Some people can look at an old photograph of a person — say, of your grandmother in elementary school — and tell whether the person is today alive or dead, according to a paper published last spring.
If that sounds too weird to be true…well, it might be. The journal editors have retracted the paper for not having enough evidence to back up its claims, despite the authors’ objections.
Here’s the retraction notice for “Prediction of Mortality Based on Facial Characteristics,” published in Frontiers in Human Neuroscience:
Two former researchers at Duke University at the center of a lawsuit by a whistleblower to recoup millions in federal funding have lost yet another paper.
Researchers have retracted their third paper due to missing original data, following an investigation at their former institution in New York. 



In a massive cleanup, Springer and BioMed Central announced today they are retracting 58 papers for several reasons, including manipulation of the peer-review process and inappropriately allocating authorship.
