This post is brought to you courtesy of Patreon. If you want to support more work like this, you can sign up here.

There were recently a pair of bad articles on rape and sexual harassment published in Skeptic Magazine. The first was on the #metoo movement, written by Carol Tavris. H.J. Hornbeck has a series of three posts on this that’s pretty good. I don’t agree with all his interpretations of the points Tavris was trying to make, but he provides a lot of background on the social phenomena and scientific points Tavris invokes.

The second article is a book review on the Sandusky child rape case. The book is written by Mark Pendergrast, whose qualification on the subject is being the author of Memory Warp; The Repressed Memory Epidemic and Victims of Memory. It was published by a small press, which isn’t what you’d expect for a book meant to overturn everything we think we know about an event which received a remarkable amount of press coverage.

I’d recommend this post by Christopher Tevuk on why and how the book fails to do what it set out to do. The short version is that the thesis of the book has already been examined and rejected by the courts.

So why would a skeptic publisher post both of these articles? Why would people who pride themselves on rational thought and examining the evidence promote it?

Some of the answers are obvious. Michael Shermer, the publisher and editor in chief of Skeptic Magazine, has a massive, undisclosed conflict of interest in publishing articles casting doubt on rape and sexual harassment. He’s a credibly accused rapist and serial harasser himself. And unlike prior instances in which skeptics have produced shoddy work that could help cast doubt on their guilt, there is no one to hold Shermer to a higher standard of skepticism in this case. He’s also the president of the Skeptics Society, which makes him his own boss unless the board wants to step in.

Other conflicts of interest are more subtle¹, at least relatively speaking. Jerry Coyne’s antipathy to highlighting sexism in the movement is pronounced enough that he’s compared people who point it out to “slavering dogs”. While Daniel Dennett usually keeps quiet publicly on the topic of rape and harassment, I have correspondence with him (after a prior occasion when he spoke out) in which he says affirmative consent laws are an example of anti-rape activism going too far.

On top of that, the skeptic movement isn’t exempt from the current wave of reporters working on stories about long-rumored misbehavior from powerful men. Those not accused themselves have at least one friendly colleague who is. Seeing these people promote bad articles on rape and harassment is disappointing, but it’s sadly not unexpected at this point.

However, there’s another class of interest strongly represented in the skeptics movement that we should address. That’s the interest in false, “recovered” memories of trauma.

Let me be very clear. The problems with induction of memories of abuse is very real. People’s lives were damaged by false memories of abuse. The skeptic movement and people now involved with the skeptic movement have done important work on this problem. They’ve changed the legal and therapeutic landscape for the better.

There’s still a problem here. That problem is that people who know quite a bit about implanted false memories don’t know necessarily know much about memory as a whole or about how memory has been shown to interact with traumatic events and with PTSD. Even some of those who do know are so invested in the fight against false memories that they ignore or deny parts of how memory is known to work in favor of seeing false memories everywhere.

Like a lot of topics in human cognition, the literature on memory is kind of messy and relatively preliminary. That doesn’t mean we know nothing, though. It means we’re still choosing between overarching theories of how memory works, with the possibility several of them are partially correct. It means we’re still figuring out what count as extraordinary events in memory formation and retrieval and what effects those have. It means we have a lot of partial answers that may be influenced by our limitations in ethical study designs. But it doesn’t mean we know nothing.

So what are some of the things we know about false recovered memories specifically?

• Not everyone is susceptible to them. Brewin and Andrews, writing for The British Psychological Society, characterize the situation thus: “Rather than childhood memories being easy to implant, therefore, a more reasonable conclusion is that they can be implanted in a minority of people given sufficient effort.” Estimates in the studies they look at (including Elizabeth Loftus’s work) show an effect in, on average, 15% of study participants, though they caution actual belief in those memories may be lower.
• In addition to effort, implanting memories appears to require substantial trust and/or vulnerability from those being affected. Laboratory studies have involved lying to subjects about what their parents recall as childhood events. Cases of recovered memories that have been examined in lawsuits have involved hypnotherapy, drugs, unprofessionally emotionally intimate relationships with therapists, and subjects who were already experiencing significant emotional distress.
• False recovered memories may occur spontaneously. In this case, the “memory” may be a dream or other vivid experience that is mistaken for reality. If someone receives reinforcement that their false belief this is a memory, they may become convinced it really happened. This reinforcement is not substantially different from part of the process of inducing false memories, though it doesn’t require as much work.
• Major professional organizations around the world have been working for a couple of decades to reduce beliefs and practices among therapeutic professionals that may contribute to the creation of false memories. That doesn’t mean no professionals continue those beliefs or practices, but those who do are a distinct minority and fewer than some surveys indicate.
• Over the same period of time, large-scale efforts have been underway to educate government and other agencies that deal with these issues on the problem of false recovered memories. This ranges from people who do interviews of possible abuse victims up through the courts. Experts who work on false memories often offer expert testimony in related cases.
• Not all memories described as recovered are false. According to the American Psychological Association, “[M]ost leaders in the field agree that although it is a rare occurrence, a memory of early childhood abuse that has been forgotten can be remembered later.”

