Place cells: How your brain creates maps of abstract spaces
Вставка
- Опубліковано 5 чер 2024
- In this video, we will explore the positional system of the brain - hippocampal place cells. We will see how it relates to contextual memory and mapping of more abstract features
OUTLINE:
00:00 Introduction
00:53 Hippocampus
1:27 Discovery of place cells
2:56 3D navigation
3:51 Role of place cells
4:11 Virtual reality experiment
7:47 Remapping
11:17 Mapping of non-spatial dimension
13:36 Conclusion
_____________
REFERENCES:
1) Anderson, M.I., Jeffery, K.J., 2003. Heterogeneous Modulation of Place Cell Firing by Changes in Context. J. Neurosci. 23, 8827-8835. doi.org/10.1523/JNEUROSCI.23-...
2) Aronov, D., Nevers, R., Tank, D.W., 2017. Mapping of a non-spatial dimension by the hippocampal-entorhinal circuit. Nature 543, 719-722. doi.org/10.1038/nature21692
3) Bostock, E., Muller, R.U., Kubie, J.L., 1991. Experience-dependent modifications of hippocampal place cell firing. Hippocampus 1, 193-205. doi.org/10.1002/hipo.450010207
4) Christopher D. Harvey, Forrest Collman, Daniel A. Dombeck, David W. Tank, 2009. Intracellular dynamics of hippocampal place cells during virtual navigation. Nature. doi.org/10.1038/nature08499
5) Grieves, R.M., Jedidi-Ayoub, S., Mishchanchuk, K., Liu, A., Renaudineau, S., Jeffery, K.J., 2020. The place-cell representation of volumetric space in rats. Nat Commun 11, 789. doi.org/10.1038/s41467-020-14...
6) Jeffery, K.J., 2011. Place Cells, Grid Cells, Attractors, and Remapping. Neural Plasticity 2011, 1-11. doi.org/10.1155/2011/182602
7) Latuske, P., Kornienko, O., Kohler, L., Allen, K., 2018. Hippocampal Remapping and Its Entorhinal Origin. Front. Behav. Neurosci. 11, 253. doi.org/10.3389/fnbeh.2017.00253
8) Laura Lee Colgin, Edvard I. Moser, May-Britt Moser, 2008. Understanding memory through hippocampal remapping. Trends in Neurosciences. doi.org/10.1016/j.tins.2008.0...
9) Meredith Minear, Tesalee K. Sensibaugh, 2020. Neuroscience of Navigation. doi.org/10.1002/9781119458555...
10) Moita, M.A.P., 2004. Putting Fear in Its Place: Remapping of Hippocampal Place Cells during Fear Conditioning. Journal of Neuroscience 24, 7015-7023. doi.org/10.1523/JNEUROSCI.549...
11) O’Keefe, J., Burgess, N., 1996. Geometric determinants of the place fields of hippocampal neurons. Nature 381, 425-428. doi.org/10.1038/381425a0
12) Robinson, N.T.M., Descamps, L.A.L., Russell, L.E., Buchholz, M.O., Bicknell, B.A., Antonov, G.K., Lau, J.Y.N., Nutbrown, R., Schmidt-Hieber, C., Häusser, M., 2020. Targeted Activation of Hippocampal Place Cells Drives Memory-Guided Spatial Behavior. Cell 183, 1586-1599.e10. doi.org/10.1016/j.cell.2020.0...
13) Wohlgemuth, M.J., Yu, C., Moss, C.F., 2018. 3D Hippocampal Place Field Dynamics in Free-Flying Echolocating Bats. Front. Cell. Neurosci. 12, 270. doi.org/10.3389/fncel.2018.00270
Icons by flaticon.com and biorender.com
_______
Socials:
VK: atpsynthase
I have old cognitive maps in my brain from I think my childhood. I sometimes walk trhough them in my dreams and it's funny because everything is recognisable and has the character of the place it resembles but is totally blown up in size and filled with extra stuff.
Very much relatable
Yea it's a strange thing if you think about it... That in your dreams there are some aspects of real experiences mixed with stuff that completely doesn't belong there. Like mixed and confused maps the cortex is confabulating together or something...
