This is an excerpt from Vision and Goal-Directed Movement by Digby Elliott & Michael Khan.
Normal functioning in everyday life requires humans to interact with moving and static objects within a visually dynamic environment. Moving around inside and outside the home, such as when preparing food in the kitchen or climbing a stairway, requires us to pick up, intercept, or avoid surrounding objects. Failure to perceive the changing relationship between our body and an object of interest could lead to ineffective behavior that has the potential for serious consequences. For example, misperceiving the speed of an approaching car while waiting to cross a road could possibly result in collision. It is perhaps not surprising, therefore, that the human visual system has evolved highly specific mechanisms for processing motion that are capable of extracting very precise information about an object's position, direction, and speed as it moves across the retina (Anderson & Burr, 1985; McKee, 1981).
Although the human visual system can provide accurate object-related information from retinal input alone (i.e., when the eyes are fixated), when an object of interest moves relative to the retina, either because the self or the object is moving, our normal response is to move the eyes and head in an attempt to maintain the object image on the fovea. This can be achieved using image-stabilizing eye movements of the vestibulo-ocular reflex (VOR) and optokinetic reflex (OKR) or gaze-orienting eye movements such as saccades, smooth pursuit, and vergence. As well as keeping the object of interest in the region of high acuity to discriminate object characteristics such as size and shape, motor signals sent to move the eyes (i.e., efference copy) provide extraretinal information. This information is critical in interpreting retinal stimulation generated as the eyes pursue an object moving against a cluttered background (i.e., did the object move or did the observer move?).
Extraretinal information not only provides a stable perception of the world around us but also provides advanced information related to upcoming object motion. This information allows the observer to exhibit predictive eye movements that are not driven reflexively by online visual feedback. These predictive eye movements help the observer to overcome the delays involved in processing retinal feedback that would otherwise limit the ability to pursue a fast-moving object. In addition, predictive eye movements allow the observer to initiate and perpetuate the pursuit response when tracking an object in the absence of visual feedback. This ability is particularly important in situations where an object undergoes transient occlusion that does not permit integration of consecutive visual samples and the formation of a continuous percept of the object trajectory. Such situations are commonplace in everyday life (e.g., when the goalkeeper's view of the incoming free kick is occluded by the wall of defending players). Until recently, little was known about how the eyes move when visual feedback of a moving object is occluded and, therefore, about the underlying control mechanisms.
In this chapter, we describe the gaze-orienting eye movements that are used to keep the object image on the fovea. We then turn to the use of these eye movements in tracking smooth object motion as opposed to step changes in object position that elicit saccadic eye movements (for a discussion of saccadic eye movements in reaching and grasping, see chapters 10 and 11) or changes of object position in depth that require vergence eye movements. The main findings are from experiments that have examined the ocular response in situations where there is a temporary absence of visual feedback of the moving object. This should create problems for a control system that, according to traditional dictum, is under reflexive control driven by retinal input. Particular focus is given to work on ocular pursuit in which visual feedback from the moving object is temporarily unavailable, such as if the object undergoes transient occlusion or extinction. Using the findings in our laboratory as well as those of others, we present a model of ocular pursuit, including a novel arrangement of extraretinal input that accounts for both reflexive and predictive control of the eyes. Finally, we refer to work that has revealed the underlying neural substrate and then discuss potential directions for future research.