By Emma Young
How do we change beliefs that are contrary to the scientific consensus? Given that such misperceptions can be harmful to the believers, their families, and even to broader society, research in this area is vital. Now Aart van Stekelenburg at Radboud University and colleagues report preliminary but promising work finding that a brief training exercise on the value of scientific consensus, and how to look for it, can help. Their paper in Psychological Science suggests that this could be a more effective approach than just communicating what the scientific consensus is — at least, for some false beliefs.
For one of their studies, the team recruited 854 US-based participants who all believed that genetically engineered food is worse for our health than non-GM food. Some participants were given an infographic that explained how a scientific consensus develops and why it’s very useful in deciding whether or not a new claim is likely to be true. It also provided a three-step guide to evaluating a new claim (as might be found in a news article, say):
- Look for a statement indicating consensus
- Check the source making the consensus statement
- Evaluate the expertise of the consensus
The training also included a short practice session, with feedback on how to apply these steps.
These participants, plus another group who were not given this three-step guide, read a 2017 news article from Science about a new fungus-resistant, genetically engineered banana. For some participants, the article mentioned that a survey had found that 92% of “working PhD biomedical scientists” said that GM food was just as safe to eat as non-GM food. It did, then, communicate the scientific consensus on this topic — and overall, participants who saw this statement felt more positive about GM food than those who did not. But importantly, those in the training group showed even bigger shifts in their beliefs towards the consensus that those who hadn’t received the training. The team argues that the findings represent “extreme evidence in favor of boosting consensus reasoning”, to help correct misperceptions.
However, when the team ran a very similar study on climate change sceptics, they did not get this result; the consensus reasoning training did not change these participants’ belief that current climate change is not primarily down to human action.
There are a few possible reasons for the different results, the team thinks. Trust in climate scientists in the US is relatively low; if you don’t trust the scientists in the first place, you’re presumably less likely to be interested in or swayed by any consensus. Also, the scientific consensus on climate change has been widely highlighted in regular and social media. So it’s possible that this group already knew all about the consensus, and held their beliefs anyway — while perhaps the GM food group started out less aware of the consensus that genetically engineered food is safe, and so were more responsive to the messaging, the team suggests.
Overall, though, there are arguments in favour of boosting consensus reasoning in any case, the team writes. One is this: rather than telling someone what they should believe, the training is designed to “empower individuals to be able to understand and make the best use of the available information regarding a scientific consensus”.
Clearly, this is preliminary research, and the results were mixed. But as what seems to be the first experimental work to test messaging about scientific consensuses among people who hold contrary beliefs, it’s a useful step forwards.