Once an electoral candidate is in the lead, the public views them as the winner. If the candidate then loses, fraud seems a likely explanation, and even the winner’s supporters might be inclined to be suspicious about the final result. This is because of the cumulative redundancy bias (CRB). The CRB is a cognitive effect that makes it difficult for us to ignore information that we have already processed, even if the previously processed information is redundant to the most recent observation. “This effect plays a major role in the communication of election results in the USA in particular,” explains Dr. Moritz Ingendahl from the Social Cognition Lab at Ruhr University Bochum with regard to the presidential election in 2020. The researchers report their findings in the journal Psychological Science from July 24, 2025.

During the presidential election in 2020, Donald Trump was initially ahead of his competitor Joe Biden in many states, even after most votes had already been counted. However, towards the end, the margins became closer, and eventually Joe Biden took the lead. “STOP THE COUNT”, Donald Trump demanded on Twitter/X once he began to fall behind his opponent. Many of his supporters backed his accusation that the election had been rigged, resulting in the storm on the Capitol in January of 2021. A few years later, one third of all Americans still believe that the election had been fixed.

Trump’s fraud allegations were likely fueled by the cumulative redundancy bias, as shown by the team in Bochum working with Dr. André Vaz, Dr. Moritz Ingendahl, and Dr. Hans Alves in seven studies. During the summer of 2024, the researchers conducted online studies with about 200 participants each from the United Kingdom and the United States. They used a mixture of fictional and real interim and final results of vote counts for these studies.

The research team showed the participants the interim and end results of a vote count in which one candidate first had a clear early lead over the other, but later lost the lead and the election. “The subjects had a more positive view of the initial favorite, even if they eventually lost the election,” reports Ingendahl. “The winner was viewed more poorly if they took the lead late in the count.” When the researchers presented the same count results but in reverse chronological order, the participants’ views flipped.

In one follow-up study, the researchers took the same approach but let the participants know after the count was final that there were rumors of possible election fraud. “As expected, the participants found it more likely that fraud had been committed during the vote if the winner took the lead late in the count,” says Ingendahl.

If the researchers confronted the subjects with the rumor of fraud during the count, the participants thought it was likely that the fraud had been committed to the benefit of the now leading candidate. “This perception was not affected by the candidate one supports,” Ingendahl emphasizes. In one study, the researchers presented American participants with the real vote results from the state of Georgia in chronologically correct or in reverse order. In 2020, Joe Biden took the lead here very late and won the state after trailing behind Donald Trump for much of the count. “In our study, Democrats and Republicans were equally likely to be influenced by the interim vote count results, and whether they made either Donald Trump or Joe Biden have an early lead in the vote counts,” says Ingendahl.

Reporting on electoral results is thus proven to impact trust in the election itself. “False beliefs concerning illegitimate elections could be combated by only announcing the results after all votes have been counted,” the researchers summarize. They also recommend more transparency for the public regarding the factors that lead to certain developments in the vote count. Better prediction algorithms could also prevent the announcement of interim results that do not align with the final result.