Get the FREE DIGITAL BOOK: The Case for Killer Robots
Mind Matters Reporting on Natural and Artificial Intelligence
senior woman with family photo on tablet pc screen

What Neuroscientists Now Know About How Memories Are Born and Die

Where, exactly are our memories? Are modern media destroying them? Could we erase them if we wanted to?

At one time, neuroscientists believed that there must be a “seat” of memory in the brain, something like a room with a door marked Memory. They settled on two structures called hippocampi, on either side of the brain’s base. The illustration shows the the hippocampus of the right hemisphere (public domain).

But memories turned out to have no fixed address. Neuroscientist Matthew Cobb, author of The Idea of the Brain (2020, excerpt here), tells us,

But the hippocampuses are not the site of memory storage. Rather, these brain regions are the encoders and the routes through which memory formation seems to pass. The memories that are processed by the hippocampuses seem to be distributed across distant regions of the brain.

Matthew Cobb, “Where Do Our Memories Live?” at The Scientist (May 1, 2020)

Synapses, the connections between neurons, become stronger with learning. If they fire together, they wire together, as the saying goes. That includes memories. Still, Cobb acknowledges,

The search for memory’s seat in the human brain seems even more complex today. Researchers now know that memories may not be found in a single place, but precise cells and structures play a key role in memory formation and recall. Memories are often multimodal, involving place, time, smell, light, and so on, and they are distributed across the cortex through intricate neural networks. Our brains might be like computers in terms of how they sometimes process information, but the way we store and recall our memories is completely different. We are not machines, nor are we like any machine we can currently envisage.

Matthew Cobb, “Where Do Our Memories Live?” at The Scientist (May 1, 2020)

That really shouldn’t be a surprise. The cells that currently comprise our brains are all alive but none of the parts of a computer are. Differences like that would probably make a difference in how each entity does things.

How do our neurons keep track of time and place in memories?

Some interesting information has come to light about how our memories encode realities of time and place. For example, experiments with rats suggest that our neurons use a sort of virtual map and a “time stamp” to encode a sense of time in our memories.

Studies at the Kavli Institute in Norway focused first on the sense of place. In studies of rats, Edvard Moser and May-Britt Moser discovered “grid cells,” neurons that create a virtual map of an area that an animal is crossing when the animal crosses the part that the cell represents. In a discovery that earned them a share of the Nobel Prize for Physiology or Medicine for 2014, they found these cells in the medial entorhinal cortex (MEC), which exports neurons to the hippocampus, from which memories are distributed.

But what of time? As Catherine Offord tells it at The Scientist, one of their students, Albert Tsao, hypothesized that a brain region beside the MEC, the lateral entorhinal cortex (LEC), fed time information to the hippocampus. Another neuroscientist, Marc Howard, used a mathematical model to show that neural circuits might use “time stamps” while memories are formed, to enable the brain to sort past experiences by when they happened. Meanwhile neuroscientist Howard Eichenbaum of Boston University showed that, in a rat’s brain, certain cells fire in sequence, one after another, during completion of a task, “as if the hippocampus were somehow marking the passage of time.” Later experiments with monkeys showed the same phenomenon.

Applying this information to humans, however, hasn’t been straightforward. One finding of interest has been that, in general, our neurons’ sense of time depends on whether events are perceived as interesting or dull. Not surprisingly, we have a much better memory of a few minutes of terror than of months of tedium:

That the sense of time in episodic memory might be dependent on neural activity rather than on a traditional clock reinforces some researchers’ belief that the brain perceives time rather differently from how people imagine it to. Buonomano and New York University neuroscientist György Buzsáki have independently argued before and since Tsao’s work that neuroscientists should rely less on preconceived notions of time and instead think more about how time-related information might be used by the brain. “The sole function of memory is to allow animals to better prepare for the future,” says Buonomano. “Sometimes the field forgets that detail.”

Catherine Offord, “How Time Is Encoded in Memories” at The Scientist (May 1, 2020)

Of course, it may not be correct to think that the “sole function” of memory is to better prepare animals for the future. For one thing, that type of reductionism may not help us understand human memory, which meets a variety of needs. But, in general, this area of research may help us pinpoint causes of specific types of memory loss in an aging population and point to possible therapies.

Are modern media really destroying our memories?

Modern media are certainly changing our memories, says Jared Cooney Horvath, an educational neuroscientist at the University of Melbourne in Australia.

