JMIR Publications

ABSTRACT

Background:

AI-enabled decision support systems are crucial tools in medical practice; however, their reliability is not absolute, necessitating human oversight for final decision-making. Human reliance on such systems can vary, influenced by factors like individual psychological traits and levels of experience among physicians.

Objective:

This study aimed to explore the psychological traits impacting subjective trust and reliance on medical AI advice, specifically examining relative AI reliance and relative self-reliance to assess appropriateness of reliance.

Methods:

A survey involving 223 dermatologists included lesion image classification tasks and validated questionnaires assessing subjective trust, propensity to trust technology, affinity for technology interaction, control beliefs, need for cognition, along with inquiries into medical experience and decision confidence.

Results:

Participants' accuracy significantly improved with AI support (t222=-3.3; P<.001; Cohen d=4.5) but showing only a 1% average increase. Reliance on AI was stronger with correct advice than incorrect advice (t222=4.2; P<.001; Cohen d=0.1). Notably, participants demonstrated a mean relative AI reliance of 10.0% (139/1384) and relative self-reliance of 85.1% (487/569), indicating a high level of self-reliance but only a low amount of AI reliance. Propensity to trust technology influenced AI reliance, mediated by trust (indirect effect = .024; 95% CI 0.0-0.0; P<.001), and medical experience negatively predicted AI reliance (indirect effect = -.001; 95% CI -0.0 to -0.0; P<.001).

Conclusions:

Findings underscore the need to design AI support systems in a way that assists less experienced users with a high propensity to trust technology to identify potential AI errors, while encouraging experienced physicians to actively engage with system recommendations and potentially reassess initial decisions.