I have only ever seen one car crash and I remember it with crystal clarity. I was driving home along a motorway and a car heading the opposite way simply veered into the central reservation. Its hood crumpled like so much paper, its back end lifted clear off the tarmac and it spun 180 degrees before crashing back down in a cloud of dust. All of this happened within the space of a second, so the details may be different to what I remember. But the emotions I felt at the time are still vivid - the shock of the sight, the fear for the passengers, the confusion over what had happened.
Many studies have shown that peoples' memories become particularly clear when it comes to traumatic or shocking events. Even learning about a shocking event, rather than witnessing it first-hand, can produce unusually clear recollections. Many of us still remember where we where when we learned that famous figures like Princess Diana or John F. Kennedy had died (I found out about Diana on the toilet).
Scientists have suggested that this type of event triggers a process that produces a very specific and exceptionally vivid type of memory called a 'flashbulb memory'. This concept has been kicking around since the 1970s, but the evidence that flashbulb memories actually exist is inconsistent.
Tali Sharot and colleagues from New York University decided to find some proper answers by studying the brain activity of people remembering a traumatic event. Doing such experiments would normally be ethically impossible - you cannot after all willingly traumatise someone in the name of science. But Sharot did not need to - unfortunately for us, the twenty-first century has already provided its fair share of traumas.
On September 11, 2001, the people of New York experienced terror and devastation on a massive scale. If any event led to the formation of flashbulb memories, this one would. Sharot recruited 24 people who had witnessed the World Trade Centre attacks first-hand and asked them to remember either the attacks, or a random event from another summer.
She found that people who were in Downtown Manhattan near the attacks had distinctly different memories than those who were twice as far away in Midtown. The Midtown group recalled their 9/11 memories in the same way as their generic ones. But the Downtown group remembered their 9/11 experiences more vividly, strongly and confidently and gave both longer and more detailed descriptions. They reported seeing the towers "burning in red flames", smelling the smoke and hearing "the cries of people".
Not content with relying on descriptions, Sharot used a brain scanner to see if these differences were mirrored in the volunteers' brains - specifically, in a small region called the amygdala, the brain's emotional control centre. The amygdala affects how memories are stored in the long-term and animal studies have shown that this storage is influenced by stress hormones. Sure enough, when asked them to think about 9/11, the Downtown group showed much greater activity in their left amygdala, while the Midtown group did not.
Despite these results, Sharot is still tentative about concluding that flashbulb memories, as they are classically defined, exist. Nonetheless, her results clearly show that experiencing a shocking event yields a very different and exceptionally vivid type of memory than the humdrum occurrences of daily life. These memories - flashbulb or not - are formed through a special mental route which involves the amygdala.
And Sharot's brain scans turned up something more unexpected. When the Downtown group thought about 9/11, they also showed much lower activity than normal in the parahippocampal cortex. This part of the brain is thought to be involved in processing and recognising details of a scene or event. If its neurons are dimmed during shocking situations, this could explain why people who experience surprising events remember how they felt, but cannot reliably provide details.
During the tragic shooting of Jean Charles de Meneses in 2005, eyewitnesses proved to be wildly inaccurate, with first-hand accounts of his clothing, police action, and the number of shots fired clearly contradicting each other. If emotions are prized over details in memories of shocking events, how much value can we truly place on eyewitness accounts?
Reference: T. Sharot, E. A. Martorella, M. R. Delgado, E. A. Phelps (2007). How personal experience modulates the neural circuitry of memories of September 11 Proceedings of the National Academy of Sciences, 104 (1), 389-394 DOI: 10.1073/pnas.0609230103