Neuroscientists have known for over a century, thanks to the research of psychologist Hermann Ebbinghaus (1850–1909), that how we encounter information impacts how we remember it. We remember less, for example, if we encounter a mass of information all at once than if we learn smaller portions over shorter periods. Binge watchers, take heed:

Unfortunately, when information exposure is constant and ceaseless, our ability to hold onto information naturally diminishes. In fact, as colleagues and I demonstrated in a recent study, individuals asked to binge-watch the entire season of a television series remembered significantly less about the plot and characters than individuals who watched the same series on a nightly or weekly schedule. Human beings have always had a limit to the amount of information they could meaningfully encode in any given day. Modern technologies have not changed this; they simply push us beyond this limit more frequently than media of the past.

Jared Cooney Horvath, “Is Modern Media Destroying Our Memories?” at The Scientist (May 1, 2020)

At one time, binge consumption was much more difficult. Media were not as readily available and people usually had to wait a week or a month for the next instalment or episode, which gave them plenty of time to reflect on the last one.

But then there’s the “Google effect”: Horvath notes that in times and places where everyone can readily access the internet, we retain many fewer facts in our heads. As he says, we “outsource memory storage.” Cut off from our phones, we’d be pretty vulnerable.

One problem is that it’s not clear in exactly which phase of neuron activity memories are formed. Some researchers, Horvath tells us, think they form in the encoding phase, others in the storage phase, but he argues for the “final retrieval” phase:

Put simply, memory is constructive: the more you retrieve a memory, dredging it up from the depths using your own cognitive faculties, the easier it becomes to recall in the future. This is likely why we remember so many TV jingles—we retrieve these songs each time we sing them—and why we don’t remember so many ubiquitous logos—very few of us have ever retrieved these images.

Jared Cooney Horvath, “Is Modern Media Destroying Our Memories?” at The Scientist (May 1, 2020)

In Horvath’s view, our memories capacity is not undergoing a change due to high tech and AI—but the way we use our memories is.

What if we want to erase our memories?

We’ve talked about encoding, storage and retrieval of memories. But some people are crippled by dreadful memories. McGill University psychiatrist Alain Brunet tries to help people forget. In a recent attempt to help 60 volunteers forget abandonment or betrayal (or, as he terms it, “adjustment disorder”), he used detailed recollection under the influence of a blood pressure pill. He claims to have succeeded in that volunteers told him, “I feel like I’ve turned the page. I’m no longer obsessed by this person, or this relationship.”

His work is therapeutic. But other neuroscientists are working on fixes to enable us to just forget things, period:

What was once purely science fiction is moving ever closer to clinical reality. Researchers are working on techniques and drugs that might enable us to edit our memories or at least seriously dull their impact — to make the intolerable bearable — by, say, swallowing a pill to block the synaptic changes needed for a memory to solidify. A pill that could be taken hours, even months or years after the event.

Sharon Kirkey, “If you could erase the worst memory of your life, would you? Scientists are working on a pill for that” at National Post (November 1, 2019)

These researchers are banking on a theory in psychotherapy that memories can be changed for a few hours after they are evoked (memory reconsolidation).

But there’s a problem. We might want to forget many things. But is it good for us to do so? Kirkey asks,

Would it bleed into not-so-disabling disorders? If someone misbehaves at a cocktail party and would really sooner forget what happened, is that an appropriate use? Isn’t it good to be embarrassed by your past behaviour, to keep you from doing it again?

Sharon Kirkey, “If you could erase the worst memory of your life, would you? Scientists are working on a pill for that” at National Post (November 1, 2019)

Indeed. Addiction recovery groups explicitly incorporate bad memories into a therapeutic technique called “Remember when?” The person in recovery is encouraged to remember the dysfunctions of a life dominated by substance abuse or addiction now and then, to protect against resuming that lifestyle.

For now, experiments in erasing memories from mice have had a mixed record when translated to humans, Kirkey reports. So we may be stuck with just learning from our bad experiences after all. That’s not the only function of our memories but it’s an important one.

You may also enjoy:

The mind is the opposite of a computer. Matthew Cobb, a materialist, only scratches the surface when he explains why your brain is not a computer. (Michael Egnor)

We will never “solve” the brain. A science historian offers a look at some of the difficulties we face in understanding the brain. In a forthcoming book, science historian Matthew Cobb suggests that we may need to be content with different explanations for different brain parts. And that the image of the brain as a computer is definitely on the way out.


Why the brain is not at all like a computer. Seeing the brain as a computer is an easy misconception rather than an informative image, says neuroscientist Yuri Danilov.

Mind Matters News

Breaking and noteworthy news from the exciting world of natural and artificial intelligence at

What Neuroscientists Now Know About How Memories Are Born and Die