"Hours" are less and less relevant unfortunately. Especially for FO's flying long haul routes. Somewhere back in one of the previous threads this was covered with the approximiate math that the PF probably had something on the order of 9 hours a month of actual "PF" time with actual stick time literally measured in minutes (if any at all). So over the course of a year he might have 90-100 total flying hours with possibly a dozen takeoffs and landings and potentially less then an hour of "hands on" line flying.
So how unexpected is it really that under extreme stress and in totally unexpected circumstances he would fall back on the protections that had been heavily stressed in a low altitude stall recovery....in effect he appears to have been instinctively flying a missed approach type "recovery".
So the reality is that a bright carefully selected 300 hr (actual type specific with significant hands on stick/sim time) would be more likely to respond correctly then a more experienced pilot whose training was far in the past.
Personally I think this incident highlights how important actual hand-flying is toward maintaining situational awareness. My thought is that the over reliance on automation decreases the ability of the "PF" to actually be mentally "flying" the airplane which can lead to a mental "brain lock" when suddenly forced to assume flying responsibility. From all I've seen the AP disconnect left the plane relatively trimmed and stable with a slightly banked and nose down attitude at (or immediately after) disconnect. The flight control inputs required should have been very minimal and consistent with normal inputs in mild turbulence.
I've got to belive that the combination of surprise and a focus on low altitude upset recovery procedures led to an ingrained response that was completely incorrect for the specific circumstances.