@@alexharvey9721 maybe it's reality that's being confabulated
@@Beanybag2 no, its the brain reusing the same circuits and neurons for other tasks, ever wonder how the brain can use so little energy and compress so much information, it is reusing what it knows.
which is why when you are learning new things, the easiest way to learn anything is to relate to things you already knew before. This is a thing that the stupid schooling system totally doesn't do, because it randomly reorder the schedule of things to a fixed order (that's good for them, not for you, their problem is one of logistics, lots of kids have to get the same content at the same time, but that's bad for individual kids).
I'm a self-learner, you have to learn things in a order that makes sense to you, much like exploring a physical place. You can't just pick a book and go page by page, it doesn't work like that, specially not like a traditional school curriculum, that doesn't even make any sense whatsoever.
A "sparse" neural network, as Numentia suggest
Some of the other students in my intro to neuroscience class were asking my professor about place cells. I recommended your video, since it helped me so much. I found your channel a year ago and started watching as a hobby and now I'm going into the field! So fascinating. Love your work!
Really interesting! Would love to see this studied in people who play competitive videogames / speed-run. How their brains keep track of the environment and objects in game vs real life would be particularly interesting.
so cool to see this topic being materialized. one of the things that led me to neuroscience was this elusive subjective feeling of one location in space feeling like it's registering as a different location. standing in the street outside my house, for example, at night it feels as though it's a physically different place than the same place in daylight. i can remember a hard to explain feeling of locations, or even episodic memories, having a quality of being mapped, ie my street at night is, say, up somewhat, to the left this far, and a bit further back than a representation of my street during the day. that isn't to say that there's a strong empirical hypothesis for some kind of within-brain-circuit interoception or spatial navigation. but the experience and the subsequent questions became sort of an internal cosmology of the mind, and this is my first introduction to actual research that touches on it. great channel! shit like this keeps me excited/motivated for grad school - earning my bachelors in neuro next week
I definitely have felt this before, never had it put to words. Awesome, thank you.
Its been 2 yrs. What’s your progress?
Thanks for such high quality neuroscience content! This channel is truly amazing.
hi artem, im a neuro phd student. i've worked on the hippocampus and fear conditioning. this is the best explanation on the topics i've seen.
Lovely neuroscience content 💙
Exactly what I’ve been needing to see to help an internal line of thought progress. Thank you so much.
I'm a programmer, and the fact that I describe what I do as navigating spatially through things that are like total abstraction always seemed odd.
Now I have a way of describing what I feel happens internally to other people.
It doesn't only feel spatially, it is spatially because it literally uses the same structure with the same purpose.
Crazy, considering the variables they represent must change over time with learning. I wonder if place cell representations change universally (in all contexts) as a function of *learning abstractions of more important environmental variables.* If so, do we have to re-orient our place cells upon returning to a *familiar* environment after learning has occurred in the interim since last visit? Or, does learning update place cells in a way that is somehow consistent with prior representations? Does the hippocampus have much neurogenesis in adulthood? If not, this seems to point at a unique struggle for learning in adulthood: substantial learning that is incoherent with prior understanding might seriously disorient us in familiar environments as we might be forced to re-calibrate place cell representations in a complex way, given that new environmental information cannot simply be appended through growth of new place cells.
Great video! Please make more of these
This video is pure gold. There are so many people out there that still don't know this channel but they are craving contents like this
I’m One of them
I need MORE! Such great videos! I appreciate you putting the references to the studies on screen as you are discussing them
Great video, thanks a lot! I've been reading about spatial navigation in animals for some time, but the concept of the place cells being really used for task-specific variables blew my mind. Hope you do more of these!
Thank you very much for your time Artem :)
Amaizing topic. Please keep up the great work :D
jesus christ what an amazing quality...keep up the great work Artem. Cheers from Italy
I discovered your channel yesterday: its pure gold
I LOVE THIS VIDEO, IT'S SO AMAZING!
From the concept that's presented to the very own editing and the many animations in the video, it is stunningly beautiful.
Fascinating information like this is exactly why I love watching your videos
Dude, your Videos are stunning! Can’t believe you’ve only got 7k subscribers so far. Keep it going! ✌🏻
Fascinating subject! Great video man, you're doing a great job with those visual explanations, they're really on point.
Awesome videos! These are the hottest scientific works out there today. Thanks for such in depth discussions of these papers.
best neuroscience content on the internet. love the math lens you approach it with
Wonderfully clear a inviting to learn. Those animations are improving wildly fast, at least to me.
