Smell helps the brain form memories during sleep | Hacker News Reader

Smell helps the brain form memories during sleep

medicalnewstoday.com

48 pointsprostoalex16 comments

There are a tiny handful of smells (perhaps 3) that I encounter very rarely (maybe once every few years), and for whatever reason, they bring back childhood memories of a particular place more vividly than just about any other form of recall I've ever experienced.

It's always the same smells, and it's always the same place, and it's happened enough times that I dont think I'm imagining it, and nothing else does this.

Has anyone else experienced something like this?

I happened to have watched this old scene recently from a TV show. It's apparently well established that smell is one of the strongest triggers of memory we have.

https://www.youtube.com/watch?v=TpUjnfF6ZhQ

Of course!

Definitely. Be it childhood memories, or even just something from few years ago, memories seem to work very well with memory, much better than picture-type images.

Yes. The smell of the gas meter under the stairs at my nan’s house in the early 1980s. If I smell that ever I get instant recall of staying there when I was a kid.

Known as Madeleine de Proust in french. https://en.wikipedia.org/wiki/Madeleine_(cake)#Literary_refe...

https://en.wikipedia.org/wiki/Involuntary_memory

Pierre Gloor's book The Temporal Lobe and Limbic System has an excellent discussion of this pages 316-318. I'll try to type it in by hand and stick it on pastebin later since it is unfortunately a rare out of print book (Google Books has it but it seems to show much less text these days than it used to). Gloor suggest that because low level olfaction is not able to distinguish parts of a smell it is more heavily associated with context than other senses. He notes a study where recognition of olfactory cues are not as immediately reliable as visual or auditory cues but there is very little decay over time of that recognition ability in olfaction while visual and auditory cues are recognized no better than chance after 3 months. He compares the Proust description to electrical stimulation of the amygdala in an epilepsy patient and suggests that particularly strong memories may be amygdala assisted recall (and notes that similarly vivid recall can happen with other senses but seems to be less common).

Ok, took a bit longer than I expected to get to it but here it is:

https://pastebin.com/UKdMz5zJ

I think he overstates the inability to pick out individual smells (presumably from later processing comparing them with memories). Also, I seem to do at least as well recalling odors as I do recalling images, but mostly that is because I am unusually bad at recalling images. I don't think this really affects the overall argument all that much.

I remember getting off a plane in Florida and smelling the air... and it brought back SO vividly the last time I had visited florida over 10 year before.

It was almost like playing an old video taken on a bright sunny day, in smell-o-vision.

I once had a dream of being in a ballroom of a fancy hotel, and the smell was overpowering. I was thinking that there must have been a murder and they've bleached the carpet. Turns out, my wife was already taking a shower and used a new brand of shampoo. I couldn't stand the smell of it. Somehow, contemporaneous reality entered my dream, rather than a memory of a past event.

This can have implications for skill and knowledge acquisition. Imagine if a different smell is released for every new branch of activity you perform on a computer or digital device (memorizing an anatomy text, playing a new keyboard concerto, learning to code Python, watching finance lectures). At night, you wear an EEG headband connected to a scent machine which releases scents connected with the activities of the day. The more important the task, the more you are exposed to the corresponding scent (kind of like spaced repetition but with smell).

Here are some questions I have in light of the study, assuming it can be replicated:

- Is this technique biased towards certain types of knowledge, such as tangible, locational knowledge as opposed to abstract skill acquisition?

- Can a certain smell be over-saturated with knowledge, such that it becomes less effective for knowledge reinforcement? If so, can the “saturation point” be predicted using measurable variables?

- Does this technique only work for many discrete pieces of knowledge, or does it also work for branches of knowledge? (in the example of the finance lecture, there might be 30 major discrete facts throughout the lecture, but they can be bucketed into three major concepts - it might only be practical to use three scents).

- Are more recent associations with a scent significantly more salient to the brain than older ones, or do we subconsciously cycle through the old associations with roughly equal weight?

- Can we synthesize related knowledge sets by broadcasting multiple scents simultaneously?

I am fascinated by the possibilities here. We are advanced primates after all. Our brains are “designed” for more basic things than most of what we do. Paradoxically, it seems possible to supercharge the abstract by engaging the primordial. Brb, I’m buying wallflowers at Bed Bath and Beyond.

Interesting fact: the nerves in your nose for smelling are the only part of your brain that is in direct contact with the outside world. Ancient stuff!

Source: https://www.cell.com/current-biology/fulltext/S0960-9822(20)...

I wonder if this also has some bearing on Alzheimer's. My father who has a notoriously bad sense of small developed Alzheimer's a few years ago.

I've also heard that lack of / loss of sense of smell and Alzheimer's are correlated.

pretty cool

Reply on news.ycombinator.com