SCIENCE@home

Numerate people are less likely to be biased by regular science reporting: the critical roles of scientific reasoning and causal misunderstanding

Olivia D. Perrin,Jinhyo Cho,E. Cokely,J. Allan,Adam Feltz,Rocio Garcia-Retamero

Published 2025 in Cognitive Research

ABSTRACT

Numerate people tend to make more informed judgments and decisions because they are more risk literate (i.e., better able to evaluate and understand risk). Do numeracy skills also help people understand regular science reporting from mainstream news sources? To address this question, we investigated responses to regular science reports (e.g., excerpts from CNN Health), testing a cognitive model linking numeracy, scientific reasoning, judgment biases, and causal theory errors (i.e., interpreting correlational information as causal). In Study 1 (n = 200), structural equation modeling indicated that more numerate people were less likely to exhibit judgment biases because they were better at scientific reasoning, which helped them avoid causal misinterpretations. Study 2 (n = 342) cross-validated findings from Study 1, indicating that the link between numeracy and scientific reasoning was also associated with improved cognitive self-assessment (e.g., reduced overconfidence on comprehension judgments). Results indicate that more numerate people may generally be less likely to confuse correlation and causation in regular science reporting. Results also suggest that numerate people are more likely to have acquired scientific reasoning skills that more generally support risk literacy and knowledge acquisition, consistent with Skilled Decision Theory. Discussion focuses on implications for risk literacy research, and includes a Risk Literacy Difficulty Analysis indicating that more than half of the USA adult population may be likely to misunderstand common types of regular science reports. Are some people more likely to misunderstand reputable science reporting, such as secondary reports about scientific findings distributed by mainstream news sources? In two studies, we measured cognitive skills (e.g., risk literacy as measured by the Berlin Numeracy Test) and investigated people’s susceptibility to potentially costly biases by asking young adults to review secondary science reports based on actual reporting (e.g., CNN Health). Consistent with previous research we found that many people confused correlational evidence with causal evidence (i.e., causal theory errors), which increased their susceptibility to a wide range of biases (e.g., judgment errors, overconfidence, misinformation propagation). However, we also found that risk literacy skills were consistently associated with scientific reasoning quality and reduced bias susceptibility. Results suggest that risk literacy skills may generally predict how likely it is that people will experience confusion and bias when reading secondary science reports. While misinformation and disinformation continue to threaten informed decision-making, the current findings indicate that even reputable science reporting may routinely bias a large proportion of adults (e.g., those who are not highly risk literate).

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-100 of 113 references · Page 1 of 2

12
Next

CITED BY