Weirdly enough, people would rather listen to an advisor who's usually 'right,' even if following their advice actually makes things worse for them.
When choosing between a human and an algorithm, people prioritize 'frequency of being right' over 'total value gained.' This means we will trust a source that gives us small wins often, even if it causes us to lose more money or resources in the long run compared to a less frequent but more accurate advisor.
Learning to choose between advisors, algorithmic and human, over repeated interactions.
PsyArXiv · uqbce_v1
People increasingly consult algorithmic aids repeatedly, yet most evidence on algorithm aversion/appreciation comes from one-shot decisions. Across five preregistered incentive-compatible studies (Prolific; N=1,351), we examine how people learn whom to trust when advisors disagree. Study 1 elicits advice from experienced participants, revealing a bias towards the option that is better most of the time, even when it’s worse in expectation. Studies 2–5 then paired this human advice with algorithms