More than scientists: How message and messenger attributes influence viewers’ climate change intentions

PLOS One | https://doi.org/10.1371/journal.pone.0331672 September 10, 2025 6 / 20 chimneys and factories. Höijer [72] examined representations of climate change in a Swedish newspaper and on Swedish television, and argued that imagery may influence public engagement by anchoring the issue in emotions of fear, hope, guilt, compassion and nostalgia [72]. To our knowledge, no previous study has investigated whether imagery featured in a public corpus of climate change online video messages may predict viewer responses. Current study The current study uses LIWC and visual coding to investigate the motivational impact of messages from the More than Scientists website (https://morethanscientists.org), which contains short videos of climate scientists explaining why cli- mate change is personally relevant to them. In particular, we were interested in determining which specific linguistic and messenger attributes within the video were most effective in motivating viewers to take action against climate change, and whether these effects varied as function of viewers’ political orientation (i.e., right or left leaning). Based on previous research on linguistic framing, we predicted videos with greater analytic thinking, clout, authenticity, and positive emotional tone would elicit stronger climate change motivations in viewers than messages that scored lower on these variables. We further predicted that messages that emphasized affiliation (e.g., collectivism and community) and climate-change risk

would elicit stronger motivation to act in left-leaning viewers, and the messages that emphasized power, achievement, reward would elicit stronger motivations for right-leaning viewers [76]. Finally, we also explored the impact of five messen- ger attributes on viewers’ climate motivations: age, sex, perceived attractiveness, presenting in natural setting or not, and presenting with family members or not. We predicted that more attractive messengers would elicit stronger motivations to act in viewers than less attractive messengers. Given the inconsistency in the literature about the effects of other messen- ger variables on persuasion and motivation, we did not make specific hypotheses about the other messenger attributes. Method Data were gathered by a Qualtrics web-based survey administered after human ethics approval. The study was approved by the University of New England Human Research Ethics Committee and complied with the ethical standards outlined in the Declaration of Helsinki. All participants provided written, informed consent prior to study participation. The dataset can be accessed through the Open Science Framework Repository. Participants Participants were recruited via Qualtrics from an online panel of individuals residing in demographically and geograph- ically diverse areas of the USA. To be eligible to participate, respondents had to be US residents and aged 18 years or older. Participants understood that the project aimed to improve climate-change communication and that the survey would

take approximately 30 min. A total of 581 US residents (50% male) fully participated in the study. Additional details about the sample are provided in the results section. Design and procedure Respondents first completed measures assessing demographics, political orientation, and personal values. They next viewed a random sample of 6 (of 41) climate change videos selected from the More than Scientists website (https://more- thanscientists.org). To avoid order effects, the videos were presented in a random order for each participant. After viewing each video, respondents rated the attractiveness of the main speaker(s), how the videos made them feel, and the extent to which the videos motivated them to seek out more information about and take action against climate change. Video selection. At the time of selection, the More than Scientists website contained 184 video messages from a total of 44 presenters. We limited our selection to no more than one video per presenter. We randomly selected 44 videos, each by a unique presenter. Of these, 3 were excluded for exceeding 3 min (a limit set to ensure the overall length of