Examination of goal-directed aiming tasks has revealed that rapid, discrete human action is amendable to online control. This control affords humans a margin of error in movement planning and execution as well as a means of acquiring their goals when the body and/or the environment are extrinsically perturbed. For over a century, the models of online movement control that have best described the trajectories and outcomes of goal-directed reaches hold that these movements are composed of two distinct components. The first component moves the limb from its resting position towards the target. The second component is a corrective movement that is formed on the basis of a visual referencing of the moving limb and target positions. As such, the temporal and spatial characteristics of these discrete movement changes have been attributed to the limits of visual information processing. Furthermore, the absence of any discrete movement changes in the portions of movements outside of the temporal and spatial limits of vision led many investigators to conclude that first component impulses are ballistic and uncontrollable. However, recent studies involving environmental perturbation and within-subject trial-to-trial spatial variability analyses have evidenced that initial impulses are privy to online control. Because the corrections made early in movement impulses occur quicker than purely afferent visual information can be processed this form of control has been attributed to the use of forward anticipatory processes. The four studies presented here examine the nature of initial impulse control through kinematic analyses of reaches made to targets against various combinations of limb, target, and environment perturbations. This was done in order to evaluate anticipatory control's relationship with visually-regulated control and the relative influence the two processes have on the movement trajectory and performance outcome. The first study examined target-directed reaches made against illusory moving background and target relocation perturbations. The results showed the presence of early anticipatory and late visually-regulated control. Non-interactive main effects of the two perturbations on outcome accuracy revealed that the processes operate independently. The second study tested the applicability of an air discharging stylus as a tool for perturbing reach velocities. The results showed that the stylus effectively perturbed limb velocity and highlighted the presence of a limb forwarding response to either an advancing or hindering perturbation. The findings evidence the non-specific nature of anticipatory control responses. The third study examined reaches made against combinations of actual limb velocity and target position perturbations. The interactive effect of the two perturbations on reach trajectories and outcome accuracy indicated that the perturbations were salient enough to prompt parallel operation of the two control processes. Again, the control of initial movement portions was highlighted by non-specific responses to the perturbations. The fourth study examined reaches made against combinations of illusory and actual perturbations to both the anticipatory and visually-regulated control processes. Interestingly, performers withheld responses to the illusory perturbation unless they were also responding to an actual perturbation. This finding suggests that anticipatory control responses are biased during movement preparation. Furthermore, combined illusory and actual perturbations to target position had interactive effects on visually regulated control. Overall, the studies evidence that target-directed movements are mediated by two modes of control. There is an anticipatory mode of control that operates continuously and, given that reaches are made within the spatial and temporal limits of visual processing, there is also a feedback driven discrete mode of control that overlaps with the continuous mode. / Thesis / Doctor of Philosophy (PhD)