Unique memory 'barcode' allows bird to find stored food

A fantastic find for foragers...
28 March 2024

Interview with 

Selmaan Chettih, Columbia University

CHICKADEE.jpg

Chickadee

Share

Memory is a funny thing, and comes in various different forms. One is episodic memory, the ability to store information about unique experiences and when they occurred. This kind of memory is pivotal in animal species that squirrel away - and remember the location of - thousands of food items during the good times for when it’s much less plentiful. And now scientists from the University of Columbia have made a breakthrough in understanding how they manage this feat of memory. By monitoring the brain patterns of chickadees, small members of the tit family of birds, they saw that at the moment of making a cache, the brain’s hippocampus activated a unique assembly of neurons corresponding to each individual food store.  Selmaan Chettih explains…

Selmaan - It looked to us like when the bird made a cache, the hippocampus activity entered this different kind of state. It just looked really different than when the bird was just navigating around. And in this sort of different caching state, there were a couple things going on and the most interesting one was that there was kind of a unique pattern that happened for each cache. So each time the bird cached, you got sort of like a unique kind of label and this is what we were calling the barcode. It's a different pattern of activity that occurred for each cache.

Will - If these birds are going round and storing thousands, even hundreds of thousands, of caches per year, do they have a unique barcode for each one?

Selmaan - Yeah, that's exactly the thought. You know, like we're studying them in the lab, but the finding that they can generate these really unique labels like that was pretty new. We didn't expect that before we went in. And it really matches well with what we think these animals are doing in the wild because they have to keep track of so many different things and you don't want those different memories to interfere with each other or blur into one another.

Will - Yeah, that does sound like an immediate advantage for this memory system because if you do have thousands of individual caches to remember, you can't afford to blur those memories or else you starve.

Selmaan - Yes, exactly. Yeah. Or you might think, you cache a food item on the left and then you cache another food item on the right. Then if you blur them together, start thinking there's a food item somewhere in between those two. But you really have to keep all the memories separate and very precise. And we think that this barcode mechanism might be a way for them to do that.

Will - Do we know why this has become so distinct and separate from their regular navigational memory? Because I would've assumed that caching in this way requires knowing how to navigate to individual sites.

Selmaan - Yeah, that's a really interesting question. So another striking thing about these sort of barcode patterns is that they're a really transient thing that happened when the bird cached, which is like a second long. So it's like a quick blip almost in the signal that you have to know exactly when to look to see it. When we first started, that wasn't the framework we were going into the study with at all. There's been a lot of work on the hippocampus of many species for a long time showing these famously called 'place cells'. And these cells represent where an animal is. And the thinking was based on previous work that when the animal forms a memory, there'd be a change in place cells and that maybe you would have more place cells representing a location where you made a cache. We didn't really see that and it caused us to sort of look in a different direction. It's a really interesting question. I think it's kind of an ongoing question of research first, now. The speculation is that there's really a different kind of memory between remembering something that's kind of like a stable, like a feature of the environment. Like there's a lot of food here or in this place, this is like a scary sort of part of the environment or this is the regular path I take to work. That's a different kind of memory than like an individual event that just happens once and you kind of want to store that episode. And so we think that maybe the brain has different kinds of mechanisms for these different kinds of memory.

Will - When you say this episodic memory flashes for a short time, it almost feels like when you go to the shops and come home and see a gap in your fridge store and you think 'milk, I forgot to get milk.' It feels like the inverse of that, that enough contextual prompts kind of ignites short episodic memory for them.

Selmaan - Yeah, it's very much like that. We've kind of looked at them, they form the memory and their recall of the memory is like they're basically back at the site and they're about to retrieve that cache. But that's actually like something that we're really interested in looking forward to in the future. Sometimes you recall a memory like that, but sometimes, to use your metaphor, you're in the grocery store and you remember then, 'ah, I need milk.' And we actually haven't really shown yet if this barcode pattern is also underlying that kind of recall. So that's kind of like an ongoing project that we're working on now.

Will - Do we think this is unique to chickadees? Could this expand out into the wider animal world?

Selmaan - We certainly don't know the answer, but we think it could be more widespread. There's no evidence particularly that this would be really specific to chickadees. Caching is a really clear behaviour. It's really easy to identify when a memory is formed by looking at when a bird caches. Usually it's very hard to figure out exactly when an animal is forming a memory. So we think it just could have been hard to see in a lot of previous experiments. And actually there are a couple studies actually in humans that are sort of hinting at something similar. So we're excited about the idea that this could really generalise across organisms.

Comments

Add a comment