Effects of emotional narratives and uncertainty visualization on non-experts’ trust in climate change forecast maps

Climate change is an ongoing environmental and societal challenge. Communicating its ramifications and related uncertainties clearly to stakeholders and the public is an imperative task for time-critical decision-making. Public communication about climate change often includes maps, aimed at facilitating the understanding of complex scientific findings and making these more accessible to non-specialist audiences. This is especially important when difficult concepts such as inherent uncertainties related to climate predictions are involved.

While climate change communication may appear abstract and distant to non-experts, climate change discourse often involves strong emotional responses from the public. Engaging visual storytelling with climate change maps may be a useful strategy to reduce the psychological distance of the public. However, elicited emotions may influence how people perceive the presented information and thus their willingness to trust the maps.

We aimed to investigate the effect of emotional narratives on map readers’ trust in visualized (un)certainty information in static climate change forecast maps. We applied a 3x2 mixed factorial, map-based study design, including electrodermal activity measurements and eye-tracking. We designed three versions of climate change prediction map stimuli, inspired by the Swiss Climate Scenarios CH2018. Uncertainty was operationalized as a within-subjects independent variable such that participants viewed 18 map stimuli in total, showing different climate variables in randomized order, equally distributed across three conditions: (1) without uncertainty information, (2) uncertainty visualized as black gridded dots, or (3) uncertainty visualized as black randomly distributed dots. Following prior research, we used the term ‘certainty’ in our map stimuli, as it is better understood by the audience than ‘uncertainty’. We used narrative instructions as the between-subjects independent variable, with participants randomly assigned and matched across groups to one of the two conditions: (1) emotion or (2) control. In the emotion condition, each map stimulus was accompanied by an emotion-inducing verbal narrative and a human cartoon character. In the control condition, participants viewed the same map stimuli accompanied only by a factual verbal narrative.

We recruited 61 participants (30 females, 31 males, average age = 30 years) from the Department of Geography at the University of Zurich to participate in the study. After viewing each map stimulus, participants were asked (without any time restriction) to select one of the six predefined locations shown in the maps that they predicted to be most/least affected by climate change. Finally, they indicated their trust in each stimulus type using a standardized questionnaire.

Preliminary results suggest no significant differences in participants’ overall average trust ratings across the two narrative conditions. However, participants significantly trust climate change prediction maps more when certainty information is also included, regardless of the narrative condition they were assigned to. Conversely, we found no significant difference in trust ratings between the map stimuli that contain certainty information visualized as gridded or randomly distributed dots.

These novel empirical findings stress the need to visually communicate (un)certainty information to support people’s trust in climate science and climate change forecast maps. The use of cartoon characters to emotionally engage the public in climate change communication remains to be further empirically investigated.