How would you feel if all of a sudden you were arrested and declared the prime suspect of a terrorist attack? You are accused of being involved in a train bombing that killed 193 people. “That’s ridiculous”, you might think. “I’m completely innocent, why would they suspect me?”. Oregon lawyer Brandon Mayfield may have thought something similar when this exact hypothetical scenario became his reality back in 2004 during the Madrid Train Bombings. He was detained by the FBI because they had found that a partial fingerprint preserved from a bomb fragment matched one of his fingerprints. But Mayfield was completely uninvolved, so how come his fingerprint was on the bomb?
The truth is: it wasn’t. Forensic experts had falsely identified a match between the bomb fragment fingerprint and Mayfield’s fingerprint after a software had indicated that the two prints were similar. However, the software also found 19 other people whose fingerprints were similar. In reality, none of the prints were perfect matches, so why did the forensic experts still assert that Mayfield’s print was undoubtedly a match?
Part of the explanation is that they were biased by contextual information. When looking into Mayfield’s background the investigators found that he had recently converted to Islam, and represented a client who tried to join the Taliban, meaning he had a connection to terrorists. Additionally, he had taken flying lessons before, and he had looked at maps of Spain on his computer. In the eyes of the FBI, these findings were reason to cast serious suspicion on Mayfield. Knowing contextual information like this can lead to certain biases, stereotypes, and expectations, which consciously or unconsciously impact subsequent decision-making. This is because we are more likely to search for and favor information that is in line with our beliefs and expectations.
Forensic experts are not exempt from this bias. The influence of contextual information can be found in many forensic analysis techniques, and all different investigation stages: it can influence which data is collected at the crime scene, which testing strategies are used, and which interpretations and conclusions are made regarding the findings. To say it in the words of French forensic science pioneer Edmond Locard: “Physical evidence cannot be wrong […]. Only its interpretation can err. Only human failure to find it, study and understand it can diminish its value.”