INHIBITION IN ON-OFF DIRECTIONALLY SELECTIVE GANGLION-CELLS OF THE RABBIT RETINA

1. We have investigated the inhibitory mechanisms modulating the extra cellularly recorded responses of ON-OFF directionally selective (DS) g anglion cells of the rabbit retina. Our investigations used both movin g spots and apparent motion. The latter was produced by both prolonged light steps, which simulate movement of an edge, and light flashes, w hich simulate movement of a spot or slit. 2. Within the excitatory rec eptive-field center of DS ganglion cells, apparent motion with prolong ed light steps elicits null-direction inhibition whose strength rises to 90% of maximum in 160 +/- 110 ms (7 cells) and then decays slowly, remaining above baseline longer than 2,000 ms for short interslit dist ances. 3. Prolonged light steps are generally effective for inhibiting any given excitatory receptive-field locus from an ovate-shaped area that extends asymmetrically in the direction that would be previously traversed by null-direction moving objects. This inhibitory area is ty pically larger than one-half the size of the receptive field center. T he strength of the inhibition is greater at short than long distances within this area. 4. The rise and fall times of the null-direction inh ibition elicited by apparent motion using prolonged light steps are so mewhat faster at large than short interslit distances. 5. Short light flashes (at sufficiently long interslit delays) elicit inhibition not only from the same asymmetric, ovate-shaped inhibitory field as long s teps of light, but also from loci completely surrounding the second sl it. This implies that the asymmetric, null-direction-specific inhibiti on is due to a temporally sustained mechanism. The symmetric inhibitio n elicited by short flashes may be due to the presence of the antagoni stic surround mechanism within the receptive-field center. The apparen t absence of this surround inhibition for preferred-direction apparent motion during prolonged light steps may be due to masking by facilita tion that is strongly evoked by long steps, but not flashes of light ( see accompanying paper). 6. The relatively slow rise time and sustaine d time course of the inhibition elicited by null-direction apparent mo tion within the excitatory receptive field center appears to distingui sh it from the inhibition elicited by stimulation within the receptive field surround, which has a much faster rise time and more transient time course. However, the sustained, null-direction inhibitory mechani sm that can be elicited by prolonged light steps within the excitatory receptive field center extends into the surround on the side of the r eceptive-field center previously traversed during null-direction motio n. 7. The contrast dependency of both the null-direction and the surro und inhibition is division-like, rather than subtraction-like. That is , the effect of both types of inhibition is to divide the response-ver sus-contrast curves by a constant factor rather than to subtract a fix ed value. 8. Use of prolonged light steps shows that the interaction b etween excitation and null-direction inhibition is largely segregated between the ON and the OFF pathways. Thus, for null-direction apparent motion, light onset of the first slit strongly inhibits response to t he onset, but generally not the offset, of the second slit. Light offs et of the first slit similarly tends to inhibit the response to light offset, but not onset, of the second slit. Thus selectivity for direct ion of motion appears to be computed independently for light and dark regions of moving objects.