Extended Data Fig. 5: Responses of RGCs to moving light bars in NW arrays-attached blind retina ex vivo, and choice-box-based behavioural test using moving light bars.
a, Schematics of ex vivo patch-clamp recording with moving light bars stimuli. b-d, Responses of representative RGCs to moving light bars stimulus (DMD, wavelength: 400 nm, light intensity: 10 μW·mm−2, width: 120 μm, speed: 200 μm/sec) in normal, blind and NW-array-attached blind retinas. e, Responses of a representative RGC to moving light bars stimuli with different width (width: 20–180 μm). f-g, Average firing rates and fraction of responses of RGCs in normal (n = 9 RGCs from 6 retinas), blind (n = 10 RGCs from 8 retinas) and NW-array-attached blind retinas (16 RGCs from 10 retinas) using moving light bars with width ranging from 0.64 to 5.76 degrees. h, Schematics of the choice-box-based moving bar detection test using moving and static light bar stimuli. i, Correct rate of normal mice (n = 6 mice), blind mice with glass implant (n = 6 mice) and blind mice with NW arrays implant (n = 8 mice) in moving bar detection test (5.25 degrees in width at 7.85 degrees/sec). j-k, Correct rate of normal mice (n = 3 mice), blind mice with glass implant (n = 3 mice) and blind mice with NW arrays implant (n = 4 mice) in moving bar detection test with moving light bars of different widths and speeds (width: 2.63–7.88 degrees, speed: 5.23–10.47 degrees/sec). All data are expressed as mean ± SEM. Statistical significance was determined using one-way ANOVA with Tukey’s multiple comparisons in g and i, two-way RM ANOVA with Tukey’s multiple comparisons in j and k. * indicates a significant difference between blind mice with NW arrays and blind mice with glass implant, # indicates a significant difference between blind mice with NW arrays and normal mice. * and # P < 0.05, ** and ## P < 0.01, *** P < 0.001.
