SPATIAL FREQUENCY EFFECTS ON BINOCULAR INTERACTION ASSESSED WITH VISUAL EVOKED POTENTIALS *

We are conducting studies on binocular interaction to grating stimulation in the human visual evoked potentials (VEP) and this preliminary report describes the effects of changes in spatial frequency of the stimulus. Neurophysiological studies in animals have shown that ganglion cells in the retina,’.* and neurons in the lateral geniculate body3 and in the striate cortex4” exhibit selective sensitivity to spatial frequency. In animals with normal binocular functions, signals from the two eyes converge and activate the same cells in the striate cortex, provided the spatial characteristics of the stimuli to the individual eyes are similar.6 It is reasonable to assume that these conditions apply also to the human visual system. This would explain results of psychophysical experiments in humans, showing that the interaction between the responses from the two eyes is largest when they are stimulated with identical grating patterns, and that the interaction decreases with increasing difference in spatial frequency content of the two stimuli.’ These studies have indicated a selectivity of human binocular interaction of approximately two octaves around the conditioning spatial frequency. Electrophysiological evidence of these mechanisms in humans might be obtained in recordings of visual evoked potentials to grating stimulation. It has previously been demonstrated that binocular interaction might be evaluated in VEP to checkerboard stimulation and that subjects with poor binocular functions can be identified with this method.’ It was also shown that binocular VEP was dependent on the spatial content of the stimulus to the two eyes, much in the same way as indicated by the psychophysical experiments.’ However, VEP binocular interaction to the grating stimulation, used in the psychophysical experiments, has not been previously studied. The experiments were performed on two subjects with normal monocular and binocular vision. Vertical grating patterns with sinusoidal luminance profile and 50% contrast were generated on the screens of two identical oscilloscopes (Tektronix model 5440 with P31 phosphor). The mean luminance was 25 cd/m2. The spatial frequencies of the patterns on the two screens could be varied independently. With the aid of a prism in front of one eye, the images of the two screens could be fused, although the patterns stimulated the individual eyes separately. The two patterns reversed at slightly different temporal frequencies (1.9 and 2.1 reversals per sec), allowing transient VEP for each eye to be recorded separately from the other, even during binocular viewing.8 The signals were amplified and 128 sweeps averaged in the conventional manner. In both subjects the VEP to sinusoidal grating stimulation consisted of a