Memory “can sometimes be used (in court) as a shield and sometimes used as a sword,” observes defence lawyer Marcy Segal.
“In the Ghomeshi trial, it was used as a sword: if you say it happened, you’d better remember correctly,” she said.
As part of their closing arguments, the defence argued that the complainants could all recall “rich detail” from the time of the assaults, but could not form a consistent or coherent narrative of the assaults themselves — all of which goes to show the complainants are unreliable.
In the trial of Officer James Forcillo, memory was also a key part of the defence strategy, but in a different way. The defence consistently pointed out inaccuracies in the memory of eyewitnesses to the traumatic shooting, leading up to Forcillo himself testifying that he remembers — it turned out clearly mistakenly — seeing Sammy Yatim sitting up after being shot three times, leading him to fire another six shots.
“It’s don’t blame these officers for not remembering everything, it was a traumatic event and sometimes your memory plays tricks on you,” Segal says.
Why we remember certain details and not others, why our recounting of memories can change in fundamental ways over time, and the impact of stress and trauma on memory are all areas that arise frequently in the courts and have been and continue to be closely studied.
Memories of Trauma
When someone is in a traumatic situation, the body prepares for “fight or flight,” and prompts certain changes, like an increased heart rate. It also results in the part of the brain that captures explicit memories being deactivated, leaving only part that captures sensory memories.
That means memories of a traumatic event are not processed as a narrative, explains Toronto-based clinical psychologist Lori Haskell, who researches trauma and sexual violence. Instead, they are “sensory fragments: smells, images, sensations in the body.”
The perceptual field narrows, focusing only on what is needed for survival, she says, sometimes resulting in extremely vivid “super-encoded” memories.
She recalls a case where a woman was sexually abused as a teenager by a group of men. She could clearly remember what each man was wearing during the assault, though she could not, when asked by the defence, recall other details about the day like what she had for breakfast, Haskell said.
But witnesses may not understand why they can’t recall certain things and can get into trouble is in trying to answer questions about details they cannot remember.
“They don’t know they’re making s--t up,” Haskell observes. “We all fill in the blanks.”
Haskell, who teaches Crowns and police officers across the country about how traumatic memories are processed and how to question someone who has experienced trauma, suggests that Crowns spend more time in court asking the witnesses to share their visceral memories of the key event.
“What that moment felt like, how did she make sense of it, how did she experience it.”
Sharing as much detail about the sensory memories, like what the witness smelled or what the witness focused on in the room, can make the testimony more credible, she says.
“We have to get away from this narrative of peripheral details, it is just not how the brain is wired, it is not how memories are encoded. And there is a lot of attention on that.”
Why memories change over time
“People do believe that memory works much more like a video-recording or a DVD or some kind of high-fidelity recording of our experience than it actually does,” says Christopher Chabris, a psychology professor at Union College in the U.S. and co-author of the book The Invisible Gorilla: How Our Intuitions Deceive Us.
“Memory is even more changeable than people realize and it can be distorted over time even without our being aware of it. So a memory we retrieve one month, one year, ten years later can be a very inaccurate representation of what actually happened with people being completely unaware of that.”
When giving a police statement or testifying in court, people can’t just press play on their memories, he says.
“What you are doing is reassembling and reconstructing the memory at the time when you are retrieve. And each time you retrieve the memory you are recreating it fresh as though you had to go to a bunch of different books and look up all the facts instead of just reading it out of one book.”
That means events that happened at another time, or another piece of information can get mixed in, leading to different versions of a story being told. It’s like a game of broken telephone, he says, where a memory is being passed along through a series of people, changing a little each time in the retelling.
“I don’t think there is a solution to this,” he says. “You really need to go back and find other corroborating information in the past.”
Another issue is that people can express great confidence in their memories even when they are incorrect, says Chabris. So even though accurate people are more likely to be confident, confidence is not the most reliable indicator of accuracy.
Aaron Benjamin, a psychology professor researching human memory at the University of Illinois, notes that “repeating the same information over and over again, perhaps under consistent questioning, can make you more confident in the information you produce, but the accuracy of the information isn’t changing at all.”
Both researchers point to the importance of other forms of evidence to the justice system — such as cellphone videos, and perhaps even more so with the introduction of police body cameras.
A string of high-profile wrongful convictions centring on an eyewitness confidently — but wrongly — identifying the perpetrator of the crime has led the courts to “clearly acknowledge the frailties of eyewitness identifications,” as a federal Department of Justice report that recommends ways to prevent miscarriages of justice puts it.
In both the widely watched Netflix documentary series Making a Murderer, and in the oft-cited case of Jennifer Thompson, rape victims sincerely and confidently identified their attackers, only for DNA evidence to show they were wrong years later.
But this doesn’t mean eyewitness identification, often critical evidence for the prosecution, should never be relied upon.
John Wixted, an experimental psychologist at the University of California, San Diego, explains that there is only one chance to test the memory of an eyewitness — so making sure the testing conditions are adequate is crucial. If the memory is contaminated through a biased lineup or some kind of suggestion, it can never be tested again.
But, he argues in his latest research paper, if the test is conducted right with an uncontaminated memory, and there is a “high confidence” identification, it can be considered reliable.
“Perhaps even more important to understand is that low confidence ID’s are unreliable. Think of it like an inconclusive DNA test,” he says.
Wixted also suggests that, contrary to what may be commonly thought, a confident eyewitness initial identification given six months after a crime is no less reliable than one given shortly after.