Public opinion polls have shown that beliefs about climate change have become increasingly polarized in the United States. A popular contemporary form of communication relevant to beliefs about climate change involves digital artifacts known as memes. The present study investigated whether memes can influence the assessment of scientific data about climate change, and whether their impact differs between political liberals and conservatives in the United States. In Study 1, we considered three hypotheses about the potential impact of memes on strongly-held politicized beliefs: 1) memes fundamentally serve social functions, and do not actually impact cognitive assessments of objective information; 2) politically incongruent memes will have a “backfire” effect; and 3) memes can indeed change assessments of scientific data about climate change, even for people with strong entering beliefs. We found evidence in support of the hypothesis that memes have the potential to change assessments of scientific information about climate change. Study 2 explored whether different partisan pages that post climate change memes elicit different emotions from their audiences, as well as how climate change is discussed in different ways by those at opposite ends of the political spectrum. Keywords: climate change, memes, metaphor, politics, beliefs, topic models

When making decisions, humans often rely on information from their social networks through a process termed social sampling. Prior work suggests that when drawing social samples, people search through their contacts in a sequential manner based on structured social categories. We examined whether the problem domain impacts how one categorizes their social contacts and which social categories they sample from. In our study, participants answered questions about the relative popularity of national parks or social media platforms, respectively associated with offline activities and online activities. Participants then provided frequency information about the number of their contacts who have engaged in these activities. Adopting a hierarchical Bayesian modeling approach, we compared two social sampling models: one defining social groups based on closeness of social relations, and one defining social groups based on contact mode. Our findings suggest that people sample from different members of their social network depending on the type of decision they are making.

While narratives have shaped cognition and cultures for centuries, digital media and online social networks have introduced new narrative phenomena. With increased narrative agency, networked groups of individuals can directly contribute and steer narratives that center our collective discussions of politics, science, and morality. We report the results of an online network experiment on narrative and hashtag generation, in which networked groups of participants interpreted a text-based narrative of a disaster event, and were incentivized to produce matching hashtags with their network neighbors. We found that network structure not only influences the emergence of dominant beliefs through coordination with network neighbors, but also impacts participants' use of causal language in their personal narratives.