�����鶹��Ƶ

The language that elected members of the US Congress use in debate increasingly includes words such as “phony” and “doubt” over words such as “proof” and “reason”.

This linguistic trend away from evidence in favour of intuition was revealed in an artificial intelligence analysis of millions of congressional speech transcripts. It also coincides with both greater political polarisation in Congress and a decline in the number of laws that get enacted through Congress, says at the University of Bristol in the UK.

“We can think that truth is something we can achieve based on analysis of evidence, or we can think of it as the result of intuition or ‘gut feeling’,” says Lewandowsky. “Those notions of honesty and truth are expressed in how we use everyday language.”

Adapting an off-the-shelf AI language model, Lewandowsky and his colleagues analysed the words used in transcripts of 8 million congressional speeches given between 1879 and 2022. They compared the congressional text wording with representative groups of 49 evidence-based keywords such as “logic” and “reason” and 35 intuition-based keywords such as “suspicion” and “gut.” They then computed a score showing if a given congressional speech leaned toward evidence or intuition.

They found that Congress has increasingly favoured intuition-based language over evidence-based language since the 1970s. Before then, intuition-based language also spiked in speeches made during the Gilded Age in the years 1899 to 1901 and the Great Depression in the years 1933 to 1935.

“The findings fit with other impressions of rising anti-intellectualism, populism and a rejection of scientific experts in recent decades,” says at New York University.

A particular strength of the research is that it evaluates the context in which words appear instead of just tracking their frequency, says at Eötvös Loránd University in Hungary. “These models can capture deeper, often subtle associations between words, even reflecting cultural meanings and social relations,” she says.

Next, Lewandowsky and his colleagues plan to look for similar language shifts for individual lawmakers both in congressional speeches and in social media posts. They are also looking to compare similar trends among other parliaments throughout history, including legislator speeches from Italy and Germany.