False memories of abuse are a problem. They remain a problem even after decades of work. However, they’re rare and difficult enough that they probably shouldn’t be the first assumption when someone talks about remembering abuse. There are specific circumstances that should alert us to the possibility a memory is false, but we need to be careful to differentiate between those and the basic ways that memory works.

What do we know about memory of traumatic events generally?

• For the most part, memory of traumatic events works much like memory of anything else (pdf). The big thing that means is that we don’t remember traumatic events as a smooth narrative of events integrated with sensory experiences and meaning. We hang our memories of events on a few key points and reconstruct them as we re-experience or tell them. It also means people with better memories tend to have better memories of trauma.
• Memory is suggestible, but those suggestions tend to affect the details of retelling rather than the core narrative. That has implications for how we treat eyewitness accounts that are already discussed among skeptics, but none of those implications are “throw up our hands and declare it never happened”.
• Long-term memory as we think of it tends to start when children are significantly verbal, at age two or three. Memories that are supposed to occur before this point are generally not reliable.
• Various parts of memories may be more salient than others at different times and more or less likely to be remembered, including as our understanding of events changes. We don’t have to create new memories to come to a new appreciation of, for example, the social pressure we may experience to make certain people happy.
• The trend in memory is toward forgetting over time, even for traumatic events. We have fewer cues to memory in our lives as they and we change.
• Memories that challenge our core beliefs appear to be particularly subject to forgetting. In terms of trauma, this means that people may be less likely to remember abuse by close family members, as it challenges the belief that family members love them.
• Not everything we informally call “forgetting” involves actual loss of memory. We may claim to have forgotten things we don’t want to talk about. Some of us may be particularly good at directing our attention away from memories we don’t want to deal with. Neither of these means the memories involved actually become inaccessible at any point in time.
• One of the things people forget is that they’ve previously remembered things. “Oh, I haven’t thought about that forever!” isn’t necessarily an accurate statement. Recalling memories is a trivial act, not likely to create strong new memories in itself.
• We’re most suggestible when we have reasons to believe something is true. We may integrate secondhand memories of our childhood, but we’re more likely to do so when they’re trivial and characteristic of our behavior.

Now, that’s general memory of traumatic events. It doesn’t cover “abnormal” response to trauma, or PTSD. What do we know about memory and PTSD?

• The primary way that PTSD changes memory is that memory of relevant traumatic events is experienced in flashbacks. These are very vivid sensory and emotional memories that can overwhelm intellectual processing of memories, including the ability to talk about the event.
• People with PTSD have shown both susceptibility and greater resistance to suggested memories. As with most memory, this is probably related to how well the suggested memories fit with existing memories. This means details may be incorrect but the core of the event in question is likely remembered correctly.
• PTSD interferes with memory consolidation. In practical terms, some details of the event are stored with an unusual degree of fidelity, but a broader narrative of the event is not. This is not forgetting, though it’s not normal memory either. Once PTSD flashbacks are managed, consolidation and construction of memory narratives can still occur. A therapist guiding a patient through the emotional landmines toward more normal memory consolidation is not “recovering” memories, though they do have to be careful not to suggest too much, as any therapist should.

When we look more generally at how memory works, it quickly becomes apparent that focusing exclusively on the recovery of false memories produces lessons that aren’t generally applicable for evaluating memories of traumatic events. We need to continue to be on our guard for the circumstances that produce induced memories, and we have skeptics to thank for very important work on that topic.

However, it’s equally important that we, as skeptics, don’t fall into thinking every memory that people haven’t been shouting from the rooftops from the moment of trauma is induced. Recovered false memories are unusual events that happen under unusual circumstances. Abuse is a common occurrence, typically subject to normal rules of memory.

If we’re doing this because we care about reality, we need to make sure we don’t confuse the two. We also need to demand that those who publish on the topic declare interests that may bias their views on the topic. If you’d like to read a more responsible take on claims of false recovered memories, I suggest this piece instead of anything recent from Skeptic.

Want to see more work like this? Support me on Patreon.