Abstract

Influential models of visuo-spatial attention (Rizzolatti et al., 1987; Rizzolatti et al., 1994; Schneider, 1995) postulate that the programming of goal-directed motor actions is inevitably accompanied by a spatially congruent shift of attention toward the target of the upcoming movement (premotor shift of attention). This presumed obligatory attentional selection of motor targets has been attributed to motor programming itself, either as a by-product of motor programming (Rizzolatti et al., 1987; Rizzolatti et al., 1994) or as a prerequisite for motor programming to enable the extraction of the target's spatial information (Allport, 1987; Neumann, 1987; Schneider, 1995). Although the linkage of spatial attention and motor programming is well established in different types of goal-directed motor actions, there is mixed evidence regarding the proposed obligatory nature of the attention-action coupling. This raises the possibility that the premotor attention shift phenomenon is mediated by mechanisms unrelated to motor programming. In this dissertation, two experimental studies are presented that both investigated whether attentional selection of the target of an upcoming motor action can also be attributed to a habitual expectation that motor target locations contain visual information of high behavioral relevance. To this end, participants in both studies performed a psychophysical dual-task in which they had to identify a task-relevant visual stimulus (an attention probe) while concurrently preparing a pointing movement (first study) or saccadic eye movement (second study) toward a motor target. In a training phase, the position of the attention probe was manipulated so that participants learned to expect either spatial congruence or incongruence between the attention probe and the motor target. In a subsequent test phase with randomized attention probe position, possible training effects on attention allocation were assessed. Overall, both studies yielded similar results. Spatial attention allocation was markedly biased toward the expected position of the attention probe, regardless of whether this position matched the motor target or not. The observation that attentional resources can be shifted to a position other than the target of an upcoming motor action indicates that the link between spatial attention and motor programming is more flexible than generally assumed. It further suggests a crucial role of habitual top-down modulations of spatial attention in the generation of premotor attention shifts.