Through a Glass, Darkly

Memory and trauma.

Posted Sep 26, 2018

The human brain is not a recording device. It’s a sense-making device. A predictor of future events, and an assimilator of past ones. This is Psychology 101. Almost the first cog sci class we teach to students demonstrates that even using different words to describe video footage of a car crash can dramatically affect their memory of the event. Just using the word "bump" rather than "smash" can change reported speeds by 25 percent, or have people seeing damage where there was none. And this is an unemotional event, with no personal stake. What happens when an event is one of the most traumatizing imaginable—rape or attempted rape? Is it really the case that our memories of such events can be unreliable?

Suddenly the media and the Internet seem to be awash with psychology experts, many of whom are suddenly very sure of some highly technical and, in some cases, poorly understood mechanisms of memory storage and retrieval. This could be seen as gratifying to a psychology lecturer. Is it the case that we have produced a generation of informed and nuanced psychologists, alert to the myriad difficulties of relying on uncorroborated eyewitness testimony, while showing the human warmth and meaning-finding needed in a therapeutic alliance with victims?

Not exactly.

I am not going to link to them, but I cannot imagine that any readers of this are unaware of the claims and counter-claims surrounding Dr. Christine Blasey Ford's accusation that Supreme Court nominee Brett Kavanaugh attempted to sexually assault her while she was 15 and he was 17. Calls to always believe the victim have been met with accusations of memory distortion or outright invention.

One prominent citizen has opined, "I have no doubt that, if the attack on Dr. Ford was as bad as she says, charges would have been immediately filed with local Law Enforcement Authorities by either her or her loving parents,"

In response to this cacophony, the APA has brought out a statement to the effect that:

  1. Sexual assault is very under-reported
  2. False claims of sexual assault are low
  3. The developmental stage of the victim in question means that she was even less likely to report a sexual trauma (especially to parents) and that
  4. Extreme events play merry hell with the storage and retrieval systems of human memory, so that common sense assertions regarding what others think she should or should not recall cannot be relied upon

All of this is true.

Memory Science

However, experience (both in person and online) has taught me that this is unlikely to be enough to silence some armchair experts. Some have gotten hold of the idea that rape is an event especially linked to false memories or misidentification, so I think it’s worth exploring some of the data on this. What does the science say about the ways our memories can fail us?

Classically, there are seven ways that memory can go wrong—transience, absent-mindedness, blocking, misattribution, suggestibility, bias, and persistence. (People eager for more detail are encouraged to check out Daniel Schacter’s classic (1999) review.)

Forgetting and Wishing to Forget

Transience is the decay of unused memories over time: Do you recall the telephone number of a restaurant you reserved a table at two years ago? Likely not, unless you used it again, and repeatedly. Memories decay over time as classically demonstrated by Ebbinghaus in the 19th century.

Absent-mindedness is a failure to properly encode the memory at the moment. It’s only when I find myself about to put the cat in the fridge that I realize that I put the milk outside for the night.

Blocking is most commonly experienced as a tip-of-the-tongue phenomenon. You know the one, where you are trying to recall, Oh, you know, err… that word… it’s Greek, I think… or Russian… begins will an “L”… or an “M”… you know the one. That thingy.

These are all errors that occur when a person tries to recall something and it just isn’t there. That’s not what we are talking about with trauma victims. They are much more likely to suffer from persistence—the intrusive recall of painful memories that one would rather not recall. We find this with trauma victims all the time. That may well be an issue in the Ford case, but is not the focus of the criticism. This leaves us with three things—bias, suggestibility, and misattribution.

Let’s deal with each of these in turn.

Distorting and Creating

Since the early days of cognitive science we have known that pre-existing beliefs can affect memory encoding and retrieval. One of the early heroes of cognitive science—Frederick Bartlett, used to go around Cambridge telling people stories and then button-holing them days, weeks, or months later, to see what their pre-existing schemas had done to the memories. This is known as bias: We subtly alter things to make them make more sense to us, such as recalling that we always thought that the guy later revealed as the murderer in the detective story was untrustworthy. These are alterations to make our memories fit with our schemas of self and world. But, is it possible to generate memories of things that never happened? Yes, and no.

