Section snippets

Overview of the present work

In the present work, we used evidence accumulation modeling, specifically the drift-diffusion model (DDM; Ratcliff, 1978; Ratcliff et al., 2016), to examine the cognitive mechanisms behind wishful thinking in predictions. Here, we define predictions as an inferential process that involves determining which of two uncertain outcomes will occur. The DDM, designed for speeded two-choice decisions, can separate evidence-accumulation processes from judgment-level processes (see Fig. 1), providing a

Task overview

Our experiment used a speeded prediction task, in which participants were presented with bi-colored stimuli—grids composed of squares—that varied in color discriminability (e.g., some grids had many squares of yellow and few squares of purple, other grids had a roughly equal number of yellow and purple squares). On each trial, one of the squares in the grid was randomly selected by a computer. Participants were asked to predict the color of the square that would be selected by the computer.

Data exclusions and manipulation checks

To ensure adequate data quality, we excluded participants who failed to respond on more than 10 trials or gave more non-normative than normative responses on filler trials. In total, 13 participants were excluded based on these criteria, resulting in a final sample size of 147. On the trial level, we excluded trials with response times below 200 milliseconds and trials with response times above/below three standard deviations from the mean response time for that participant. To improve general

Discussion

When and why outcome preferences bias people’s expectations are of substantial concern (de Molière & Harris, 2016; Harris & Hahn, 2011; Logg et al., 2018). A key finding from prior research is that people are especially prone to biased estimations when making predictions, as opposed to judgments of likelihood (Park et al., 2023; Windschitl et al., 2010). However, the mechanism by which desirability impacts predictions has remained unclear. In the present work, we tested how outcome preferences

Preregistration

Author note

This work was supported by Grant SES-1851738 (PW) and SES-212112 (TJP) from the National Science Foundation. All work was completed at the Psychological and Brain Sciences Department, University of Iowa, Iowa City, IA.

CRediT authorship contribution statement

Jeremy D. Strueder: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Conceptualization. Inkyung Park: Writing – review & editing, Methodology, Formal analysis, Data curation, Conceptualization. J. Toby Mordkoff: Writing – review & editing, Supervision, Formal analysis. Timothy J. Pleskac: Writing – review & editing, Formal analysis. Paul D. Windschitl: Writing – review & editing, Supervision, Methodology, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

All other authors declare no conflicts of interest. Funding: This work was supported by Grants SES-1851738 (PW) and SES-212112 (TJP) from the National Science Foundation. Artificial intelligence: No artificial intelligence assisted technologies were used in this research or the creation of this article. Ethics: This Study was approved by the University of Iowa Institutional Review Board (#202102151).

© 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.