Simpler is better for everyone: causal diagram complexity and the role of perceived knowledge in health decisions

Abstract People frequently make decisions about their health in contexts where information is incomplete, complex, or difficult to interpret. Although a range of tools have been developed to support these decisions, including visual and causal representations, it remains unclear how the structure and complexity of such information affect decision quality. Causal representations are often assumed to be beneficial because they convey relationships between risk factors, behaviors, and outcomes, potentially supporting more effective decisions. However, prior work suggests that this benefit is not guaranteed: for everyday decisions, complex causal models can impair performance and, in some cases, lead to worse decisions than providing no information at all. Little is known about how individuals’ prior knowledge shapes their use of such information. In particular, the gap between perceived and actual knowledge may influence how new information is interpreted and applied during decision-making. We systematically investigated the relationship between information complexity, self-rated knowledge, and objective knowledge in the context of decision-making about Type 2 diabetes. In an online experiment (N = 1077 U.S. adults), causal diagrams led to higher decision accuracy than no diagram, but performance declined as complexity increased. Self-rated diabetes knowledge was correlated with actual knowledge (DKQ-24) but not decision accuracy. Diabetes experience was associated with both overestimation and lower decision accuracy, suggesting that experience may contribute to both calibration and poorer decision-making. Our findings suggest that health decision-aids should be concise rather than comprehensive, and that individuals who appear the most confident may be the most in need of targeted decision-support.