Another problem for gauging the reliability of canines is the bias of their handlers. In a 2011 study published in Animal Cognition, the sniffing accuracy of 18 trained dogs was tested over two days. The dogs’ handlers had experience in drug and bomb detection. They were falsely told that the scents of drugs and bombs had been planted in rooms of the church where the test took place and that some of these points were marked by a piece of red paper. Recorders, who weren’t told the purpose of the study followed the dogs to write down where they raised an alert. Out of 144 searches, 123—involving 17 of the 18 dogs—raised a false alert. Most strikingly, the handlers were most likely to claim their dogs picked up a nonexistent scent when they saw a piece of red paper.