British scientists have found that liberals and conservatives use different parts of their brain when they respond to risk.

Researchers working with American colleagues were able to predict if people voted for the Democrats or Republicans with 83 per cent accuracy just by studying their brain activity.

A group of 82 US volunteers were asked to play a simple gambling game while their brains were scanned.

All were registered members of political parties whose affiliations could be obtained from public records.

Republicans and democrats were no different in terms of the risks they took during the game, the scientists found.

But there was a marked contrast in the way their brains dealt with risk-taking.

Democrats showed significantly greater activity in the left insula, a brain region associated with social and self-awareness. Republicans had a more active right amygdala, a region involved in defensive fight-or-flight responses.

Brain activity in these two regions alone was enough to predict whether a participant was a Democrat or Republican with 82.9 per cent accuracy.

The scans provided a better clue to political allegiance than the party loyalty of an individual’s parents, which was only 69.5 per cent accurate. They also outperformed forecasts based on genetic differences.

The research was published in the online journal Public Library of Science ONE.

Co-author Darren Schreiber, from the University of Exeter, said: “Although genetics have been shown to contribute to differences in political ideol-ogy and strength of party politics, the portion of variation in political affiliation explained by activity in the amygdala and insula is significantly larger, suggesting that affiliating with a political party and engaging in a partisan environment may alter the brain, above and beyond the effect of heredity.”

The results could pave the way for new research on voter behaviour, yielding better understanding of the differences in how liberals and conservatives think, Schreiber said.

He added: “The ability to accurately predict party politics using only brain activity while gambling suggests that investigating basic neural differences between voters may provide us with more powerful insights than the traditional tools of political science.”