The article mentions that there is a weird problem among African American students at Berkeley: the middle of the SAT distribution and low end of the income dis

The article mentions that there is a weird problem among African American students at Berkeley: the middle of the SAT distribution and low end of the income distribution were doing better than the high ends of either. They chocked this result up to "disorientation". Treisman himself seems to emphasize the importance of the group aspect. Shouldn't we pull back a little on the claims about study then? It seems possible, though without much data, that hours of study and hours of group study aren't perfect substitutes, at least for some subset of the population.