Strategic behaviour and manipulation resistance in Peer-to-Peer, crowdsourced information gathering

Abstract

To realize the potential of crowdsourcing to tap into collective intelligence, one should consider how “crowds” are incentivized. One model, widely applied in blockchain oracles, constructs a coordination game where “true reporting” should be a focal point. Beyond well-known strategic behaviour associated to Keynesian Beauty Contests, we see this model provokes an additional layer of strategic behaviour. Specifically, for voting and payoff systems that satisfy minimal attack resistance assumptions, there always exist situations where participants are incentivized to provide rankings between three or more alternatives that neither reflect their honest assessment of the truth, nor correspond to their expectations of which alternatives are likely to win the vote. This result nuances our understanding of this class of Beauty Contests and mirrors issues in preference aggregation.