The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Political Ignorance

Survey Shows Most Extreme Opponents of GMO Foods Know the Least—Yet Think They Know the Most

The result is consistent with lots of other evidence of widespread ignorance and bias influencing public opinion on political and scientific issues.


A recent study finds that the most extreme opponents of GMO food have the lowest levels of relevant scientific knowledge, but also tend to believe they know much more than they actually do. The Guardian has a helpful summary of the results:

The most extreme opponents of genetically modified foods know the least about science but believe they know the most, researchers have found.

The findings from public surveys in the US, France and Germany suggest that rather than being a barrier to the possession of strongly held views, ignorance of the matter at hand might better be described as a fuel.

"This is part and parcel of the psychology of extremism," said Philip Fernbach, a researcher at the University of Colorado and co-author of the 2017 book The Knowledge Illusion. "To maintain these strong counter-scientific consensus views, you kind of have to have a lack of knowledge."

Fernbach and others analysed surveys completed by nationally representative samples of the US, French and German public. Those who took part were asked about their attitudes to GM foods and given instructions on how to judge their understanding of the topic. Next, they completed a scientific literacy test. Among the statements the participants had to wrestle with were: "Ordinary tomatoes do not have genes, whereas genetically modified tomatoes do" (false), and "the oxygen we breathe comes from plants" (true).

The results from more than 2,500 respondents revealed the curious trend. "What we found is that as the extremity of opposition increased, objective knowledge went down, but self-assessed knowledge went up," Fernbach said.

The study itself is available here. As the authors point out, scientists overwhelmingly conclude that GMO foods are no more risky than "natural" ones, yet many in the general public continue to believe they should be severely restricted or even banned.

This is far from the first study to show widespread public ignorance about either GMO foods specifically or scientific and public policy issues, generally. For example, surveys conducted in the US in 2014-15 and 2016 found that some 80% of Americans say there should be mandatory labeling of foods containing DNA (despite the fact that DNA is the basic building block of all life on Earth, and nearly all our food contains it).

Some of this is simply the result of what scholars call "rational ignorance": Most people have little incentive to spend much time learning about government, public policy, and policy-relevant science, because the chance that their votes will make a different to policy outcomes is infinitesimally small. Many people don't have time to study the science of GMOs. Thus, they simply do not know that GMO foods are no more dangerous than "organic" ones, and that most organic foods are themselves the result of centuries of genetic manipulation by humans.

But that does not, by itself, explain why people who most oppose GMOs are not only the most likely to be ignorant, but also unusually confident about the extent of their knowledge. That has to do with bias, not simply ignorance. Studies repeatedly show that many people—especially those with strong views—are highly biased in their evaluation of political information, often acting as "political fans" cheering on their preferred party or ideology rather than evaluating information objectively. That is true of ordinary voters, activists, and even government officials. The most committed partisans and ideologues also tend to have the strongest biases, and thus are probably the most likely to overestimate the extent of their knowledge and understanding.

While such biases occur in many areas of life, they are particularly strong when it comes to political disputes. Both politicians and ordinary citizens are much more biased in evaluating information on political issues than otherwise similar data on nonpolitical ones. A big part of the reason why is that we have stronger incentives to try to keep our biases in check when we make decisions in the private sector than when we vote or otherwise influence government policy. In the latter situations, our actions either have very little chance of making a difference (if we are voters) or are likely to have their biggest impact on other people with whom we have few ties (if we are policymakers). That may help explain why many people hold ill-informed and foolish views on policy issues related to GMO foods, yet also continue to happily eat them.

One possible solution to these sorts of problems is for voters to defer more to scientific experts, at least on technical issues such as the risks posed by GMO foods. Such deference may often be useful. But it is also often difficult to figure out when it is appropriate, and when not. Figuring out who is a real expert and where the limits of their expertise are may itself require considerable knowledge and insight. Moreover, the same flawed incentives that lead voters to make biased assessments of evidence may also cause them to dismiss the views of experts whose findings cut against the voters' own preconceptions.

Another standard strategy for overcoming public ignorance about science is for government to disseminate accurate information about risks (or, in the case, of GMO foods, the lack thereof). Such government warnings can be useful, as in the famous case of the Surgeon General's warning on smoking. But they also have risks of their own. Governments have a long history of spreading inaccurate information about various types of risks, and their incentives to do so are often exacerbated by the very same public ignorance that government-produced information is supposed to cure. While it is unlikely that the US government will start mandating warning labels on food containing DNA, they do have a track record of requiring other misleading warnings, that often impose real costs on society.

In sum, we have good reason to be wary of the influence of public ignorance on government policy—especially when it comes to scientific issues. But there is also good reason for skepticism about the standard solutions usually proposed to fix that problem.