Decapitated worms regenerate heads with old memories

Yup, you read that right.

The worm in question is the Planarian flatworm. Compared to C. elegans, the flatworm doesn’t get as much love in neuroscience. But to regenerative medicine, it is a truly incredible gem.

You see, planaria harbors adult stem cells that imbue them with astonishing regenerative abilities. If you decapitate a worm, the tailpiece can regenerate a COMPLETE head with a fully functioning brain within a few days. What makes this even more incredible is that – unlike C. elegans that have a distributed nervous system* – planaria has a centralized brain in the head region, just like you and me. Planarian neurons also talk to each other in ways similar to ours, with the same majority of neurotransmitters. They can also learn simple associations and keep the memory. Oh, and they look like this:


Science is only starting the tease out the mechanism behind planarian’s regenerative abilities. But to me, an even more tantalizing question is this: what happens to all the memories stored in the chopped-off old brain after a new one takes its place? Does planaria revert to a state of tabula rasa, or does it carry with it memories of its merry old life?

Shomrat T and Levin M. 2013. An Automated Training Paradigm Reveals Long-term Memory in Planaria and Its Persistence Through Head Regeneration. J Exp Biol. Doi: 10.1242/jeb.087809

This is the question this paper set out to answer. Planarians have this feeding quirk: when fed in a new environment, they tend to be more “cautious”, taking longer to go after the yummy liver morsels that they love. Once they’ve been fed many times in the same environment, they “feel safe” and go right after the food.

Screen Shot 2013-07-11 at 1.46.50 PM

Scientists trained a group of planarians to associate feeding with a rough-floored Petri dish (pictured on the left)– significantly different from the smooth-floored one they’re kept in. 4 days after the final training session, scientists put both trained and untrained worms into the rough-floored dish, with one extra twist: the food was now illuminated by light shining through the dish. Planarians hate light; in order to get the food they’d have to be VERY comfortable with the environment.

As you can see from the red line below, 4 days after the last familiarization session, trained (right side) planarians took much less time to grab the food compared to their untrained (left side) peers. This was also observed 14 days after training (black line), meaning that the memory of the familiar rough-floored dish lasted at least that long.

Screen Shot 2013-07-11 at 1.47.19 PM

Results! Fig 3 from the paper.

Scientists then decapitated the worms (both trained and untrained), and waited patiently while the worms regrew their heads. Roughly a week later, scientists pre-fed the worms to satiety in their home dish, and 4 days later tested them for memory of the rough-floored Petri dish. As you can see from the green line above, the trained-and-beheaded worms seemed to have lost the memory of the feeding environment, taking just as much time to go after the food as the untrained-and-beheaded worms.

Is the memory completely lost? Worms trained to associate food with an environment can re-learn the same association much faster than naïve-untrained worms. (You can brush up on a rusty skill much faster than learn it from scratch.) To see if a hint of the old memory remained, scientists pre-fed both trained and untrained decapitated worms in the rough-floored Petri dish. To the familiarized worm, this is a previously encountered environment; for the unfamiliarized, this is a first introduction to the dish. Previously it took these worms 10 days of training to form a food-environment memory, so this one-time training session shouldn’t result in significant learning.

Scientists tested the worms for memory of the rough-floored dish 4 days later. As you can see from the blue line in the figure above, the trained worms (right) quickly re-familiarized with the feeding environment, taking much less time than the untrained ones (left) to feed. This suggests that maybe the memory is not all gone – it’s just not easily accessible without reactivation.

Screen Shot 2013-07-11 at 4.18.13 PM

Here’s a summary. After decapitation and head regrowth: Pre-feeding in home dish = no difference between trained and untrained. Pre-feeding in rough-floored (training) dish = previously trained worms remember better.

What to make of all this? Can old memories be re-grown along with the head? My first reaction was maybe the decapitated worms had some sort of modification going on in the peripheral nervous system, which resulted in their sensitization to food-environment learning. By comparing the blue and green lines, you can see that both untrained (left) and trained (right) worms learned and remembered the feeding dish after one-time training. However, peripheral modifications doesn’t explain why previously trained worms learned and remembered the feeding environment BETTER.

The authors also designed their experiment very cleverly. In the test dish, the worms had to recognize food and the feeding environment, and make a decision to move towards it against their natural preference (stay away from light). This cautious approach strongly argues that the brain is involved, ie it’s not a simple reflex.

Memory is stored in neuronal communications in the brain. Could it be that a rough correlate is also stored in stem cells of the planaria? This way, when stem cells divide to form a new brain, the memory would return. Although this scenario sounds like sci-fi, it actually could occur through epigenetic mechanisms (changing the pattern of gene expression). There’s just not a lot of evidence for it yet.

While still skeptical, I have to admit the idea that memory can survive decapitation and brain regrowth is tantalizing. