Sunday, November 16, 2025

Trying to Explain Memory, a Neuroscientist Gives Us Bad Descriptions and Vague Circuitous Hand-Waving

A recent NPR interview offers us an example of how empty-handed today's neuroscientists are when it comes to explaining how a brain could produce effects related to memory. You can read a transcript of the interview here. An interviewer announces that the subject will be "Why Do We Remember?" We are told that the person will be a cognitive neuroscientist specializing in memory, who has been researching memory for 25 years.  The person is Charan Ranganath who "directs the Dynamic Memory Lab at the University of California, Davis, where he's a professor of psychology and neuroscience." He has written a book called "Why We Remember."

I can understand why a neuroscientist writing a book on memory might pick a title such as "Why We Remember" rather than a title such as "How We Remember." The reason is that no neuroscientist has any credible detailed theory as to how a brain can produce any of the main phenomena of memory.  No neuroscientist can credibly explain how a brain could ever store information such as the information you learn in a school lesson.  No neuroscientist can credibly explain how a memory could persist for decades in a brain subject to such very rapid turnover of protein molecules, and subject to such constant remodeling of things like synapses and dendritic spines, which do not last for years.  No neuroscientist can credibly explain how a memory of something experienced long ago or learned long ago could ever be instantly retrieved by a brain, which has none of the things humans put in devices they manufacture to allow instant information retrieval (things like addresses, indexing and sorting).  So lacking any credible tale to tell of "how we remember," a neuroscientist might want instead to change the subject to "why we remember." 

But switching the subject to "why we remember" is a pretty futile task for any materialist, because so very much of human memory makes no sense from any kind of evolutionary standpoint.  For example, the history of musical achievement is full of cases of people who can play or sing many thousands of notes from memory (such as pianists who can play from memory all of the piano notes in a long piano concerto, or opera singers who can sing from memory long roles requiring the correct retrieval of thousands of notes, and also corresponding words). Why would anyone have such an ability, which has no survival value in the wild? And there are also many people who have memorized very long bodies of text such as the entire Quran or the entire text of an epic poem such as the Illiad or Paradise Lost ( retired teacher John Basinger memorized all 60,000 words of Paradise Lost). Why would anyone have such an ability, which has no survival value in the wild?

In the long interview on the topic of memory, neuroscientist Ranganath says scarcely a word about the brain. What he says about the brain is sometimes wrong. He states this:

"So the human brain is much more like the high-efficiency but also high-performance sports car. We're not designed to carry tons and tons of junk with us. And like you said, I don't know that anyone would want to remember every temporary password that they've ever had. So I think what it's designed for is to carry what we need and to deploy it rapidly when we need it."

The analogy is an extremely misleading one, because brains do not have any moving parts, and cars do have moving parts, such as the wheels and the steering wheel.  And by saying "we're not designed to carry tons and tons of junk with us"  and that we're designed to "to carry what we need," he is misspeaking very badly.  The average human can remember endless thousands of details about things he experienced in the past,  songs he heard, TV shows and movies he watched,  and a huge variety of things he does not need to know.  And humans can do things like memorizing hundreds of pages of text and memorizing many very long opera roles (as Placido Domingo did), and very long roles such as the role of Hamlet. It is not at all true that human memory merely "carries what we need."

Neuroscientists commit this type of error very frequently. They speak as if their ideas about how humans mind performed were derived from their dogmas about a neural basis of the mind, rather than from a thorough study of the best performances of human minds. 

Later we have more bad misstatements from Ranganath. He states this:

"Stress makes it harder to pull out the information you need when you need it. It basically kind of shuts down the prefrontal cortex."

No, it is in general not true that under stress a person's memory freezes up. When you watch the TV show Jeopardy, you see people under stress performing memory recall with lightning-fast speed. And every person who does well on a final exam at college is doing memory retrieval  very well under a stressful situation. There is zero evidence that the prefrontal cortex ever shuts down during conscious experience, except in the cases of near-death experiences, where people may have the most vivid and memorable experiences when the entire brain has shut down during cardiac arrest.  To the contrary, the neurons of the brain keep firing at a rate between about 1 and 100 times per second, during both normal waking consciousness and sleep. 

Ranganath then tries to make it sound like he has some understanding that stress damages memory. He states this:

"Then there's the issue of chronic stress, where we know that chronic stress can be actually neurotoxic for areas of the brain that are important for memory, like the prefrontal cortex and another area called the hippocampus. And that is really, I think, part of the problem that you see in people with PTSD, for instance."

