong (120 msec) subjects zoomed covert visual attention on the cued loc ations. Report of frequent cognitive failures was significantly associ ated with greater zooming (r = 0.45). Nevertheless, zooming led more t o costs than to benefits. The data are discussed in terms of Norman an d Shallice's (1986) model of attention control.