The “Black Cloud” on Call: A Field Study of Cognitive Bias in Workload Perception Among Physicians

Abstract Background The "black cloud" phenomenon, the belief that certain physicians disproportionately experience higher workloads during on-call shifts, is a recurrent anecdote in medical communities worldwide. However, empirical evidence is scarce as to whether this phenomenon reflects actual differences in workload or is a subjective belief rooted in cognitive biases. This study aimed to investigate the objective basis of the "black cloud" perception and its underlying psychological mechanisms by analyzing on-call data from physicians working in the same clinical environment. Methods A retrospective single-center observational study was conducted on four urologists who completed at least 90 on-call shifts each at a tertiary teaching hospital in Seoul from March 2024 to June 2025. All shifts were standardized to 12-hour units. The primary outcome was the number of emergency calls per shift. Secondary outcomes included the frequency of early morning calls (00:00–05:30) and high-call shifts (≥ 5 calls). Statistical analyses included descriptive statistics, one-way analysis of variance (ANOVA), negative binomial regression, and sensitivity analysis. Results A total of 493 on-call shifts were analyzed. There was no statistically significant difference in the mean number of calls per shift among the four physicians (1.25 vs. 1.12 vs. 1.21 vs. 1.22; F = 0.228, p = 0.877). The negative binomial regression analysis also confirmed no significant differences in the incidence rate of calls among individuals. Notably, the physician who self-identified as a "black cloud" experienced numerically higher mean calls and more frequent early morning and high-call shifts than their colleagues, but these differences were not statistically significant. Conclusions This study suggests that the self-perception of being a "black cloud" is likely not an objective reality but rather a psychological construct shaped by the interplay of cognitive mechanisms such as the availability heuristic, confirmation bias, the peak-end rule, and negativity bias. This implies that even highly trained medical professionals are not immune to cognitive biases amid uncertainty and stress. The findings underscore that data-driven feedback and education on cognitive biases could contribute to physician well-being and burnout prevention.