Effects of the source of advice and decision task on decisions to request expert advice
summary
A lab builds a Search & Rescue video game (four versions: lives vs oil, drone vs human advice) that generates advice with a 80% accuracy rule via a random-number generator. Sixty-five participants produce data (five always requested advice, eleven never), and players who ask for and follow advice score higher. Participants rely on the automated drone more consistently than the human confederate; no interaction appears between task type and source. Limitations include a single in-room confederate and possible task undervaluation.
AI-written summary · about