Probability overestimation induced by icon arrays

Datos Bibliográficos

ID21284030
AutoresKazuhisa Nagaya (0000-0002-7889-816X, Yamaguchi University, autor de correspondencia), Hiroyuki Shimizu (0000-0002-5926-2713, Kobe Gakuin University)
Año2024
Volumen248
Páginas104352
Fecha de publicación2024-08-01
Peer ReviewedSí
Open AccessSí
TipoARTICLE
RevistaActa Psychologica (JOURNAL)
Identificadores de la revistaISSN: 0001-6918 • E-ISSN: 1873-6297
EditorialElsevier BV (PUBLISHER)
DOI10.1016/j.actpsy.2024.104352
PMID38889658
OpenAlexW4399748953
IdiomaEN
Referencias citadas41

In decision making under risk, where the consequences of choices vary depending on specific probabilities, a proper interpretation of the presented probabilities is essential. However, because people are not good at evaluating numerical values, such as percentages, efforts have been made to accurately convey presented probabilities using visual aids represented by icon arrays. However, previous studies have suggested that icon arrays produce an overestimation of probabilities under specific conditions. Especially, (a) the arrangement of highlighted icons and (b) the severity of the outcomes represented by highlighted icons can lead to an overestimation of probabilities. By considering the absolute estimation of probabilities indicated by icon arrays and manipulating the aforementioned factors, this study investigated whether these factors could elicit an overestimation of probabilities in an additive manner or whether they have interactive effects. Two studies (total N = 1171) consistently showed that the arrangement of highlighted icons and the outcome severity they indicate induced a perceptual bias in the estimation of the probabilities presented in icon arrays in an additive manner. When the highlighted icons were random and the outcomes they indicate were severe, the probabilities presented by the icon arrays were overestimated. Accordingly, the study not only confirms the robustness of the findings of previous studies but also expand our understanding of the probability overestimation caused by icon arrays. Here, we discuss our findings in terms of the impact of risk-related communication

Icon · Advanced Statistical Modeling Techniques · Computer Science · Distributed Sensor Networks and Detection Algorithms · Optimal Experimental Design Methods · Psychology