March 29, 2023

Testing alibis is not as straightforward as it seems

Testing alibis is not as straightforward as it seems
Testing alibis is not as straightforward as it seemsTesting alibis is not as straightforward as it seems

IN 1985 RONALD COTTON, a resident of North Carolina, was falsely convicted of rape and burglary, and sentenced to life imprisonment. Nearly a decade later, he was exonerated on the basis of DNA evidence. Not only did the victim make an error in identifying him as the perpetrator, but Mr Cotton had also provided an alibi that could not be corroborated.

Listen to this story

Enjoy more audio and podcasts on iOS or Android.

This, it turned out, was because he was wrong about where he had been when the crime was committed. A subsequent investigation showed that he had, in error, instead told the police where he was at that time on the same day during the week before. An unfortunate confusion, then, with serious consequences. But, if research just published in Psychological Science is to be believed, one that might be commoner than imagined.

Aware of Mr Cotton’s case, and of estimates that about 1% of America’s prison population are innocent of the offences they are serving time for, Yim Hyungwook of Hanyang University, in South Korea, and Simon Dennis of the University of Melbourne, in Australia, wanted to understand better how people remember details about where they were when in the past. They hoped they would thus be able to identify common mistakes that people make about such matters, so that they could bring these to the attention of detectives and defence lawyers, and therefore help keep innocent people out of jail.

To this end, they recruited 51 volunteers and asked them to download an app onto their smartphones, to collect information about their lives over the course of four weeks. The app tracked the phone’s location using GPS, and made recordings, from time to time, of nearby sounds.

A week after the experiment ended, participants were asked to take a memory test. They were presented with electronic maps that had four markers on them, and were asked 72 questions of the form, “click the marker where you were on Thursday morning 8am on August 15th”. They were allowed to zoom into the maps and take a look at the names of streets and venues in the area, but were not allowed to use the “street view” facility or to refer to any other information, such as past appointments noted on the calendar of their phone. After selecting a marker, they were asked to rate how confident of their answer they were, on a scale that ranged from one (not at all confident) to five (very confident).

Dr Yim and Dr Dennis found that participants got things wrong about a third of the time, and further examination of the data revealed that these errors were not random. People were more likely to remember events incorrectly if they were similar in time or space than they were if they were not. Specifically, they chose the right day of the week but the wrong week 19% of the time, and the right hour of the day but the wrong day 8% of the time.

Analysis of the audio files revealed that they were also more likely to make errors when two locations had similar acoustic environments. And when participants were confident of their answer, they were almost always right, and when they were not at all confident, they were much more likely to be wrong.

None of these findings is, it is true, that surprising. But they are, nevertheless, novel—and are the sorts of things that detectives and lawyers need to be aware of. How common it is for mistaken imprisonment, or worse, to happen as it did to Mr Cotton on the basis of alibi claims that could not be proved because they were themselves mistakes, is unknown. But that is because, until now, no one has had reason to ask.

A version of this article was published online on June 2nd 2021

This article appeared in the Science & technology section of the print edition under the headline “Where were you on Thursday the 15th?”