That's my primary reason today.

The other major one, which the article/study perhaps misses, is collaboration. I and others who modify the pages I track will sometimes contribute after someone gets the ball rolling with a change.

However, the study may have accounted for this by looking at the content of these "most sociable editors" user pages. Here's the working paper version of the cited study, which I haven't read yet: http://blog.mikezhang.com/files/chinesewikipedia.pdf