In the past decades, research in cognitive science has increasingly improved our understanding of human cognition and decision-making processes. On the one hand, the “heuristics and biases” research program highlighted how human reasoners often fail to conform to normative standards of rationality, falling prey to systematic and predictable “cognitive illusions”. On the other hand, further research explored how to design communication strategies and reasoning tools that are more ergonomic - that is, that make easier and more transparent the processing and evaluation of complex information. Here, we report on a preliminary analysis (including a pilot study) of understanding and communication of statistical information concerning COVID-19-related risks among the public. Our tentative conclusions are as follows: first, in line with previous literature, transparent communication using a “natural frequency” format (instead of percentages or probabilities) does improve understanding; second, many important institutions do not employ cognitively ergonomic communication strategies, leaving much room for further improvement.