The shift-effect in retinal ganglion cells of the rhesus monkey

McIlwain's periphery effect [7, 1, 5, 6] and the shift-effect of Krfiger and Fischer [2, 4, 6] are elicited by moving patterns, projected outside and at a long distance from the receptive field of retinal ganglion cells. The shift effects studied in the cat's retinal ganglion cells [4, 6] and lateral geniculate neurons [2] point to a visual information processing which is different from that performed by the classical receptive field mechanism and requires long range neuronal connections in the retina. So far, these effects were reported for cats only. This research report gives the first description of the shift effect in the monkey's retinal neurons. In the present experiments the monkeys faced a projection screen. Artificial pupils 5 mm in diameter and contact lenses checked by retinoscopy were used. During the recordings both eyes were open. Extracellular spikes from axons of retinal ganghon cells have been recorded in the optic tract using the same surgical preparation, anesthetics, muscle relaxation and optical stimulation as in cats (for detailed description see [4]). 79 optic tract fibres have been recorded from two adult rhesus monkeys and in 63 of these the receptive fields could be located and classified as 40 on-center and 23 off-center neurons: In all these 63 neurons we observed phasic discharges following grating shifts in the periphery. These responses were elicited by eae displacement of a grating well outside the receptive field. An example is shown in Fig. la. The receptive field of this on-center neuron was adequately illuminated by a steady bright spot of 1.8 deg. in diameter separated by a large dark surround from the moving grating as indicated at the right of Fig. la. (Stationary area 40 deg. in diameter; amplitude of displacement, 2 deg. ; velocity of displacement, 400 deg./see). Under these conditions the response latencies were about 50 ms. In most neurons an adequate steady visual stimulus was necessary to obtain the shift-effect, i.e7 bright spots of suitably chosen size against a dark surround for on-center neurons, dark spots against a bright surround for off-center neurons, provided a steady background activation on which the shift response appears. In a homogeneous field of brightness or darkness the shift-effect was almost absent or at least weak. Reversing brightness and darkness in the stationary field illumination ("inadequate illumination") abolished the shift-effect in all units tested and in two neurons we observed a weak inhibitory shift-effect. Thirteen neurons whose receptive fields could not be located and which therefore could not be activated showed no shift-effect under homogeneous illumination. This does not prove, however, that they had no shift-effect.