Figure 2.

Motion training was ineffective in dark-reared animals. a, Imaging fields of direction selectivity before and after bidirectional or unidirectional motion training for dark-reared animals. b, c, Cumulative density histograms of direction index values before and after training for dark-reared animals (b) and light-reared animals (c). Differences in DI were not significant for dark-reared animals (Kruskal–Wallis, p = 0.2), but were significant for light-reared animals (Kruskal–Wallis, p < 0.001).