The science: Lovely study. Confirms and extends previous work in rabbits and mice to primates. This paper basically says in a rigorous way: We finally found wh

The science: Lovely study. Confirms and extends previous work in rabbits and mice to primates. This paper basically says in a rigorous way: We finally found what we knew should be there in primates. Hurray. This makes a ton of sense because rabbits, mice, and primates are members of the mammalian clade called “superprimates” or more formally the euarchontoglires. Why throw away a perfectly good set of cells and a bunch of brain regions that all ither mammals have? Rodents and rabbits are evolutionarily more closely related to humans than most other mammals—-dogs, cats, bats, cows, etc. https://en.wikipedia.org/wiki/Euarchontoglires https://en.wikipedia.org/wiki/Euarchontoglires Given some of the comments here, it is worth pointing out that the retinas are part of the central nervous system. They are bilateral forward-facing out-pouching of the prosencephalon. The retinas and optic nerves you are using to read this are the only parts of the brain that can be directly inspected with an ophthalmoscope. Retina it is also the ultimate biologically tractable neural network in that input and output channels are very accessible. If you want any biological inspiration for DNNs look no farther. Highly recommend the first three reference of this new paper to anyone interested in how wet neural network can works. Focus on starburst amacrine cells. 1. A great review of this topic https://pubmed.ncbi.nlm.nih.gov/22314444/ https://pubmed.ncbi.nlm.nih.gov/22314444/ 2. Hard core analysis in mice. And yes, we have learned a great deal about vision by studying mice, arguably mouse work is now trend setting even in vision research. This is an example. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3818553/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3818553/ 3. Open article on link up between directionally selective retinal ganglion cells and the rest of the brain. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0004320 https://journals.plos.org/plosone/article?id=10.1371/journal...