There just need to be a post on my recently accepted paper: Pulvinar Retinotopy of a Prosimian (bush baby) The basic idea is this: people had looked at the pulvinar of two and half primate species. The two species (macaque and cebus) have been reported to have different retinotopic organizations. The difference would be interesting for evolution people, but would pose a problem for people who just started unveiling the function of pulvinar. Now, let's look at an earlier species. Which species does its pulvinar look like more? What pulvinar features are common for all three species?
It turns out that bush baby pulvinar is quite like that of the macaque monkey, with two maps and all. If these two agree, what happens with the third species (cebus)? I believe it's just plotted differently, as the raw data presented in that paper actually agree with the two maps in bush baby and macaque, with a third dorsal area which looks like macaque Pdm, the region so many people like yet only one real report exist. So, if the cebus maps are replotted, the retinotopy reported for primate species are quite alike. So pulvinar functions found in one species have a better chance of being translated to another species.
As for the two difference between bush baby and macaque discussed in the paper, I personally believe the second order mapping of macaque lateral map doesn't really exist. Connection and chemoarchitecture strongly hint at a shell region of PL that has a separate map on pulvinar surface. Put this map together with the lateral map and you have isoelevation contours of a "second order mapping". We didn't get the shell in bush baby pulvinar because pulvinar surface is mostly horizontal here and a shell, if exist, is only briefly encountered by each penetration, and cannot be told apart from the fibers above.
The lack of columnar structure in bush baby pulvinar maps shouldn't really surprise anyone. If there is a columnar structure we should see layers already, like in LGN. There is this report saying marmoset pulvinar should have layered subdivisions based on connections. Too bad it was a somewhat crude study. Some people should look at that electrophysiologically.