Misconceptions of Memory and Trauma

Daphne Maurer, Ph.D.

Fellow, Royal Society of Canada
Distinguished University Professor,
          Department of Psychology, Neuroscience & Behaviour,
          McMaster University, Canada
Fellow, Association for Psychological Science

This is the summary only. Download the whole paper here.

How memory works. Memories are not what most people believe them to be. They are not stored in neuronal pigeonholes, waiting stably to be retrieved. They are neurochemical connections that are too unstable not to be changed by the neurochemical reactions forming recollections. For this reason, memories constantly change. On page 3 you will find an explanation of this for general readers, eight pages from one of my books.

Accurate memories. Memories constantly change. A lot of experimental evidence shows this. Events accompanied by a modest amount of emotion do tend to be remembered more accurately, but a large set of studies show that traumatic events tend to be remembered less accurately. These studies tested soldiers traumatized under standardized conditions in a military survival school. Page 12 is a three-page review of this work by two professors at the (US) National Center for PTSD, based at Yale.

Consistent memories. Consistency and accuracy are different dimensions. It is easy to be consistently wrong.

A review of 37 studies (page 15) found that in some of them, traumatic memories tended to become a little more consistent with time, but the difference was slight and not of practical significance. The authors conclude:

Results of this review have implications for legal practice. Memory is a reconstructive process, which is prone to errors. Therefore, we cannot fully rely on its accuracy, completeness, and consistency. Eyewitness testimonies completed in situations where a victim is showing emotional reactions may not be entirely reliable. Testimonies completed at an early stage following a crime may be incomplete and inaccurate, in particular with regard to important details of an event. However, reports of criminal events can be highly consistent over time, without necessarily being accurate. Therefore, one has to be cautious using consistency as an indicator of accuracy and drawing conclusions from a single testimony.

Recovered memories. There is a great deal of evidence that interviewers can and do induce the "recovery" of erroneous memories. An overview of this literature starts at page 34. One of its authors is Elizabeth Loftus, Distinguished Professor at the University of California, Irvine. Loftus's research on memory has earned her 59 awards from around the world, plus seven honorary degrees.

Trauma-informed investigation. When this term means sensitive listening immediately after a trauma, the technique is sensible, but it commonly means asking leading questions long after the fact, sometimes to "recover" memories. This kind of "trauma-informed" investigation is inappropriate. Elizabeth Loftus (see above) co-authored a recent paper explaining the problems. It starts at page 65.

Trigger warnings. Trigger warnings tell people that they may encounter something they find upsetting. They are supposed to reduce anxiety but they do the opposite, because they work like a laboratory bell warning a mouse that it may be shocked. In this situation, the bell comes to frighten the mouse whether it is shocked or not. This is basic Pavlovian conditioning. Similarly, trigger warnings usually become more upsetting than the triggers themselves.

The top of page 73 shows this. It is a table by researchers at Harvard reporting results from the six studies of trigger warnings that the authors had been able to find as of 2020. One of the six did show a reduction in anxiety afterwards, but another one found an increase, and both of those effects were negligible. On the other hand, the three studies that measured it found the warning itself to trigger anxiety. Moreover, trigger warnings are almost never heeded. An example of this is on the bottom of page 73.

In short, trigger warnings do harm rather than good.

False charges. Supposedly, no woman would lie about rape because no woman would subject herself to the inquisition that follows. However, a meta-analysis (page 74) based on police records found convincing evidence that across four anglophone countries, roughly 5% of rape reports were false. Apparently some women, like some men, want to take revenge upon somebody, and/or enjoy being the centre of attention, and/or develop psychotic distortions of reality.

Five percent may sound like a low error rate, but in this context it is high. Imagine that after convicting 19 criminals, a judge orders a cop to pull a man off the street at random and throw him into jail. That is the equivalent of what happens when 5% of rape reports are false.

Moreover, 5% is a conservative estimate. Many of the less clear cases cases were undoubtedly false as well, so the real rate of false reports will be higher. Note, too, that less serious sexual assaults are easier to fabricate and less onerous to report, so falsifications of lesser claims are likely to be more common.

This is the summary only. Download the whole paper here.