Эвристические стратегии принятия диагностических решений в неотложной неврологии

The aim of this study was to identify the heuristic strategies employed by emergency neurologists in making initial diagnostic decisions. Drawing on Gerd Gigerenzer’s theory of Simple Stopping Rules and One-Reason Decision Making [Gigerenzer et al., 1999; Todd & Gigerenzer, 2000], which posits that a decision between two or more alternatives can be made based on the value of a single cue, we tested the hypothesis that physicians apply the heuristic strategy Take The Best [Gigerenzer & Goldstein, 1996]. According to this strategy, in order to determine which of two diagnostic alternatives is more valid, one must: (a) examine cues in order of their validity, (b) stop searching once a cue discriminates between the alternatives, and (c) choose the alternative favored by that cue. Twenty-three interviews were conducted via video conferencing with practicing neurologists of varying levels of professional experience. Participants were presented with clinical cases in the domain of emergency neurology and asked to request additional information in order to make a diagnosis. During the analysis of interview transcripts, cue-seeking statements were identified that indicated the evaluation of competing diagnostic hypotheses. Results showed that participants typically selected the diagnostic alternative supported by the first discriminative cue. This finding demonstrates that neurologists predominantly rely on the Take The Best heuristic strategy. Moreover, the use of this heuristic conferred a significant diagnostic accuracy advantage to expert clinicians compared to less experienced counterparts. These findings underscore the importance of the Take The Best heuristic in accurate diagnostic decision-making within emergency neurology. They also suggest the potential for developing interactive case-based training simulators with adjustable levels of difficulty, tailored to diagnosticians with varying qualifications. The diagnostic cases used in the current study, along with similar cases, can serve as material for such simulators.