Suggestibility: We can generate falseness in memories in suggestible individuals. That living legend among memory researchers—Elizabeth Loftus—was intrigued by the case of a 14-year-old (Chris) who was brought to sincerely believe that he had been lost in a mall at age 5 by the mischievous intervention of his older brother. Loftus managed to generate similar false memories in a quarter of participants with families who were similarly happy to join in and fool their family members. Some would happily recall meeting Bugs Bunny at Disneyland (Bugs is not a Disney character). I don’t know about you, but I don’t find this nearly as surprising as some people do. I genuinely cannot tell if a small number of my childhood memories are genuine, or the result of recalling having them fondly (or not so fondly) recalled by family members. One of the things that Loftus’ research has highlighted was the scourge of false memories of childhood abuse, the memory of which could be implanted by hypnosis, or hypnosis-style techniques, in adults previously unaware of such abuse. This caused a huge storm at the time, but Loftus has been entirely vindicated. Memory recovery techniques of childhood abuse are highly contentious. And, when I speak to adults who have suffered such abuse, I have to report (and I’m not alone in this) that the usual issue is persistence of horrible intrusive memories that they would rather get rid of. Not forgetting and sudden adult recall.

But in the Ford case we are not talking about childhood memories, or memories affected by schemas. And we certainly aren’t talking about memory-loss of trivia like unused phone numbers. We are talking about the sexual assault of a near-adult. These are not events in a misty childhood, much of whose reality is generated by family interactions long after the event. What is left of the seven ways memory can go wrong? Misattribution—the attaching of a memory to wrong referent.


Are there misidentifications in eyewitness testimony, even in a crime so violent and personal as rape? Indeed, there are. Let’s consider a couple of the most dramatic cases. The Australian psychologist Donald Thomson was accused of rape on the basis of a detailed description of him given by the victim. He had a pretty unassailable alibi: He was on live TV at the time. As it happened, this was a live television interview that had been playing (and watched by the victim) just before the rape occurred. Ironically, his TV interview was about memory distortion.

In another famous case, Ronald Cotton served 11 years for the brutal rape of Jennifer Thompson. A photofit composite produced at the time resembled him, and he was a suspect who had served 18 months for sexual assault. However, he had not raped Jennifer. By an odd coincidence the actual rapist, Bobby Poole, whom he only barely resembled (both were tall and black but did not have much else in common) came to work in the prison kitchen and boasted that Ronald Cotton was “doing his time." Subsequent DNA evidence led to Cotton being released—but only after serving 11 years. Although Jennifer Thompson had seemed calm and collected, her identification via photofit had occurred while was still traumatized, and then that face became the face of her attacker.

Finally, there is the disturbing case of the misidentifcation of Nazi war criminal Ivan the Terrible. Ivan Demjanjuk, was misidentified by several concentration camp victims thirty years after the event. Three years after his conviction by an Israeli court, the evidence overwhelmingly pointed to Ivan the Terrible being a different person: Ivan Marchenko. What had gone wrong? It is not as easy as it first appeared. While not being the person in question, the evidence is that Demjanjuk was a concentration camp guard at the time--and one called Ivan. He just wasnt that particular Ivan. He died while his appeal was still being heard so we will never know all the details. However, he was not unconnected to events.

So, there are mechanisms that can generate false memories with emotional events. But—and this is a rather important caveat—these do seem to require some sort of link to the person being accused (they were together in a lineup, they were on TV at the time of the event, a photofit was generated that resembled them and then gets flash-bulbed into memory, they were a guard in the same camp with the same name and uniform). There are no documented cases I am aware of (contact me if I have missed one) where someone has completely misidentified someone that they knew at the time. There is always some sort of link--perhaps one that our rough and ready ancient emotional limbic system blurs some details of--but a link nonetheless.

Making things up from a whole cloth is much rarer, and looks different. It can happen: Humans are flawed beings, and there are people who have made false reports for revenge purposes. None of that is being suggested here, nor is there any evidence to suppose it in the Ford case.


