Talking to the other side

Trying to reason with climate change deniers or other types of sceptics is often considered the wrong strategy - but here is another side to the story

97% of the world’s top climate scientists maintain the evidence-informed view that the Earth is warming due to human activity.  Despite this, recent Ipsos polling found that 25% agreed that “Climate change exists but it is not caused by human activity” and 7% agreed with the statement “There is no climate change”. 

At times it can seems as if no matter how much evidence is given to people about the existential dangers we face, there is a refusal to shift from their beliefs.  It seems to be the precise opposite of a rational outlook to the world.   Even among people that understand human climate change is a given, there often seems a lack of ‘follow-though’ in terms of changing behaviour - in the same study, Ipsos found one third of people consider a solution will come through technological innovation (there is no known solution currently). 

To unpack what is going on, this week we are drawing on a paper by Bence Bago, David G. Rand and Gordon Pennycook - who have drawn some very powerful conclusions about how best to engage with those on the ‘other side of the fence’.

Political partisanship

Are we destined to be irrational and incapable of absorbing information to make informed decisions?  One of the most commonly cited reasons for climate change denial is political partisanship. In the US in particular, people on the right of the political spectrum are more likely to consider climate change is either a hoax or is not caused by human activities.  In addition, people with superior numerical ability and cognitive sophistication demonstrate increased political-allegiance differences in climate change beliefs, rather than higher agreement with the scientific consensus.  Hence, stronger cognitive capability seems to not protect against climate misperceptions; but instead, it may reinforce views that support one’s political identities.

Identity-protective cognition

Dan Kahan’s identity-protective cognition framework is perhaps the most frequently cited explanation for this.  This can be understood from the perspective of the dual process theory which differentiates two types of reasoning process: intuition (System 1) and deliberation (System 2).   It is well known that intuition is a low-effort, fast, ‘automatic’, response to stimuli, whilst deliberation requires more effort and is a time-consuming process. The identity-protective cognition framework considers that cognitive abilities are associated with greater polarization because deliberation enables motivated reasoning: when we find new evidence, participating in deliberation allows the discrediting of evidence if it is not consistent with identity and political orientation.  Given there are significant political differences concerning what evidence is considered credible, this leads to polarization in beliefs.

Translating this into the language of dual process theory, deliberative reasoning processes are sparked to rationalize identity-consistent intuitive instincts. On the specific topic of climate change, this means that deliberation leads Republicans to discard evidence in favour of climate change (to protect their partisan identity), while deliberation means Democrats reject evidence challenging climate change.  If more cognitively sophisticated people engage in more deliberation, then we would expect them to be better at aligning their evaluation of evidence about climate change with their respective political allegiances.

This theory has huge practical importance because, if it is indeed the case, strategies such as educating people, or making them deliberate, will not be effective against climate change denial.  Similarly, this is used as an argument for not engaging with other forms of misinformation and Conspiracy Theories – the notion here is that challenging these beliefs will not work and may even serve to increase partisan differences.

Are we really blinded by partisanship?

However, in their recent paper, Bence Bago, David G. Rand and Gordon Pennycook challenge this account and make the case for educating people and participating in the debate with people that hold different views to our own.  Their research suggests that people are not in fact ‘blinded by partisanship’ but are instead making a good-faith effort to form accurate beliefs.

The case for this is based on challenging two significant assumptions within the identity-protective cognition framework. First, the set out that political identity is confounded with - but is clearly distinct from - prior beliefs about climate change. Democrats are much more likely to believe that climate change is caused by human activity than Republicans (and vice versa), indicating that partisanship and prior factual beliefs are correlated. However, prior beliefs are not entirely consistent with political affiliation.   Although many Republicans believe that human activity contributes to global warming, at least to some degree, many do not.  Similarly, not all Democrats reject this notion. 

This is a somewhat problematic finding for the identity-protective cognition framework, because if our position on climate change is driven by politically motivated reasoning then we must be able to de-facto demonstrate that partisanship influences information evaluation, above and beyond the impact of prior beliefs about the specific topic. Of course, our prior beliefs may be influenced by a range of other factors such as family environment or life experiences and it may be these that also influence partisanship. Which means that effects driven by prior beliefs do not provide direct or compelling evidence for politically motivated reasoning.

In addition, past research on motivated System 2 reasoning has relied upon correlating individual differences in cognitive sophistication with degree of partisan differences on politicized issues.  However, again there is some confounding taking place – people scoring higher on cognitive sophistication scales are better at deliberation than people scoring lower on these scales, but importantly they also tend to differ in many other aspects. We cannot ignore the wider way in which people with lower (or higher) cognitive sophistication may have different social, cultural and political influences on their reasoning. 

In a series of ingenious experiments, Bago and colleagues accounted for prior beliefs (which are clearly different from partisanship per se), and experimentally manipulated extent of reasoning (using cognitive load and time pressure) rather than relying on correlations with individual differences in reasoning.

Their results are hugely important: they show that once these factors are accounted for, deliberation does not in fact increase reliance on partisan consistency - hence reasoning did not lead most Republicans to reject climate change or Democrats to reject evidence challenging climate change. 

Evaluating new evidence

Instead, they found that deliberation increased the consistency between evaluation of new climate information and pre-existing beliefs about climate change. They point out that while evaluating new evidence in light of prior beliefs is often called “confirmation bias”, it can of course  be entirely reasonable to conclude that the information source is unreliable rather than that the accumulation of all your prior knowledge is incorrect. As they point out, if a stranger tells you that they were abducted by aliens, is it not irrational to conclude that that information is probably unreliable! Bayesian style analysis demonstrates that we always start with priors in which we assess new information but that it can be updated - there is potential for change.

This suggests we need to take a fundamentally different approach to tackling misinformation and conspiracy theories.  If people are engaging in ‘good-faith efforts’ to consider information accurately and relying on their prior beliefs to guide such judgments, then educational interventions could - in the long run - move people’s priors and increase agreement with the scientific consensus.

Conclusions

Although we have looked at this issue relating to climate denial, the implications are far wider.  The widespread acceptance of the identity-protective cognition framework has arguably led to a policy of not engaging with a range of misinformation or conspiracy theorising.  The problem here, as we have pointed out previously, is that this can lead to information vacuums where people simply do not have any real opportunity to deliberate and have their prior beliefs challenged. 

Of course, as the authors point out, simply getting people to think more carefully will not in itself make people more likely to believe in the threat of climate change. But if people are indeed engaging on a good-faith basis to assess new information correctly, and relying on their prior beliefs to drive their judgments, then education could move people’s beliefs and improve agreement with the scientific consensus.

Supporting this is the way in which studies are showing the impact on the general public’s attitudes from the way the media frames climate change. A growing body of experimental research has demonstrated the way in which different frames in climate communication can affect attitudes and behaviour.

The point is that having the debate matters – we do not need to accept the bleak conclusion that we are destined to remain within the limits of our political, value-based allegiances.