Yeah, I agree--very suspicious that one would be aware of age, race, and native language, but not skin color? How does that even work? ("I can see that you're African but not that your skin is the typical color for (sub-Saharan) Africans, even though that's really the only way I know how to distinguish Africans from other peoples"??)

On the other hand, multiculturalism has some significant psychological downsides that need to be discussed rather than ignored (they were ignored in the Psychology Today article).

The worst--though not only--of these is "stereotype threat", where a group that has a negative stereotype associated with them is negatively affected when their group identity is highlighted, especially in any context where the negative stereotype is relevant. For instance, if the stereotype is "girls are bad at math", focusing on how someone is a girl right before they take a math test will tend to provoke anxiety and decrease their performance. It's not because they believe the negative stereotype--it's because they feel like they're on the spot to disprove the negative stereotype right now, which adds extra stress. They don't need the extra stress. They just need to focus on the math, like everyone who isn't suffering from stereotype threat.

Because of stereotype threat, I think there's an argument that multiculturalism is racist: by forcing people to wear the mantle of their group identity (since everyone pays attention to it), you provoke stereotype threat, increasing the anxiety of and likely reducing the performance of members of those groups who are already most historically disadvantaged. In contrast, the colorblind approach does not provoke stereotype threat. Although it may be possible to mitigate the risks of multiculturalism provoking stereotype threat, ignoring the issue while decrying the colorblind approach is not the path to mitigation.