July 25, 2013

False memories, or conflated memories? Let's not totally lean on Loftus

Very interesting. A new story in the New York Times says that false memories have been created in mice.

But, is that exactly true?

What I see from the story is that we should distinguish between a de novo false memory and the conflation of two true memories into a mashed-up false memory, which I argue is actually what happened in the mouse.

And, yes, this is important for understanding how humans understand and create either version of false memory.

Second, while Elizabeth Lofftus has been a pioneer in this field, she's not infallible, folks. Patrick Fitzgerald took her down a notch or two in the Scooter Libby trial.
Fitzgerald got Loftus to acknowledge that the methodology she had used at times in her long academic career was not that scientific, that her conclusions about memory were conflicting, and that she had exaggerated a figure and a statement from her survey of D.C. jurors that favored the defense.
As I blogged about this, the issue of false vs. true memory is more grayscale than either touters of eyewitness veracity OR Loftus types would present.

Indeed, in April, Discover magazine, talking about a similar experiment, uses the phrase "artificial blended memory." And, has some thoughts on what this means:
Old memories become active while we process new information. “You don’t learn something new without incorporating it into old information,” he says. With this new technique, Mayford will explore the basic code of memory, investigating how many neurons are required to store a memory and how cells change as memories are formed. 
It talks about the "malleability" of memory. That's also something to remember, as about no memory is likely either 100 percent true or 100 percent false.

And, that too is important to remember with this new report.

And, back to Loftus.

In the Scooter Libby trial hearing, she was, ironically, both her own best and worst expert witness. More from Fitzgerald's takedown:
When Fitzgerald found a line in one of her books that raised doubts about research she had cited on the stand as proof that Libby needs an expert to educate jurors, Loftus said, "I don't know how I let that line slip by." 

"I'd need to see that again," Loftus said when Fitzgerald cited a line in her book that overstated her research by saying that "most jurors" consider memory to be equivalent to playing a videotape. Her research, however, found that to be true for traumatic events, and even then, only 46 percent of potential jurors thought memory could be similar to a videotape. 

There were several moments when Loftus was completely caught off guard by Fitzgerald, creating some very awkward silences in the courtroom. 

One of those moments came when Loftus insisted that she had never met Fitzgerald. He then reminded her that he had cross-examined her before, when she was an expert defense witness and he was a prosecutor in the U.S. attorney's office in New York. 
The last graf is funny as hell.

The previous three are a bit more serious.

And, if we want to talk about how "repressed" memories aren't real, well, that too needs to be "nuanced."

Memories that have slid into unconsciousness still affect our brains.

So, while the finding with mice is interesting, let's not put it on the wrong pedestal, nor Loftus.

Next, let's remember that a mouse's level of consciousness is far short of a human's, and specific issues, like not having second- or third-level awareness, also have effects.

Memory is malleable, yes, and per people from David Hume to Dan Dennett, it's told in a script. But a script, to the degree it's a novel, is neither true nor false in the sense of nonfiction.

Finally, I very much support the work Loftus did in undercutting ideas about repressed memories that fed into claims of ritual Satanic abuse, a rise in claims of multiple personality disorders, and related concerns. But, let's not overstate things, and claim that memories from the past can't leave telltale signs in the brain.

To analogize from computers, deleting a Word document doesn't delete all the info related to it from your computer, as forensics experts know. Or, to analogize from Photoshop, old memories still leave EXIF tags in the brain.

And, as the lowly planarium shows us, on memory in particular and neuroscience in general, at best we're just in the Bronze Age of understanding.

Now, this all said, I'll confess I have personal reasons, anecdotal yet quite empirical, to say what I do about Loftus overreaching in her efforts.

No comments: