Eye movement training and recording in alert macaque monkeys: 1. Operant visual conditioning; 2. Magnetic search coil and head restraint surgical implantation; 3. Calibration and recording.

PURPOSE: To describe current methodology used to: a) train alert adult macaque monkeys to perform visual fixation tasks, b) implant binocular magnetic scleral search coils and attach a head restraint, and c) precisely record their fixation and pursuit eye movements.
METHODS: Animals are trained to sit in a primate chair and manipulate a lever. The animal turns on a laser spot (< 1 degree of arc) by pulling the lever. The spot dims after a variable, randomized period of time (luminance decrement 30-80%) and if the monkey releases the lever within 150-500 msec of dimming, a small bolus of juice is squirted via a servo into the monkey's mouth as a reward. The small size of the spot, the difficulty of detecting the dimming, and the short reaction time required for reward act in concert to assure attentive foveal fixation. After training, a search coil is implanted subconjunctivally in each eye and a polycarbonate head restraint device is attached to the skull. With the animal at the center of magnetic field coils, eye position is then calibrated precisely by requiring the animal to perform the dimming task at known positions of gaze. Fixation, vergence and pursuit eye movements are recorded by rewarding the animal for positioning the fovea of the eye within a small fixation 'window' encompassing the position of a stationary or moving target. A cover test of binocular eye alignment is performed by having the animal view through liquid crystal shutters that can be switched instantaneously from transparent to opaque. These methods have been used to train and to record eye movements in over 20 monkeys, and a representative sample of our experience in eight animals is reported in the Results.
RESULTS: Normal monkeys, and monkeys who had strabismus or amblyopia, required an average of 53 days to learn to fixate steadily for a minimum of 5 sec with a dimming-detection performance of at least 75% correct. Implanted coils provided good signals for durations ranging from 4 months to longer than 3 years (average greater than 1 year) before signs of breakage or extrusion. Head restraints under daily use have lasted an average of 11 months (range 5 months to greater than 2 years) before spontaneous detachment.
CONCLUSION: The training and surgical techniques described provide an efficient, straightforward method for recording precise binocular eye movements in awake monkeys.