What is the upshot? Do not rush to judgement, one way or another. Let people tell their stories and explore them with as much critical sensitivity as we can bring to bear. Modern cognitive interviews are much better at elicitng information than combative or coercive techniques. Find corroboration, such as other accusers who have credible stories. Do not assume that because someone did not immediately bring things to the police then the events did not occur. The police are trained to deal much more sensitively to these things than they used to be, but this is never going to be an easy thing to report.

One way that our brains make sense of the world is to compartmentalize horrible events, especially ones where we feel (rightly or wrongly) that we might be blamed or shamed, despite being the victim. This is where Freud went wrong, I think: Mistaking unwillingness to relive trauma, for genuine repression from memory. The systems for doing this are deep in the lymbic system--and things that are deep the brain tend to be the oldest systems, and the systems least amenable to reason and control--hence the intrusive persistent memories characteristic of PTSD. And contrary to film and fantasy, humans often freeze when attacked. Maybe in 70% of cases.

I started by saying that the human brain evolved to be a sense-making device. It is disturbing to realize that our memories are not objective records, but it is possible to go overboard with this realization of a subjective element. It might be more accurate to think of memory as being facultative rather than objective: As generating useful information to an organism in its environment of threats and opportunties. Sense is not exactly the same as objective truth, but it has to track it to some degree. An organism that cannot make sense of its environment will not survive. Trauma should be seen as an adaptive mechanism. Like pain. One that warns you to stay away from places, people, or situations like this one that caused the pain. But, like so much else, this system evolved on the savannah plains of Africa, in small groups of mostly kin, to warn us about things that were likely threats then and there. A dangerous local to warn the women about, a hungry lion by the watering hole, a poisonous plant. It didn’t evolve in circumstances where a TV might be playing as your attacker assaulted you, melding the face on the TV with theirs. It didn’t evolve in social circumstances where suspects were lined up while you were still traumatized and asked you to pick which of these was the attacker. It didnt evolve in circumstances of organised death camps with humans systematically torturing others in huge groups over years. And we need to be alert to these facts about human brains. And also alert to the fact that we do not want to believe certain things about certain people because another feature of our fallible memories is to distort inconvenient facts to suit our schemas. Schemas like, "He’s always seemed decent to me," or, "He's on my team." Most of history's worst monsters seemed decent to someone.


Loftus, E. F., Miller, D. G., & Burns, H. J. (1978). Semantic integration of verbal information into a visual memory. Journal of Experimental Psychology: Human Learning and Memory, 4, 19-31.

Schacter, D. L. (1999). The seven sins of memory: Insights from psychology and cognitive neuroscience. American psychologist, 54(3), 182.

Bailey, C. H., & Chen, M. (1989). Time course of structural changes at identified sensory neuron synapses during long-term sensitization in aplysia. Journal of Neuroscience, 9, 1774-1781.

Ebbinghaus, H. (1885). Uber das Gedachtnis [Memory]. Leipzig, Germany: Duncker and Humblot

Bartlett, F. C. (1932). Remembering. Cambridge, England: Cambridge University Press.

Loftus, E. F., & Ketcham, K. (1994). The myth of repressed memory: False memories and allegations of sexual abuse. New York: St. Martin's Press.

Loftus, E. F., Feldman, J., & Dashiell, R. (1995). The reality of illusory memories. In D. L. Schacter (Ed.), Memory distortion: How minds, brains and societies reconstruct the past (pp. 47-68). Cambridge, MA: Harvard University Press.

Loftus, E. F., & Pickrell, J. E. (1995). The formation of false memories. Psychiatric Annals, 25, 720-725.

Thomson, D. M. (1988). Context and false recognition. In G. M. Davies & D. M. Thompson (Eds.), Memory in context: Context in memory (pp. 285-304). Chichester, England: Wiley.

Some folk have expressed distress at not being able to find the search terms that they wish such as "Rape" "Misidentification" "Witnesses" and so on. In other words, they are skeptical that the misrepresentations I have discussed above actually occured. So, as a service to these people, here is some assistance:

More Posts