https://www.theatlantic.com/ideas/archive/2025/01/amanda-knox-murder-slander-trial/681457/
The lie that I was at the house when the crime occurred led to repeated instances of what is called forensic confirmation bias. The lie colored the collection and analysis of all the other evidence. It led police to ignore exonerating evidence, such as my lack of a motive or any history of violence or mental illness, my alibi, and the virtual impossibility of participating in such a brutal murder without leaving a single trace of DNA in the room. And it led them to distort and magnify the significance of trivial evidence, such as the fact that my DNA had been found in the bathroom where Guede attempted to clean off Meredith’s blood. Of course my DNA was found there—that was my bathroom too.
This phenomenon has been demonstrated by the cognitive neuroscientist Itiel Dror. In a 2006 study, he gave six fingerprint experts pairs of prints that, unbeknownst to them, they had previously judged in their own casework as matching or not. The experts were given some made-up context for each pair of prints. For the nonmatches, they were told that the suspect had confessed to the crime; for the matches, they were told that the suspect had an ironclad alibi. This fictional information resulted in two-thirds of the experts changing some of their original judgments. Believing that a suspect confessed alters the supposed objectivity of scientific experts. Dror went on to show a similar effect in other forensic domains, including DNA analysis and forensic pathology.