Figure 3.

The modulatory effect of light on test performance depends on both rods/cones and melanopsin (sample phases at 350 lux). The background irradiance was manipulated at test (10 or 350 lux) but the visual context remained unchanged (white arena). (a) Schematic shows the spontaneous object recognition task under the 350 lux → 10 lux and 350 lux → 350 lux conditions. In both conditions, a mouse was allowed to explore two identical replicates of an object (indicated by red squares) in the white arena for 10 min in the sample phase. After a 5 min delay, a novel object (indicated by green circles) was introduced, and the animal was allowed to explore the familiar and novel objects for 2 min in the test phase. For animals in the 350 lux → 350 lux condition, both the sample and test phases were performed under 350 lux, measured at the centre of the white arena. For animals in the 350 lux → 10 lux condition, the light level in the test phase was reduced to 10 lux. All other aspects of the task were identical under the two conditions. (b–d) Object recognition ratios in WT mice, rd/rd cl, and Opn4^−/− mice, respectively. Performance in WT mice was sensitive to the background light level at test. Recognition ratios were higher when the test was given at 10 lux than when it was given at 350 lux (n = 10 in the 10-lux test; n = 19 in the 350-lux test; (b)). No effect of light on performance was found in rd/rd cl and Opn4^−/− mice (n = 5 per genotype in the 10 lux test; n = 10 per genotype in the 350 lux test; (c,d)); however, these mice could discriminate between novel and familiar objects. In the diagrams in (b–d), R/C, rods/cones; M, melanopsin-expressing pRGCs; asterisk: significant effect of test irradiance (p < 0.05); dagger: significant object recognition performance (above zero; p < 0.05); error bars denote standard error of mean.