Feedback Information on Cumulative Payoff in a Bandit Experiment: Meaningful Learning in Weighted Voting