The members of the Bundestag and the state parliaments have become part of an experiment.

To this day they don't know anything about it.

At the beginning of the year, a researcher from the Institute for the German Economy in Cologne sent e-mails to all 2503 representatives.

He posed as "Alexander Müller" or "Peter Schmidt", claimed to live in the respective constituency and to be insecure about a certain political issue.

Then the fictional voter asked if certain information that he had heard was true.

However, this information was wrong, classic fake news.

The dates for this fake news were not fictitious.

The year before, the researcher, his name is Matthias Diermeier, had already seen in a representative survey that voters misjudge certain issues - and that this depends on their preference for certain parties.

AfD voters overestimated the proportion of unemployed migrants, on average they believed it affected 48 percent and not 14. Green voters, on the other hand, underestimated the proportion of renewable energies in Germany.

On average, they believed it was 35 percent when it was 45 percent.

How often do MPs answer?

Diermeier took advantage of these wrong numbers. He discussed with the Ethics Council of the University of Duisburg-Essen and drafted three different inquiries. One on migration, one on renewable energies and a neutral one on the economy. It was a coincidence who received which request, but each MP only got one. So the Greens also the migration request, the AfD also the request about renewable energies. Diermeier wanted to find out: How often do MPs answer their potential voters? Did that have to do with the issues? And: Do the MPs correct the fake news - even if the fake news fits their election platform? Finally, and that was his main question: Do the right-wing populists behave differently from the other parties?

Diermeier had two hypotheses before starting the experiment: First, he believed that the subject of the query would also determine the response rate.

So that MPs of the AfD reported back more often when a voter was interested in the topic of migration.

Second, he suspected that tolerance for fake news increases the more radical an MP is about an issue.

An AfD MP, so the assumption, would let more fake news about migration go through without correcting it, a green one more about renewable energies.

AfD less close to the people than claimed

Both hypotheses were ultimately wrong. It was found that an average of half of the MPs responded to the voter request, more in the Bundestag than in the state parliaments, and there again more when an election was due. A pretty good cut compared to other states. The established parties were not very different from one another, but the AfD was an outlier: only 39 percent of the MPs reported to their potential voters. And that although they always presented themselves as the party that is particularly close to the people.

When it comes to migration, the AfD's core issue, almost 37 percent respond, less than all other parties. It was also shown that AfD MPs tolerated fake news much more often. One in three of their MPs left the false facts of the fictitious voter as they were, and on the issue of migration it was even half. In contrast, for the other parties, an average of 95 percent corrected the information. The most defensive was the SPD: Here it was 98 percent. Diermeier had suspected that there would be a tendency among the Greens to correct the request for renewable energies less strongly. That was not the case. And interestingly enough, the AfD also overtook here by a distance.

Diermeier therefore believes that the AfD has a “structurally different response behavior”.

Your answers "often completely missed the misinformation requested".

One MP wrote, for example: "The unemployment rate among migrants is the best-kept secret in Germany".

Diermeier found the number on Google with one click.

He finds the result of his experiment "refreshing".

“Otherwise you can't look into such private communication.” Except for the AfD, the answers were consistently benevolent, helpful, understanding and not very susceptible to fake news.