Dimensions of decision-making: An evidence-based classification of heuristics and biases. (1st August 2019)

• Record Type: Journal Article
• Authors: Ceschi, Andrea; Costantini, Arianna; Sartori, Riccardo; Weller, Joshua; Di Fabio, Annamaria
• Is Part Of: Personality and individual differences. Volume 146(2019)
• Page Start: 188
• Page End: 200
• Publication Date: 2019-08-01
• Subjects: Heuristics and Biases -- Decision-Making -- Individual differences -- Mindware gaps -- Positive Illusions -- Negativity effect -- Anchoring and Adjustment; Personality -- Periodicals; Individuality -- Periodicals; Individuality -- Periodicals; Personality Development -- Periodicals; Personnalité -- Périodiques; Individualité -- Périodiques; 155.205
• DOI: 10.1016/j.paid.2018.07.033
• Languages: English
• ISSNs: 0191-8869

Abstract

Abstract: Traditionally, studies examining decision-making heuristics and biases (H&B) have focused on aggregate effects using between-subjects designs in order to demonstrate violations of rationality. Although H&B are often studied in isolation from others, emerging research has suggested that stable and reliable individual differences in rational thought exist, and similarity in performance across tasks are related, which may suggest an underlying phenotypic structure of decision-making skills. Though numerous theoretical and empirical classifications have been offered, results have been mixed. The current study aimed to clarify this research question. Participants (N = 289) completed a battery of 17 H&B tasks, assessed with a within-subjects design, that we selected based on a review of prior empirical and theoretical taxonomies. Exploratory and confirmatory analyses yielded a solution that suggested that these biases conform to a model composed of three dimensions: Mindware gaps, Valuation biases (i.e., Positive Illusions and Negativity effect), and Anchoring and Adjustment. We discuss these findings in relation to proposed taxonomies and existing studies on individual differences in decision-making.