Abstract
Despite the overwhelming scientific evidence, many people still remain skeptical about climate change and refuse to take actions to mitigate the adverse impacts of climate change. Here we propose a motivated attention framework to explain public skepticism and inaction. We propose that personal motivations (e.g., political orientation) shape attention to climate change information, which alters the perception of climate evidence and shifts subsequent actions to mitigate climate change. In Study 1 (N=700), participants viewed a graph representing the annual global temperature change from 1880 to 2014 and estimated the average temperature change. We found that participants gave a higher estimate when the data were framed as global temperature than when the temperature label was removed (in a neutral frame). Furthermore, political orientation predicted participants' estimation in that conservatives under-estimated the temperature change compared to liberals. In Study 2 (N=214), we eyetracked participants' gaze when they viewed the temperature graph, and found that liberals focused more on the increasing phase of the curve, which was associated with a higher estimation of the global temperature change. However, conservatives focused more on the flat phase of the curve, which was associated with a lower temperature estimation. In Study 3 (N=104), we found that the total amount of gaze fixations of liberal participants on the graph predicted their willingness to donate to environmental organizations and their donation amount. These results provide initial evidence for the motivated attention framework, highlighting an attentional divide between liberals and conservatives in the perception of climate data, which can further explain their polarizing beliefs about climate change, as well as the actions these individuals take to address climate change. The current findings have important implications for the visualization of climate data and communication of climate science to different socio-political groups.
Meeting abstract presented at VSS 2018