Is there any research on the usability of the pull-to-refresh gesture?

While the original thought behind pull-to-refresh was to take advantage of the scroll motion, which a user would use to see if any new messages appeared at the top or bottom of the list, to refresh a list and see if there isn't something new.

Unfortunately, this pattern is being used where that no longer applies -- for example, to refresh Chrome pages -- and it's not very discoverable. Is there any research on whether people have problems with this gesture, and, conversely, whether they expect it in all apps with refresh now?