In the United States, a poll by the Pew Research Center found that 88 per cent of scientists thought GM foods were safe to eat, while only 37 per cent of laypeople thought so, the largest gap for any of the issues tested.
The report, published in Nature Human Behaviour, said that while research on opposition to GM foods had primarily focused on what people actually knew, it was also important to consider what they thought they knew.
Self-assessed knowledge was a strong predictor of attitudes, and people tended to be poor judges of how much they knew.
They often suffered from an illusion of knowledge, thinking that they understood everything from common household objects to complex social policies better than they did.
This was why people's sense of understanding decreased when they tried to generate explanations, and why novices were poorer at evaluating their talents than experts.
The researchers hypothesised that extremists would display low objective knowledge but high subjective knowledge, and that the gap between the two would grow with extremity.
Participants were asked to rate their opposition to GM foods and how concerned they were, and to rate their understanding of GM foods (self-assessed knowledge). They were then tested on scientific literacy.
The researchers found that as extremity of opposition to GM foods increased, objective knowledge of science and genetics decreased, but self-assessed knowledge increased.