An article published in Nature studied over 3,300 people’s photos to determine if there were any links between facial features and political views. The authors found that less attractive and more contemptuous women are more likely to be left-leaning.

The referenced Nature article used deep learning to analyze facial photographs and found that for females, higher attractiveness scores were associated with those identified as likely conservative, which contradicts the claim that less attractive women are more left-leaning. The study also examined expressions such as contempt but did not reduce political orientation to simple attractiveness or contempt measures alone. Additional sources clarify that the concept of beauty was not rigidly defined in the study, and the claim that unattractive women are more left-wing is an oversimplification often misrepresented in secondary reports. The study's findings are complex, involving machine learning predictions rather than straightforward causal links. Other research supports that facial recognition technology can detect political orientation with some accuracy but emphasizes ethical concerns and methodological limitations. The claim as stated omits these nuances and the contradictory evidence about attractiveness and political leanings.verificat.catverificat.catNo article in Nature magazine has shown that conservative women are more...nature.comnature.comUsing deep learning to predict ideology from facial photographs: expressions,...+3 more

Overall, the original claim distills a complex scientific analysis into a simplistic statement that does not fully align with the actual research findings published in Nature.