Communicating Scientific Uncertainty About the COVID-19 Pandemic: Online Experimental Study of an Uncertainty-Normalizing Strategy

"...a strategy aimed at normalizing uncertainty as an expected state of affairs was effective in reducing at least some aversive psychological responses to the communication of uncertainty, whereas widely used alternative strategies aimed at promoting hope and prosocial values had no such effect."

Expert guidelines on crisis communication stress the importance of acknowledging uncertainty - a defining feature of public health crises such as COVID-19 - in order to foster public accountability and trust. However, the communication of scientific uncertainty has the potential to promote negative cognitive, emotional, and behavioural responses, such as heightened risk perceptions, emotional distress, and decision avoidance. This study evaluates whether a communication strategy emphasising that uncertainty is expected/normal during crises can reduce "ambiguity aversion", and it compares the effectiveness of this approach to conventional public communication strategies that promote hope and prosocial values.

In an online factorial experiment conducted from May to June 2020, a national sample of 1,497 United States (US) adults read one of five versions of an informational message describing the nature, transmission, prevention, and treatment of COVID-19:

• The control reproduced from content from a public website produced by a government public health department, containing no explicit communication of scientific uncertainty.
• The uncertainty condition highlighted the existence of scientific uncertainty about the controllability, prognosis, and severity of the COVID-19 pandemic.
• The uncertainty + uncertainty-normalising condition combined expressed uncertainty with language emphasising the unknowability of these various aspects of COVID-19 and the expected nature of scientific uncertainty.
• The uncertainty + hope-promoting condition combined expressed uncertainty with language conveying optimism about future advances in knowledge and control over the pandemic.
• The uncertainty + prosocial condition combined expressed uncertainty with language encouraging awareness of obligations to other community members and concern for the collective good.

Participants then completed a survey to measure manifestations of ambiguity aversion (i.e., perceived likelihood of getting COVID-19, COVID-19 worry, and intentions for COVID-19 risk-reducing behaviours and vaccination).

The experiment found that communication of scientific uncertainty about the COVID-19 pandemic increased perceived likelihood of getting COVID-19 and worry about COVID-19, consistent with an ambiguity-averse cognitive and emotional response. However, it did not affect intentions for risk-reducing behaviours or vaccination. (In other words, ambiguity aversion in this study was manifest cognitively and emotionally, but not behaviourally.) The uncertainty-normalising strategy reduced aversive effects of communicating scientific uncertainty, resulting in levels of both perceived likelihood of getting COVID-19 and worry about COVID-19 that did not differ from the control message. In contrast, the hope-promoting and prosocial strategies did not decrease ambiguity-averse responses to scientific uncertainty.

Two factors, age and political affiliation, were found to moderate the effects. For example, among adults over 50 years of age, all active uncertainty communication strategies resulted in higher behavioural intentions compared to the control (i.e., no uncertainty) strategy, while for younger adults, all active uncertainty communication strategies resulted in lower intentions.

In reflecting on the findings, the researchers note that, contrary to their hypothesis that uncertainty communication might lower intentions for COVID-19 risk-reducing behaviours or vaccination, uncertainty-normalising language had no such effect; they consider several explanations. For instance, several of the behaviours (e.g., regulations requiring social distancing and use of masks) are mandated. Furthermore, uncertainties about the controllability, prognosis, and severity of the COVID-19 pandemic may increase intentions by promoting fear about the consequences of avoiding risk-reducing behaviours, thereby promoting a tendency toward action. "More research is needed to understand the factors that moderate people's behavioral responses to different uncertainties and favor either inaction or action."

In conclusion: "A major barrier to open, explicit communication of the uncertainties that inevitably exist during public health crises is a real concern about exacerbating both the perception of vulnerability as well as feelings of fear and panic among the general public....[The] findings suggest, however, that language aimed at normalizing these uncertainties can reduce aversive cognitive and emotional responses to them."