Visual attention is involved in many everyday tasks, such as finding one’s shoes, driving, or looking for a face in a crowd. Often attention must be allocated or split onto more than one location. Yet how does the visual system accomplish this, and what cost does it have in terms of performance in visual tasks? This dissertation reports the results of an investigation of the claim that attention could be allocated to two or more non-contiguous locations simultaneously. This study uses a technique, which to the best of my knowledge, has not been used by other investigators studying the splitting of attention. The stimulus used in the current thesis consisted of four possible cue locations, with each potential location located in 45 degree increments (45, 135, 225, 315 degrees) at the same eccentricity. Experiment 1 replicated the results of previous attention studies, but using the new cueing paradigm. The results showed that the spotlight of attention could change location when multiple potential locations are present. The advantage of the new paradigm is the possibility of cueing different locations within and between visual hemifields. This was explored in Experiments 2 and 3, to examine the role of hemispheric asymmetries in the debate over multifocal attention. Specifically, two locations were cued in Experiment 2, and Experiment 3 examined the differences of splitting the attentional beam within and across hemifields. Finally, Experiment 4 investigated the effect of cueing all four locations at the same time. The results from the current thesis confirm that observers can allocate attention to two or more simultaneous locations, without observing any attentional enhancement coming from the in-between cued locations. However, these results stemmed from attending to two locations across hemifields. The present thesis provides more evidence in support of the ability to split attention, and describes the impact that multifocal attention plays in visual perception.