That hyppocampus partially transparent animation is absolutely gorgeous to me. If I were you I would even try to go in that direction in general for the channel. The transparency to me feel cozy, like, 'The whole topic may be super complex, but for now let's just focus on X, and then the whole topic appears much more easy to handle'. Or maybe better put, 'don't worry about stuff around, it is not necessary, you are allowed to focus only on what it is being transmitted, don't sail away'.
Awesome video to understand better my spatial memory lessons from psychobiology classes! Good work!! Keep going!!
Maybe "setting cell" is better than place cell? As in a setting in a story. It also has a dual meaning in terms of settings on a device, because they determine behavior to some degree.
Fantastic job. Got this in my recommended which means you’re in the algorithms favour. Good luck, hope this channel grows!
Very underrated channel. Keep up the great work !
Probably the best video on place cells I've seen :)
have to write an essay on this stuff for my neuroscience degree. Been reading every paper not knowing wtf they've been talking about . This video was great, thanks!
now I have to figure out head-direction cells, grid cells , interneurons and the 8 other cells on my own :(
Your videos are a mine gold. Please make more videos regarding paper divulgation! I love the field but don't have the time to research it properly, you're doing a fantastic work conveying these things.
this is incredible! thank you for explaining this! :D
Sheldon Cooper! Did you really use a lil Lorenz's Attractor as channel mascot?! ... let me tell you that you have out-nerded most of the community with that single move. Kudos!
In his book A Thousand Brains, Jeff Hawkins proposes that this capacity to map abstract sequences is akin to what the whole neocortex does but instead of using "place cells" it uses "cortical columns". I very much hope you touch on this topic further down the series.
Thanks ;)
It's actually an asymmetric Lorenz attractor! I went with this modification so that it looks more like a brain (it's actually supposed to represent the cerebrum and the cerebellum, haha)
If you're interested, here's the reference paper: www.mathnet.ru/links/c01a6f234d34466e0f21dd8a840d1ba6/svmo700.pdf
Unfortunately, it is in Russian, but you can take a look at the figures and the equations
Your videos are so good, please keep doing them!
Great video, thanks!
Despite a few Nobel prizes being given out relating to grid/place/boundary cells etc it still feels like one of the most important discoveries of recent times and very underrated. In many respects, it's one of the first real clues to the real function mechanisms and structure of the brain.
How in the world does your channel have such low views?! Your verbal descriptions, visual aids and overall knowledge.. damn. You got a sub from me
Wow simply amazing. You explain very well. And videos very high quality.
I've got an essay on the cognitive map hypothesis and this has been extremely helpful!!
Excellent video. Thank you.
amazing job this channel gonna blow one day
Glad I found your channel
your video got in my recommendation page. good luck!
Awesome content! Can't wait for more :)
not only a rare and new area, also your language and video editing are both amazing. a pearl in yt pile of nonsense.
Great video. Thank you for explaining those experiments in an easy to understand way.
Thanks!
Wow fantastic video!
Nice explaination , Thanks!
Fascinating video. The non-spatial mapping of place cells reminds me of a Vox video (that I didn't watch) about why you have to turn down music when you are trying to navigate. Perhaps the place cells can get distracted from spatial mapping and make it difficult to navigate.
mindblowing stuff!
Staying tuned. Bring some more brain stuff.
Great video.
This is brilliant! I hope this comment can encourage the algorithm to spread it
What a great video! Thank you!!
As a beginner UA-camr, these animations are inspiring!
What program/software do you use?
Thanks!
- Eduardo
Thanks!
This is primarily Adobe After Effects for animations and Premiere Pro for editing and stitching them all together
🎵It's the cyyyycle of Liiiiife🦁
*UA-cam Algorithm presenting this channel to the masses
Great video, thank you so much! 🦑 (I had a stupid question by the way, I'll search more about it: Can overlaps and interactions between these spaces (or maybe different functioning in the place cells) cause synesthesia?
That's not a stupid question at all! I think it's a very plausible hypothesis and wonder if any studies have been done or planned, did anything come up in your search?
@Kaleb Peters I've only found one study (my research skills are terrible) & it's talking about a spesific type of synesthesia called calendar synesthesia. It's an interesting article but I couldn't find any study specifically concern about the hypothesis that I thought.
@@aelfygva interesting! Perhaps we need to find some grad students 😂
I have synaesthesia and this question makes me want to make 180-degree switch from a marketing career to a neuroscience career!
