Can Medical Diagnosis Benefit from “Unconscious Thought”?

Abstract

The Unconscious Thought Theory argues that making complex decisions after a period of distraction can lead to better decision quality than deciding either immediately or after conscious deliberation. Two studies have tested this Unconscious Thought Effect (UTE) in clinical diagnosis with conflicting results. The studies used different methodologies and had methodological weaknesses. We attempted to replicate the UTE in medical diagnosis by providing favorable conditions for the effect, while maintaining ecological validity.Family physicians (N=116) diagnosed three complex cases in one of three thinking modes: immediate, unconscious (UT) and conscious (CT). Cases were divided into short sentences, which were presented briefly and sequentially on computer. After each case presentation, the immediate response group gave a diagnosis; the UT group performed a 2-back distraction task for 3 minutes before giving a diagnosis; and the CT group could take as long as necessary before giving a diagnosis.We found no differences in diagnostic accuracy between groups (P=0.95). The CT group took a median of 7 seconds to diagnose, which suggests that physicians were able to diagnose "online", as information was being presented. The lack of a difference between the immediate and UT groups suggests that the distraction had no additional effect on performance.To assess the decisiveness of the evidence of this null result, we computed a Bayes

A critical review and meta-analysis of the unconscious thought effect in medical decision making

“…In summary, the experiments conducted so far to explore the UTE in clinicians’ decision making have yielded contradictory results, with the effect being observed in the original report by de Vries et al (2010) and in some (but not all) conditions of the Mamede et al (2010) and Bonke et al (2014) studies, but not in the Woolley et al (2015) study. Although this suggests that the effect might not be reliable, alternative explanations are possible.…”

“…Finally, in a recent study by Woolley et al (2015) , family physicians were asked to diagnose three difficult cases based on real patients. Accuracy was measured against the patient’s known diagnosis (strict measure), and also against each case’s plausible diagnoses (lenient measure).…”

“…We also included as potential moderators two features that have been manipulated in UTE experiments with medical decision tasks or that have varied substantially from one experiment to another. First, Mamede et al (2010) and Woolley et al (2015) used two different controls to assess the UTE: a CT condition, where participants responded after conscious deliberation, and an immediate condition, where participants responded immediately after the materials were presented. Some authors have argued that the UTE might actually be the product of poor performance in the CT conditions and that an immediate condition is a better control for UTE experiments ( Shanks, 2006 ; Waroquier et al, 2010 ).…”

“…Effect sizes for Bonke et al (2014) were computed from the descriptive statistics provided in their Table 4. In the case of Woolley et al (2015) , we had access to the raw data, which allowed us to compute different effect sizes for each control condition (CT and immediate) and for two levels of difficulty (Cases 1 and 2 were coded as complex and Case 3 as simple, based on participants’ mean accuracy rates of 18, 21, and 42%, respectively). These effect sizes were submitted to random-effects and mixed-effects meta-analyses using the metafor R package ( Viechtbauer, 2010 ).…”

“…A composite effect size was computed for all the related conditions in the Mamede et al (2010) and Bonke et al (2014) studies. Similarly, we computed a new effect size for Woolley et al (2015) collapsing data across the three cases (i.e., the difficulty factor) and the two control conditions (CT and immediate). The results of the second meta-analysis are depicted in Supplementary Figure S1 .…”