The superior colliculus (SC), along with its nonmammalian counterpart, the optic tectum, is well known for playing a prominent role in mediating orienting behaviors to exogenous stimuli. A partial list of species for which this capacity of the SC has been examined includes the goldfish, frog, owl, rat, cat, and monkey. In nonhuman primates, study of the oculomotor function of the SC, for all intents and purposes, began in the early 1970s with the pioneering works of David A. Robinson, Robert Wurtz, Peter Schiller, David Sparks and their numerous collaborators and colleagues. Following the lead of Edward Evarts, who combined operant training methods with

0-8493-0097-5/04/$0.00+$l .50 © 2004 by CRC Press LLC

single neuron electrophysiology to study the neural representation of limb movement in monkey motor cortex1 these investigators began to study the neural basis of visual orienting in alert, behaviorally trained monkeys. Monkeys were a natural choice as a model of visuomotor behavior owing to an extensive oculomotor range, vision as a dominant modality, and the ease with which they could be trained on relatively complex visuomotor tasks. To date, the vast majority of studies of goal-directed orienting in nonhuman primates have focused on saccadic eye movements, rapid changes in eye position designed to place the images of interesting objects on the fovea for detailed viewing. As such, most of what is known about the primate SC concerns saccade-related function, though more recent lines of evidence point toward a more general role in

shifting gaze (combined movement of the eyes and head),2‚3 posit involvement in coding for other types of eye movements (e.g., smooth pursuit),4‚5 and even suggest a role in producing arm movements.6-8

The SC is, arguably, the best understood of all of the so-called oculomotor regions. Along with a wealth of knowledge of its anatomy, progress in understanding the physiology of the SC has benefited greatly from the power of converging operations wherein results deriving from single-unit electrophysiology, electrical microstimulation, and pharmacological manipulation have provided complementary views of the SC’s oculomotor machinery (see References 9 to 13 for detailed discussion). The strength of single-unit recording lies in its spatial and temporal resolution. Clearly, when measuring extracellular action potentials, the single neuron represents the irreducible spatial quantity; equally important is the ability to monitor neural activity with millisecond temporal resolution. This is critical when attempting to relate activity to the occurrence of saccades, which quite often begin and end within the span of 50 milliseconds. Despite these strengths, single neuron recording has some limitations. First, the information represented in neural ensembles can be difficult (and sometimes impossible) to infer from consideration of serially recorded single neurons. Second, the data derived from single neuron electrophysiology is correlative, and while the correlation between a particular neuron’s activity and an experimental event (e.g., saccadic eye movement) may be compelling, it falls short of establishing causality.