EFFECT OF PASSIVE EYE-MOVEMENT ON RETINOGENICULATE TRANSMISSION IN THE CAT

1. The nature and time window of interaction between passive phasic eye movement signals and visual stimuli were studied for dorsal lateral geniculate nucleus (LGN(d)) neurons in the cat. Extracellular recordings were made from single neurons in layer A of the left LGN(d) of anesthetized paralyzed cats in response to a normalized visual stimulus presented to the right eye at each of several times of movement of the left eye. The left eye was moved passively at a fixed amplitude and velocity while varying the movement onset time with respect to the visual stimulus onset in a randomized and interleaved fashion. Visual stimuli consisted of square-wave modulated circular spots of appropriate contrast, sign, and size to elicit an optimal excitatory response when placed in the neurons' receptive-field (RF) center. 2. Interactions were analyzed for 78 neurons (33 X-neurons, 43 Y-neurons, and 2 physiologically unclassified neurons) on 25-65 trials of identical visual stimuli for each of eight times of eye movement. 3. Sixty percent (47/78) of the neurons tested had a significant eye movement effect (ANOVA, P < 0.05) on some aspect of their visual response. Of these 47 neurons, 42 (89%) had a significant (P < 0.05) effect of an appropriately timed eye movement on the number of action potentials, 36 (77%) had a significant effect on the mean peak firing rate, and 31 (66%) were significantly affected as evaluated by both criteria. 4. The eye movement effect on the neurons' visual responses was primarily facilitatory. Facilitation was observed for 37 (79%) of the affected neurons. For 25 of these 37 neurons (68%), the facilitation was significant (P < 0.05) as evaluated by both criteria (number of action potentials and mean peak firing rate). Ten (21%) of the affected neurons had their visual response significantly inhibited (P < 0.05). 5. Sixty percent (46/78) of the neurons were tested for the effect of eye movement on both visually elicited activity (visual stimulus contrast = 2 times threshold) and spontaneous activity (contrast = 0). Eye movement significantly affected the visual response of 23 (50%) of these neurons. However, spontaneous activity was significantly affected for only nine (20%) of these neurons. The interaction of the eye movement and visual signals was nonlinear. 6. Nine of the 12 neurons (75%) tested had a directionally selective effect of eye movement on the visual response, with most (8/9) preferring the temporalward direction. 7. Maximal facilitation occurred when the eye movement began from 50 ms before to coincident with visual stimulus onset, and maximal inhibition occurred when the eye movement began from 100 to 50 ms before visual stimulus onset. The facilitation began ≃72 ms after the initiation of the eye movement and, on average, persisted for 50-100 ms. The inhibition persisted for 100-150 ms, on average. 8. Facilitatory effects of appropriately timed eye movements were equally likely to be found among X- and Y-cells (X, 51%; Y, 56%; χ2 test P > 0.05). However, inhibitory effects were more likely for Y-cells (78%) than X-cells (22%), although this difference was not significant (P > 0.05). The magnitude of the eye movement effect was greater for X-cells (particularly for those with RFs located at < 5 ±) than Y-cells (X = 44%, Y = 20%, Mann-Whitney U test; P < 0.01). 9. The likelihood of an eye movement effect on a neuron's visual responses varied significantly as a function of its RF eccentricity. Facilitation was more likely among neurons with RF eccentricity within 5° of central vision (87%) versus those with RF eccentricity > 15° from central vision (26%) (χ2 test, P < 0.01). Eye movement inhibition of visual responses was observed only for neurons with peripheral RFs. 10. For 26 of 78 neurons, the responses to both phasic eye movements and steady-state eye position changes (as described in the previous report) were analyzed. For 50% (13/26) of these neurons, an appropriately timed eye movement significantly affected the visual response. Twleve percent (3/26) were significantly affected by eye position only, and the other 15% (4/26) were significantly affected by both eye movement and position signals. 11. Retrobulbar block completely eliminated the facilitation of the visual response, suggesting extraocular muscle proprioception as the source of the afferent eye movement signal. 12. The role of such afferent eye movement signals is considered with respect to psychophysical and behavioral evidence for postsaccadic enhancement of visual sensitivity and perisaccadic suppression in the geniculostriate pathway.