In Moonwalking with Einstein: The Art and Science of Remembering Everything, Joshua Foer describes his journey into becoming a memory wiz. One of the tricks in his bag: create “memory palaces” and deposit in each nook and cranny certain items from a list to be memorized. For instance, you are presented with this list of unrelated words to be memorized as fast as you can: maid, lightning, ocean, physics, burglary, and so on, 100 or more words. You are then asked to repeat the list in correct order. If you use brute force to cram the words into your brain there is no way you could recall more than 7-10 words. This is because our short term memory, called “working memory” (also called declarative memory), is limited to about 7 items, and with training can be expanded to at most 10 items. To break out of this straitjacket you need to outsmart your working memory by creating cues, or “critical lures”, that weave the list of words into a story or an image. An example: The maid opened the door and was struck by lightning, as the storm whipped up the ocean waves…you get the idea.
Now, what about a series of words that have a unifying aspect? For instance, surgeon, nurse, pharmacy, drug, hospital, disease. As you probably guessed, the task of finding an organizing principle to facilitate memorizing this list is a lot easier and doesn’t necessitate the invention of bizarre stories. But this list is different from the previous one in another aspect: the first is emotionally neutral, the second has an emotional content. Shannon McKeon, Edward F. Pace-Schott, and Rebecca M. C. Spencer of the University of Massachusetts ( PLOS One, Nov. 2012) asked volunteers to memorize a neutral list and another list of words with emotional content. Some of the subjects were tested after 12 hours of wakefulness. The other group was tested after a 12 hour period that included sleep. They then tested the “truthfulness” of recall (veridical recall” in psychological lingo, verity=truth) in both groups.
The effect of sleep
It has been well known that sleep enhances veridical memories. But how does the emotional content affect veridical memory? Quite surprisingly, the experiment indicates that sleep promotes false memories. A clue as to what happens in the brain during sleep is offered by the results. Sleep improved veridical recall of both neutral and emotional lists of words. But the group that was tested after a period of sleep “invented” critical lures. For instance, the word “doctor” was not included in the original list, yet significantly more subjects who had slept included it in their recall than the subjects who had not slept. What’s the explanation? During sleep, the brain reorganizes newly acquired information for more efficient storage and easier recall. Hence the creation of a critical lure (e.g.“doctor”) that could serve as an organizing principle for the list.
Sounds pretty esoteric research until you ponder the implications. Consider a rape victim who didn’t have a good look of the assailant and is emotionally distraught. A detective is trying to get information. Did the man have a distinctive accent? Was he over 6 ft. tall? Was his hair distinctive in some way, such as thick with tighter and smaller curls? Did he use foul the language of rap music? And so on, and so on. The victim is allowed rest, go home and have some sleep before returning to the police station to pick out a suspect out of a lineup.
What happens in the interval? The brain of the sleeping victim is trying to make sense of the barrage of detective’s questions. It needs a critical lure, an anchor to the web of data that would make a coherent picture. Is it any wonder that on the next day, or a week later, the critical lure is “a tall black man”? It pulls all the details together. And the more time elapses the more deeply embedded is the seemingly compelling scenario. Far fetched? Just check the long list of debunked eyewitness testimonies that condemned innocent people to many years of prison, including death row. These were not bad people; they were mostly normal people of good will who fell victim to a trick their memory played on their brains. Did the detective try to plant facts in the victim’s head so as to make a black person the inevitable suspect? In most cases, he was just trying to get information.
So what can be done to counter the tricks our memory plays? Science has been dealing with such unconscious biases for many years. A procedure involving controlled, randomized, double-blind study has become the gold standard of a valid study. Why not the lineup? Train detectives in the art of non-biasing investigation. Video-record all interactions with the victim (and the suspect, when identified). Have two panels, one including the suspect and the other serving as a control. Have the lineup procedure administered by a third party, not involved in the investigation. Just like in science, these steps will not eliminate mistakes, but it will help limit the damage our faulty human memory can inflict.