Affective integration in climate change communication: recency effects and negativity bias in responses to climate picture sequences

Environmental psychologists have long grappled with a central dilemma in climate change communication: while visual media such as melting glaciers, raging wildfires, and pristine green forests are staples of public awareness campaigns, policymakers and communicators rarely understand how these varied images interact when viewed in succession. A groundbreaking study published in Frontiers in Psychology sheds light on this phenomenon, demonstrating that the order in which climate imagery is presented fundamentally alters public emotional responses and, consequently, their willingness to adopt sustainable habits.

Conducted by researchers Hulda Karlsson-Larsson, Ertuğrul Asutay, and Daniel Västfjäll, the research provides empirical proof that human affective experiences—and the environmental intentions driven by them—are heavily shaped by cognitive phenomena known as recency effects and negativity bias. Rather than treating visual stimuli as isolated snapshots, the study evaluates how sequences of environmental images integrate over time to shape human behavior.

Experimental Design and Methodology

To uncover how audiences process complex visual narratives, the researchers designed a within-subject online experiment involving 464 American participants recruited via Prolific in February 2024. The final analyzed sample followed rigorous exclusion criteria to filter out technical latency issues and inattention, leaving a robust participant pool with a nearly even gender split and an average age of 42.

The experiment centered on twelve distinct trials. In each trial, participants were exposed to a sequence of three specific photographic stimuli, each presented for precisely six seconds. These images represented three distinct categories:

• Consequence pictures: Depicting severe climate impacts, specifically wildfires.
• Goal pictures: Depicting positive environmental outcomes, specifically healthy, thriving forests.
• Action pictures: Depicting individual pro-environmental behaviors, including reducing plastic use, recycling, shifting to public transport, and utilizing renewable electricity.

To test the sequence hypothesis, the images were ordered using a partial Latin Square design across three distinct combinations: Wildfire-Action-Forest (WAF), Forest-Wildfire-Action (GCA), and Action-Forest-Wildfire (AGC). Following every trial, participants evaluated their current emotional valence (pleasantness) and arousal using visual analog scales ranging from 0 to 100. They also reported their perceived risk of climate change, behavioral control, impact beliefs, and immediate intentions to engage in pro-environmental behaviors.

Chronology of Findings and Chronological Insights

Data analysis revealed that while arousal, risk perception, and behavioral control remained largely impervious to the sequence of images, emotional valence was profoundly sensitive to presentation order. Participants reported the highest positive valence after viewing the Wildfire-Action-Forest sequence (mean valence of 69.24 out of 100), followed by the Forest-Wildfire-Action sequence (67.56), and the lowest positive valence after the Action-Forest-Wildfire sequence (63.13).

Using a quasi-Bayesian causal mediation analysis, the researchers found that although presentation order did not exert a direct total effect on pro-environmental intentions, it produced a powerful indirect effect mediated entirely by affective valence. When an image sequence left participants feeling less positive, their reported intentions to act pro-environmentally dropped significantly.

Hierarchical Bayesian computational modeling further unpacked these responses, confirming a pronounced recency effect. When a picture moved one step earlier in the sequence—shifting from the final position to the second position—its predictive influence on the viewer’s emotional state dropped by 29%. The final image in any given sequence consistently wielded disproportionate authority over the final emotional summary.

Furthermore, the models confirmed a strong negativity bias. Among all individual image types, wildfire pictures exerted the strongest predictive weight on affective experience, vastly overshadowing the influence of action-oriented and goal-oriented images. This reinforces the psychological principle that negative information carries greater cognitive weight than equally intense positive stimuli.

Behavioral Disparities and Self-Reported Metrics

The post-experiment data captured a detailed snapshot of the participants’ baseline engagement with sustainable habits. While a large majority of respondents recognized the importance of climate action, practical hurdles and lifestyle dependencies varied drastically across categories:

• Air Travel: 59.27% reported traveling by airplane at least once annually, highlighting a substantial target demographic for emission reduction.
• Car Travel: A staggering 73.92% of participants relied on cars daily or most of the time, aligning with findings that public transport yielded the lowest perceived behavioral control (mean score of 58.54) and lowest intention scores (60.91).
• Recycling and Plastic: In contrast, participants reported the highest feelings of behavioral control and willingness to act regarding recycling (mean intention score of 83.88).
• Electricity Sources: Nearly a third of the cohort (34.27%) remained entirely unaware of the source of their household electricity, while 37.29% relied on fossil fuels.

When predictive mixed models were deployed, perceived behavioral control and impact beliefs emerged as the dominant statistical drivers of pro-environmental intentions. Participants who felt entirely capable of performing a green behavior and firmly believed in its carbon-mitigation potential reported the highest willingness to change their habits.

Implications for Climate Communication Strategies

The practical implications of this study are profound for environmental organizations, public policy officials, and media strategists. Historically, climate communication has relied on shock value—such as endless streams of natural disasters—or overly abstract technological solutions.

The findings argue for a sophisticated, temporal approach to visual storytelling. Because the final image in a sequence disproportionately dictates the emotional takeaway, campaigns that end on a note of empowerment, action, or positive ecological recovery can buffer against the psychological paralysis often induced by catastrophic imagery.

Moreover, the successful validation of a computational predictive model means that organizations can forecast public emotional and behavioral responses based on normative stimulus ratings, bypassing the need for intrusive real-time testing. By aligning visual sequences to leverage both recency effects and strategic emotional weighting, climate communications can successfully bridge the gap between emotional resonance and genuine behavioral intent.