There is an ongoing debate about whether and how to communicate the epistemic uncertainty of scientific findings. In this debate, skeptics that shy away from communicating uncertainty measures of...Show moreThere is an ongoing debate about whether and how to communicate the epistemic uncertainty of scientific findings. In this debate, skeptics that shy away from communicating uncertainty measures of their work, commonly name loss of trust in scientists and the information they produce as important arguments. Efforts to provide clarity in this discussion, generated insight on many aspects of epistemic uncertainty but the effect of different magnitudes, or sizes of epistemic uncertainty has rarely been the subject of studies. To collect evidence on the nature of the relationship between different magnitudes of communicated uncertainty in science and the trust scientific laymen have in the source of information, this study presents respondents to a survey (N = 235) with one of three magnitudes of uncertainty (no uncertainty, low uncertainty, or high uncertainty) across three different formats (verbal, numerical, and visual) and measures their trust in the source of information. No association was found between magnitude and trust in the source. Furthermore, exploration of the data revealed no interaction between the format and the magnitude of epistemic uncertainty on trust in the source. This study’s limitations in representability and validity restrict strong conclusions. Future research should ensure that participants comprehend what high or low magnitudes of uncertainty are in the context of scientific findings, so participants’ comprehension of the magnitudes is increased and hence also the meaningfulness of judgments they make based on them.Show less
Epistemic uncertainty refers to the doubts surrounding numerical information about past or current events. Communicators may present this information in various formats, with the numerical and...Show moreEpistemic uncertainty refers to the doubts surrounding numerical information about past or current events. Communicators may present this information in various formats, with the numerical and visual offering more precision than the verbal. However, previous research has established that even numerical and visual presentations of uncertainty differently influence individuals' understanding, with the former encouraging them to interpret the ranges of possible values in a binary way (i.e., seeing those within a numerical interval that conveys uncertainty around a best estimate as highly possible and those outside as nearly impossible). Conversely, visual displays (e.g., violin plots) promote more graded interpretations. Different magnitudes of uncertainty and differences in numerical abilities may further affect individuals' understanding. Accordingly, we examined the differences in interpreting uncertainty ranges by varying the format of presentation and the magnitude of uncertainty. We also assessed individuals' subjective indications of their own numerical abilities. Our findings suggest that visually, rather than numerically, presented uncertainty ranges encourage binary interpretation. This effect appears less pronounced in conditions with low uncertainty and is not affected by individuals' subjective beliefs about their numerical abilities. These results directly contradict previous findings, where numerically presented ranges are the ones that promote a binary understanding. The unexpected scores on our outcome variable and the underpowered nature of some analyses make it difficult to draw strong conclusions, which warrants further research.Show less
