Trained police can spot bag thieves before they commit a crime

Are professionals better than the rest of us at spotting wrong-doing? The historical evidence is gloomy: one study suggested job interviewers perform no better than novices at spotting cheaters. Several reviews have concluded that police officers and detectives have less than stellar abilities to catch lies in interrogations, with some research even suggesting chance levels of performance. However recent research has begun to rehabilitate expert abilities at interview lie detection. And now a study in Applied Cognitive Psychology suggests that the police have superior skills at spotting wrong-doing outside of the interrogation room – they're often able to identify crimes before they even occur.

Corinne Ines Koller and her colleagues from the University of Zurich worked with the city's airport police to examine people's ability to identify baggage thieves using only CCTV footage from the buildup to the crime. The research involved 12 video clips, and in each one there were between one and three thieves, mixed in with innocent bystanders – between 20 and 100 per clip. Each clip was split into three phases (each being 53-144 seconds long), and after each one, participants had to identify up to three people as suspects before going on to the next phase.

The researchers tested the detection skills of 315 participants: criminal investigators and police officers of varying experience, and two comparison groups: psychology students and fresh police recruits. All groups performed better in later phases, making fewer false accusations and more accurate ones when they had the additional data available before an imminent crime.

But crucially, the police officers and detectives were better at every phase in spotting the crooks, and using an aggregate score of detection accuracy (based on the balance between true and false identifications), they were the only ones performing better than chance in phase one. The psych students and police recruits were hopeless at this stage – the students because they were too tentative in making identifications, the police recruits because they were too trigger-happy and fingered innocent people at a high rate, a tendency that continued through all the phases and may reflect a combination of inexperience and high motivation to do good police work. After phase one, the comparison groups started to perform above chance, but never caught up to the performance of the police.

This evidence suggests that trained police can identify potential criminals more accurately, and earlier, than non-experts. Unlike hooligan gangs or drunk blowhards, baggage thieves are doing their best to keep their criminal intentions hidden, and so these cues aren't easy to detect. But Koller and her co-authors argue that the clues to recognising criminal intention from the build-up to a crime are learnable, once you know what you are looking for. Plotters are trying to look normal while being in an abnormal (aroused, tense) state, whereas normal people aren't trying, they just are. Compare that to the situation in a police interrogation, where both innocent and guilty are actually playing a similar game – convincing the police of their innocence, on pain of criminal consequence – and we can begin to understand how hard a game that is to read, which may explain the uneven picture of professionals' interrogation abilities.

The skills examined in this new research – the art of reading the build-up – are important for spotting robbers, terrorists, and sexual predators. Understanding exactly what police draw upon to make these judgments, and sharing that with others, such as CCTV operating staff, may help us on the way to a safer society.

Further reading

Koller, C., Wetter, O., & Hofer, F. (2016). 'Who's the Thief?' The Influence of Knowledge and Experience on Early Detection of Criminal Intentions Applied Cognitive Psychology, 30 (2), 178-187 DOI: 10.1002/acp.3175