Behavioral and neural evidence on consumer responses to human doctors and medical artificial intelligence

Abstract

Will consumers accept artificial intelligence (AI) as a medical care provider? On the basis of evolution theory, we investigate the implicit psychological mechanisms that underlie consumers' interactions with medical AI and a human doctor. In a behavioral investigation (Study 1), consumers expressed a positive intention to use medical AI's healthcare services when it used personalized rather than mechanical conversation. However, neural investigation (Study 2) using functional magnetic resonance imaging revealed that some consumers' implicit attitudes toward medical AI differed from their expressed behavioral intentions. The brain areas linked with implicitly apathetic emotions were activated even when medical AI used a personalized conversation, whereas consumers' brains were activated in areas associated with prosociality when they interacted with a human doctor who used a personalized conversation. On the basis of our neural evidence, consumers perceive an identical personalized conversation differently when it is offered by a medical AI versus a human doctor. These findings have implications for the area of human-AI interactions and medical decision-making and suggest that replacing human doctors with medical AI is still an unrealistic proposition.