How much should we trust eyewitness testimony? Part I

For much of human history, the best way to prove someone’s guilt was to seek testimony from someone who saw the crime occur. In fact, before forensic evidence came into its own, eyewitness testimony was sometimes the only way to secure a conviction.

To this day, juries find eyewitness testimony to be highly convincing. But how much trust should they place in the accounts of witnesses? Numerous studies in recent years have demonstrated that eyewitness testimony is far less accurate and reliable than commonly believed.

Wrongful convictions blamed on faulty witness testimony

The Innocence Project is an advocacy group working to combat and overturn wrongful convictions, particularly through the use of DNA evidence. When cases are overturned, the group also documents which mistakes caused or contributed to the wrongful conviction. After examining more than 375 such cases, the Innocence Project noted that nearly 70 percent involved mistaken eyewitness identifications.

Why are witness accounts likely to be inaccurate?

Some problems can be attributed to the shortcomings of human observation and memory errors. For instance, eyewitnesses (including victims) often have a very difficult time accurately identifying a suspect whose race is different from their own.

Another problem is that most crimes occur quickly, and witnesses may only get a brief glimpse at what happened and who was involved. Yet even based on this very limited observation, people are often overly confident about the accuracy of what they saw.

While we cannot necessarily compensate for failures in human observation, we can and should change the weight that we give to testimony in light of the fact that it is not nearly as reliable as long assumed.

There are problems with the accuracy of eyewitness testimony that can be solved through better police procedures. Many states, including New Jersey, have already implemented some of these changes. Check back as we continue this discussion in our next post.