PTSD or post-traumatic stress syndrome is a problem when people are disturbed or distressed by memories of traumatic experiences they had. PTSD is not some difficulty in remembering, as Ranganath has insinuated. Quite the opposite, PTSD is when people are distressed by unpleasant things they remember all too well.  

Later Ranganath gives us some nonsense trying to suggest that our memories are being devastated by the use of social media. He states this: 

"So there's a platform called Snapchat where the information literally disappears within, I don't know, 24 or 48 hours. And I think that's a metaphor for how technology can impact our memories in general."

No, actually, social media has not done anything to worsen people's memories.  Later we have more falsehood and nonsense from Ranganath when he says this:

"Those memories are being transformed little by little because the act of recalling them and sharing them changes it. So you tell me a memory, and it's no longer your memory. It's our memory because of the work that we put into in terms of transforming it."

No, actually when I tell you my memory of something, that memory of mine is not transformed; and it still is my memory. Later Ranganath repeats the unfounded neuroscientist legend that sleep does something to strengthen memories, an idea that is not well-supported by evidence. It is merely true that someone sleep-deprived might find it harder to learn as well as when he is sleeping normally.  He approvingly quotes some person saying the not-very-credible idea that sleep converts memory into wisdom.  

Ranganath sounds quite confused about memory in this interview.  He gives us no sign that he understands any neural basis for human memory. There is no such basis. No one has ever found the slightest speck of something someone learned by microscopically examining the tissue of that person's brain. Brains have not the slightest resemblance to a device for storing learned information, and not the slightest  resemblance to a device for instantly retrieving learned information or retrieving learned information at any speed. Memory is a spiritual or psychic or soul ability, not a brain ability. 

neuroscientist hand waving

Postscript
: I recently got further evidence of how false was Ranganath's claim below, suggesting people don't remember lots of old memories they don't need to remember:

"We're not designed to carry tons and tons of junk with us. And like you said, I don't know that anyone would want to remember every temporary password that they've ever had. So I think what it's designed for is to carry what we need and to deploy it rapidly when we need it."

I was watching a Youtube video about the 10 least popular television shows of 1966.  I performed these feats of memory recall relating to TV shows or TV personalities I have not seen or heard mentioned in about 60 years or more:

(1) Upon hearing the name of the 1966 TV show "Captain Nice" (canceled after a single season), I recalled its star was William Daniels. 
(2) Upon seeing an unlabeled photo of Judy Carne (an actress I have not seen on TV or heard mentioned in about 60 years), I instantly identified her as Judy Carne. 
(3) Hearing mention of an "Occasional Wife" TV show I once watched in 1966, and seeing an unidentified image of its star, I first guessed the person's name as William Calley. Later (still having heard no mention of the star's name) I thought to myself something like, "No, I think it was maybe Michael Calley." The actor (who did no famous work after 1966) was named Michael Callan, a name identical to my final guess, except for the last syllable. 
(4) The next night I had a dream of some office worker I had not spoken to or substantially thought of in 37 years. At first I failed to recall the name of this person I would only briefly see on any day, but soon correctly recalled his full name. 

The fact is that humans my age can recall all kinds of obscure useless memories from 50 or 60 years ago, memories never reactivated during such a period. There is no credible neural or evolutionary explanation for such an ability.  If your brain stored memories, you would not be able to remember any old memory you had not used for more than a year, because of the very high molecular and component turnover in brains. The proteins that make up synapses have an average lifetime of only a few weeks or less. 

2 comments:

  1. Where do you find this sort of info? Are there any similar sites to your blogs? I'm not aware of any that covers the problems with mainstream theories like you do, and you do it in a rigorous way without overly pushing an agenda, which is also rare

    ReplyDelete
    Replies
    1. Thanks for the kind comment. I keep up on the latest science news, using sources such as Google Science News, www.reddit.com/r/science, www.reddit.com/r/psychology, and sections of the Apple News feed I get on my I-Pad. There's a Neuroscience page and a Psychology page there I find useful. I also use Google Scholar to search for scientific papers on specific topics. As for sites like mine, I knew of few, partially because search engines such as Google are very bad at allowing people to find contrarian viewpoints on popular topics. I would like to know of such sites, so post a comment if you can find them.

      Delete