The Effect of Spatial Attention and Laterality on Information Processing

Abstract

Five experiments on the influence of spatial attention and laterality on information-processing tasks are reported. In Experiments 1 to 4, the task was memory scanning (Sternberg, 1966, 1969a), and in Experiment 5 it was a task in which letters were classified according to pronunciation (Mondor & Bryden, 1992). Spatial attention was controlled by central arrow cues. In all the experiments in which attention was manipulated overall RT was shorter for attended than for unattended locations (Experiments 1, 4, and 5). In the memory-scanning experiments (Experiment 1 and 4) the function relating RT to memory-set size for stimuli in attended locations had a significantly smaller zero-intercept than the function for unattended locations, but the slope was unchanged, implying that attention did not influence memory scanning, but had its influence at other stages of processing. In Experiment 4 stimuli were presented in both a clear and a visually degraded form. Mean RT was shorter for clear than for degraded stimuli. Visual quality interacted with attention which suggests that they influence the same stage of processing. This was concluded to be the early visual encoding of the stimulus, since visual quality presumably affects only this stage. This interaction was replicated in Experiment 5, the letter-classification task. Displaying the probe stimuli in different visual fields did not change the slope of the set-size function in Experiments 1, 2, 3 and4. In Experiment 1 there was an overall LVF advantage. In Experiment 2, which was the same as Experiment 1 except that attention was not directed, there were no significant visual-field effects. Even with bilateral presentation of stimuli (Experiment 3) there were no visual-field effects. Spatial attention did not influence visual-field differences in performance in any of the experiments. In Experiment 5 visual quality interacted with visual field, reflecting the LVF advantage for degraded stimuli and RVF advantage for clear stimuli. The study was based on Sternberg's (1969b) four-stage model of memory scanning and the additive-factors method, but such a model cannot account for all of the results and a five-stage model is proposed.