Hey man, thanks you very much for the content, I was thinking that you would bring to us your sistem of zettelkasten and how you organize it
13:03 that's exactly how I feel when I'm studying music and trying to navigate in the harmonic field. I pretty much know the CMaj field like a map, I can play things. But I can't play random things, its like I'm in a different place.
This was really cool. It got even cooler when you mentioned the frequency matching. I have synesthesia, so I can "see" sounds in my mind, if you will. I've always wondered how my brain is putting these images together, as it's the same every time. ...Could place cells be the answer?
Wow this is so intresting! Is there a similar mapping in social memory? For example, when we first meet someone we tend to conceptualize them in therms of people we already know, which is similar to how we map similar places with the same neuron.
This reminds me of multidimensional latent spaces.
The definition of the room have different aspects; the features of the room, mental (also physical when moving) location of the mouse in this space, dimensions of the room etc.
Could anything then be defined in an "abstract space" way? So anything from a sound frequency range to how square a circle is, with the observer's ralation to these/in this space mapped as well
Awesome
Very interesting topic, particular for any PKM-enthusiasts.
And nice visuals to go with the video, too.
May I ask which tool you are using to do these, and how long you've been learning doing these kind of videos?
How is this Video Related to PKM?
I wonder how important the order of doing things is for the place cells. For example, maybe the cells that changed in the experiment with the rats that were shocked changed in such a way as to indicate where it would be safe to be/move through.
Thanks for making this great video! I am researcher on mobile robots, and I am very interested in the problem about how a robot represents and remembers the surrounding environments. Could you please recommend some papers that can explain how does our brain represent the environment? Thanks
thanks
Have there been experiments testing how the place cells remap when the subject starts out in environment A, then gets moved to environment B, and then gets put back in environment A? Do the place cells remap back to roughly where they were initially, even if there's, say, a several hour delay before going back to environment A?
Under the idea that remapping is caused by the place cells encoding contextual data about the environment one would expect that the remapping does return to roughly where it was initially, but it would be interesting to see what actually happens.
It's funny because the first results that you mention seem to strongly suggest the idea of a mental model of the environment, so I was wondering how they might be interpreted from an ecological dynamics point of view.
But the section about remapping reverses all that, doesn't it?
If the brain responds not to specific places in the space, but to local characteristics of the environment, it fits perfectly with the idea that only affordances are perceived, not just simple physical characteristics such as positions.
Like, the proximity of the electrode affords, say, receiving a shock and walking around the electrode, and the perception of this set of affordances manifests as the firing of a specific group of neurons.
And so it doesn't seem far-fetched to say that in the first experiment the correlation between physical positions in the space is due to the fact that different positions bring different affordances fields (such as "going east" and "going north" in the southwest corner)
I admit that this view is not as neat when analyzing the sound reproduction setup, but still, I think it makes sense to think of affordances such as "letting the lever at this place"
I mean, those are just speculations on my part, but I feel like generally, bayesian credences could be thought of as "betting affordances" (betting "the sound I'm hearing is the sound I heard before, and I will collect the reward)
Were the place cells learning the frequency space representation, or were they actually learning the joystick angle? For example, if the rat was deaf, it could still develop an association between the joystick angle and the reward. I wonder if frequency is the relevant variable here? My guess is that both variables are learned because they are correlated. If you were to do this experiment, then put ear plugs in the rat & reset the joystick, I assume the rat would still be able move the joystick back to get the reward. The study would have to randomize/reset the joystick-to-frequency mapping between trials to control for this. I'm sure the study considers that, but I think it's an important distinction.
I’m curious about how place cells encode driving, since (esp on highways) there are very few places that are important for your route. Does your mind map just have those intersections and everything in between is nonexistent?
Or better yet, do your place cells encode information about computer screens, like what desktop you’re on or what website you’re at?
Came for SME, stayed for the fantastic neuroscience primers.
Ja stimmt, hatten wir bei unserer Laborratte auch!
I remember reading of a similar experiment that found that cats' brains move the world around them instead of moving them thru the world; would you have any more info about such differences?
Could place cells' activation and interplay with sound frequencies have to do with how echolocation functions?
So place cells can serve as models for locations in some kind of abstract parameter space as well...
I have this one street intersection that occupies two distinct place fields in my brain. The first is from passing through it while semi-lost on a detour home. The second is from moving to that town and passing through it daily. Brains are weird
What I don’t understand is why I can still do this with aphantasia. I suppose blind people can navigate quite well too, using their other senses, but I still use visual cues, despite only seeing my eyelids when I try to visualise. Makes me think that we might map out our environments mentally in a manner that is seperate to vision, but rather combines a bunch of different stimuli to create a hard coded map, almost like how you can download a map for a game. In that sense, I mean that perhaps the brain stores the data in a means that can be “loaded” as required by whatever senses are relevant for those assets.
I've definitely read a related study a long time ago about something similar, about bats navigation in total darkness using echolocation still activating visual areas of their brains even though they're navigating by sound, a mental map is still spatial I suppose? I wish I could remember more about it!
11:10 huh, I wonder if this mechanism might explain why I don't have a sense of direction. I never had a place that felt safe or familiar, especially not as a child, so I have would have no mental framing with which to determine a place where an action should take place, and therefore nothing to build or categorize. I get lost as soon as I open the door to the bathroom at like 99% of the places I go to, even at home. Like "Oh shoot, do I turn left or right?"
For the sound experiment, how do we know the "place" was the location of the joystick?
s/o from South Africa
Sounds like place cells allow immersion into mathematical spaces. I wonder if and how trained mathematicians use place cells to explore mental constructions of calculus. By translating a problem domain into geometry, we could literally make use of the intuitive, sophisticated machinery of the hippocampus to solve it. Maybe that's what Einstein did when he flew on the light ray.
I wonder how it would feel like if I had my place cells stimulated. Would it be as if I would be somewhere else? How much could it trick me?
It may be misleading to think of the rat brain’s “world” as being mapped into geographic (XY) or object (corner, lever) spaces. Rats have no need to go to Florida for the winter, or the grocery store. Their activity is based on drives - food, escape, home, sex, etc. So the end of the tunnel is “place to turn around”, and 500 Hz is ‘get reward” (and would be connected to the lever place cell). That explains the shifting of place cell activation not being highly XY or even XYZ related. In the case of the frequency-reward lever task, one might see the frequency place field neurons activated first, then when 500 Hz is activated, the place field might shift to one indicating the position of the lever relative to other objects.
So hippocampus cells are basically RAM storing floating point variables, it seems.
6:36 I am really weirded out and curious to know if would be like if someone stimulated your place neurons. Like would you suddenly go like “oh my gosh I just teleported to Starbucks! I need to start asking for a grande white chocolate mocha!”
How about when you drive somewhere new and it takes forever to get there, then the return trip feels like it takes less time, even though you go the same speed. Whats that about? Less mental effort?
Wwwowww... Location to frequency sounds like transfer learning in AI
Can anyone point me to where I could find how to highlight brain regions in a transparent model (mouse) like at 12:20 ? I would really appreciate it..
This was done using a python library called Brainrender
Here's the link:
github.com/brainglobe/brainrender
I wonder if I can program a robot in a video game to navigate along a planet surface using an analog of this
I know we gather a lot of data studying rats, and I'm grateful to the scientists that have laid all the groundwork for these discoveries... but I can't help but feel bad for the poor rats with wires in their brains :( they must get so disoriented when being tricked into thinking they're somewhere they're not
So we actually stick wires in mouses brains? We shouldn't be doing that...
If we have so little regard for life that we're willing to kill or harm to learn, we may as well go back and give Dr.Mengele a Nobel Prize for learning so much!
Man, I feel like I don't even exist. As if "myself" is just a childish dream, a cruel prank played by the 100B neurons inside my skull.
you've just come to the same conclusion as buddhism!
I wonder if AI could benefit from this idea of place cells
Context match cells...
I can barely navigate around my block. Anyone wanna scan my brain and see what's wrong with meeee?
I love science and appreciate the knowledge obtained through experiments but I can't get over the malevolence and despise of how we treat animals :(
no rats were harmed in the making of this video?
Pfff, must be nice...
replace mice with people and rule the world. oh... sorry its already done!
how "your" brain does x, but its a picture of a mouse.. :)
The "not just place" nature of the place cells reminds me of Kurt Lewin's concept of "hodological spaces." en.wikipedia.org/wiki/Hodological_space
@strealm
Lovely video! Looking forward to more on spatial navigation 🧭