Artistic representations of data can help bridge the US political divide over climate change

The United States remains divided on how to address climate change despite the Inflation Reduction Act. This division is particularly evident between liberal and conservative viewpoints regarding the urgency of action and appropriate policy directions. While data graphs provide concrete evidence, they can also trigger biased processing in skeptics, reinforcing pre-existing beliefs. To bridge this gap, scientists and artists have begun incorporating visual art into data representations of climate change, aiming to engage emotions and foster a sense of shared experience. Examples include integrating line graphs of temperature increases into watercolor paintings or transforming hurricane strength charts into abstract art. The underlying assumption is that art can overcome the perceived lack of emotional connection often associated with scientific data, enhancing public engagement and awareness. However, the effectiveness of artistic visualizations compared to traditional data graphs remains a question. Artistic abstractions could potentially compromise clarity, leading to misinterpretations or disengagement. Individuals without an interest in visual art might find such a medium less appealing or effective. Despite this, many scientists recognize the potential of visual art for communication, believing it can foster emotional connections that enhance learning. Yet, empirical evidence supporting the efficacy of artistic visualizations in knowledge retention, credibility, or bridging the political divide on climate change is scarce. This study aims to address this gap by examining how artistic representations of climate data, compared to scientific graphs, affect viewers' emotions, recall of information, perceived credibility, and whether they reduce political polarization surrounding climate change.

Existing research highlights the challenges of communicating climate change using purely data-driven approaches. Studies show that individuals tend to selectively interpret data to confirm their pre-existing beliefs, a phenomenon known as motivated reasoning. This is particularly pronounced in politically charged issues like climate change. For example, studies have shown that liberals and conservatives focus on different aspects of the same climate data graph, depending on whether the data supports or contradicts their existing beliefs. In contrast, artistic representations of data offer a potential alternative by tapping into emotions. Research suggests that anxiety, for example, can encourage consideration of opposing viewpoints and a willingness to compromise. Since climate art can evoke such emotions, it might encourage reassessment of conflicting data and reduce the perceived distance to the issue. Although anecdotal evidence exists suggesting this, a lack of empirical support remains. Additionally, prompting reflection on the visuals, encouraging thoughtful evaluation rather than biased reasoning, may further enhance their effectiveness.

This study employed a controlled online experiment with two non-probability samples of U.S. adults. The first sample (N=319) was used to compare the emotional impact, information recall, and perceived credibility of artistic visualizations versus data graphs. The second sample (N=352) investigated the role of reflective prompts on mitigating political polarization. The experiment utilized an artwork titled *Summer Heat*, 2020, by Diane Burko, which integrates a simplified version of the Keeling curve (showing atmospheric CO2 levels) with visual elements like melted glaciers and a heatwave map of Europe. Different versions of the artwork and corresponding data graphs were shown to participants, including detailed and simplified Keeling curve versions. Participants viewed standalone images to gauge their emotional responses before seeing mockup Instagram posts containing the same image with a caption. Measures included emotional responses (positive and negative), information recall (via multiple-choice questions), perceived credibility of the posts, and perceived relevance of climate change. Political leaning was assessed by combining self-reported party identification and political ideology. The experimental design involved a 2 (artistic representation vs. data graph) x 2 (detailed graph vs. simplified graph) factorial structure. For the first sample, participants were asked to reflect on the meaning and emotions evoked by the visuals. The second sample received only the mockup Instagram posts without any reflective prompt. Analysis of variance (ANOVA) and analysis of covariance (ANCOVA) were used to analyze the data, examining the effects of visual format, graph detail, art interest, and political leaning on various outcome measures.

The study found that artistic representations elicited stronger positive emotions than data graphs, although individuals with higher art interest reported more negative emotions from the artistic visualizations than those with lower interest. There was no significant difference in perceived credibility or information recall between the artistic and graphical representations. Importantly, when participants were prompted to reflect on the visuals, artistic visualizations mitigated political polarization in their perceived relevance of climate change, an effect not observed when reflection was not prompted. The detailed Keeling curve elicited stronger negative emotions compared to the simplified one. The original art piece elicited the highest positive emotions, while the edited data graph elicited the lowest positive and negative emotions.

The findings suggest that artistic visualizations can be a more effective tool for communicating the relevance of climate change to diverse audiences than traditional data graphs, particularly when reflection is encouraged. The increased positive emotional responses to the art, even when depicting negative aspects of climate change, could be attributed to the aesthetic experience and the distanced perspective adopted by participants. However, the study's specific artwork and sample population might limit the generalizability of the findings. It is also important to acknowledge that the positive effects may not be solely attributable to the artistic format, as the painting included additional visual elements not present in the data graphs. Future research needs to address these limitations and examine different types of visual art and diverse audiences. The lack of difference in perceived credibility and recall between the art and data graphs suggests that concerns about artistic representations’ lack of scientific rigor might be unfounded in certain contexts, at least when presented in a familiar context such as an Instagram post.

This study provides evidence that artistic representations of climate data can enhance emotional engagement and mitigate political polarization, particularly when paired with reflective prompts. Future research should investigate the generalizability of these findings across different types of art, audiences, and cultures, also exploring how the interplay between art and science can improve climate communication on platforms like Instagram and TikTok. It's crucial to consider how to balance emotional engagement with maintaining scientific accuracy and objectivity, and to avoid a simplistic focus on depicting loss and problems.

The use of a single artwork and the focus on U.S. audiences limit the generalizability of the findings. The positive effects of the artistic visualizations may be partly due to the additional visual elements in the painting, not solely the artistic format itself. The study also relied on self-reported measures, which are susceptible to biases. Further research with broader samples, different artistic styles, and more rigorous experimental controls is needed.