New evidence for the Gateway (mis)Belief Model: Perceived consensus shapes climate beliefs

A new study published in the British Journal of Psychology has found that misinformation about climate change can reduce people’s perceptions of scientific consensus, which in turn tends to lower belief in human-caused climate change and diminish public support for action. The researchers tested what they call the Gateway (mis)Belief Model, or GmBM, and found that even brief exposure to misinformation weakened trust in climate science and, in some cases, reduced willingness to take collective action.

The study was conducted by researchers from the University of Cambridge and the University of Oxford. Their work builds on earlier research into how public beliefs about climate change are shaped by perceived scientific agreement, a key factor that has been shown to influence attitudes, emotions, and policy preferences.

The scientific consensus on climate change is strong, with over 97% of climate scientists agreeing that human activities are the primary driver of global warming. Yet many people, especially in the United States, continue to underestimate this level of agreement. The so-called “consensus gap” has been identified as a factor that contributes to climate skepticism and resistance to environmental policy.

Previous research has shown that highlighting the scientific consensus can improve public understanding and increase support for mitigation efforts. This is the basis for the Gateway Belief Model (GBM), which describes how people use cues about expert agreement to form opinions about complex scientific topics. However, while many studies have tested the positive impact of consensus messaging, fewer have looked at what happens when misinformation distorts perceptions of agreement. The new study aimed to address this gap by proposing and testing a version of the GBM that includes misinformation as a causal factor.

“We wanted to explore the process by which misinformation can disrupt public discourse around important scientific issues,” said study author Sander van der Linden, a professor of social psychology at the University of Cambridge and author of Foolproof: Why Misinformation Infects Our Minds and How to Build Immunity. “One of the main mechanisms that disinfomers use is attacking the scientific consensus on important issues like climate change and vaccination. We wanted to explore the downstream impacts of how casting doubt about the scientific consensus on climate change can impact public opinion about the issue.”

The researchers reanalyzed data from two previous experiments involving U.S. participants. In both studies, people were shown misinformation designed to reduce their perception of scientific consensus. This included a fabricated petition suggesting that thousands of scientists rejected human-caused climate change—a widely debunked claim that has been used in public discourse for years.

In the first experiment, 207 participants were randomly assigned to view either a consensus message or a combination of consensus followed by misinformation. Beliefs were measured at three time points: before any messages, after the consensus message, and after the misinformation. The researchers found that exposure to misinformation significantly lowered participants’ estimates of scientific consensus and, to a lesser extent, their belief that climate change is happening.

Although the misinformation did not significantly affect every belief tested, structural modeling suggested a pattern of indirect influence. Lower perceptions of consensus tended to reduce belief in human causation, which in turn predicted less worry about climate change and weaker support for public action. However, these downstream effects varied in strength and significance. The overall model fit the data reasonably well, but not all relationships aligned with previous findings.

The second experiment involved a much larger sample of 755 participants and provided a stronger test of the GmBM. This time, participants were randomly assigned to receive either the misinformation message or a neutral task. Once again, the misinformation significantly reduced perceived scientific consensus. It also lowered belief in human-caused climate change, concern about the issue, and willingness to support mitigation measures.

In this second study, the model showed more consistent results. Changes in perceived consensus were associated with changes in key beliefs, and misinformation had both direct and indirect effects on support for action. In particular, reduced belief in human causation and decreased worry about climate change both contributed to lower support for mitigation efforts. The researchers noted that even among a sample that skewed liberal and highly educated—groups generally more accepting of climate science—the effects of misinformation were still observable.

To confirm their findings, the researchers conducted an internal meta-analysis across both studies and an additional dataset. This analysis supported the central idea of the GmBM: misinformation erodes perceived scientific consensus, and this erosion trickles down through related beliefs to reduce support for climate action.

The results indicate “that misinformation can have real-world consequences, such as lowering support for public action on climate change by changing people’s beliefs on the issue,” van der Linden told PsyPost. “One of the main tactics used is to cast doubt on the near-unanimous consensus among climate scientists that humans are causing global warming – e.g., by using fake experts, people should be on the lookout for such disinformation.”

“Our sample contained many people with a fairly liberal ideology,” he added. “The fact that even audiences who already strongly support the science on climate change can be derailed by misinformation about it was somewhat surprising (and concerning).”

There are a few limitations to consider. One issue is that the two experiments used different designs and measures, which makes it more difficult to compare results directly. In the first experiment, for instance, participants in both groups were initially shown a consensus message before any misinformation, whereas in the second experiment the control group received no message about climate change. This may have dampened the observed effects in the first study.

“When people learn about the strong scientific consensus on important issues such as climate change, it changes their beliefs and opinions about the matter, which in turn predict support for societal action,” van der Linden said. “This is why we refer to perceived scientific consensus as a ‘gateway’ belief. What was slightly different here is that misinformation about the scientific consensus doesn’t work in exactly the same (linear) way insofar it is less predictable what beliefs or emotions it might change – though what is clear is that in the end, the bottom line is the same: it undermines people’s support for action. One other limitation is that our samples were not representative of the population and we only tested one kind of misinformation (casting doubt about the scientific consensus) so it is unclear to what extent these results generalize to other types of misinformation.”

Another open question is whether certain individual traits make people more or less susceptible to misinformation. One candidate is metacognitive confidence—the degree to which people feel sure about their own beliefs. Previous work suggests that people with lower confidence may be more open to changing their views, while those with high confidence may resist updates regardless of evidence. Exploring these and other potential moderators could improve efforts to protect the public from misinformation.

The study, “The gateway (mis)belief model: How misinformation impacts perceptions of scientific consensus and attitudes towards climate change,” was authoed by Hannah Timna Logemann, Jacob B. Rode, Rakoen Maertens, and Sander van der Linden.