, 2008). Interestingly, our present results demonstrate a strikingly similar developmental pattern of direction selectivity in the upper layer visual cortical neurons. Thus, as in the retina, direction click here selectivity was detected at eye opening and emerges independently of visual experience. Furthermore, direction-selective neurons recorded just after eye opening in both the cortex and the retina have a similar preference for the dorsal and anterior directions of motion. This preference disappeared in the cortical neurons of adult mice. One possible

conclusion from these results is that in the mouse visual system direction selectivity emerges in the retina and is relayed to the visual cortex. This notion finds support TSA HDAC mouse in the previous observations that On-Off direction-selective retinal ganglion cells project both to the LGN and to the superior colliculus in specific laminae (Huberman et al., 2009). In line

with this anatomical evidence, direction-selective neurons were recorded in the rat superior colliculus around eye opening (P13) and, as in the mouse visual cortex, the proportion of direction-selective neurons was found to remain stable from P15 to adulthood (Fortin et al., 1999). By contrast, the relay of direction-selective information through the rodent LGN is less clear. While the receptive fields of neurons in the mouse LGN were described as center-surround with exclusively ON-center or OFF-center responses (Grubb and Thompson, 2003), direction-selective cells in mouse or rat LGN are not yet described. However, it remains unclear whether LGN neurons that receive direct projections from direction-selective retinal ganglion cells were ever studied specifically. Another possibility is that LGN-receptive fields

are more broadly tuned and that direction selectivity is generated again at the cortical level. It is noteworthy that the directional tuning of the cortical isothipendyl neurons recorded in this study is more narrow than the directional tuning of the mouse retinal ganglion cells (Elstrott et al., 2008). This result indicates that in mice the direction selectivity is refined along the retino-geniculo-cortical pathway. It is unclear whether such a possible refinement is found only in mice. Interestingly, there is some evidence for direction bias in the retinal ganglion cells of cats (Levick and Thibos, 1980 and Shou et al., 1995) as well as in the cat and primate LGN (Vidyasagar and Urbas, 1982, Thompson et al., 1994 and Xu et al., 2002). However, detailed studies in the retina and LGN of these species are needed for solving this issue. Taken together, there is accumulating evidence that the anatomical difference between the primary visual cortices of higher mammals (ferrets, cats, or primates) and rodents, i.e., columnar organization versus salt-and-pepper structure, is paralleled by functional differences during development.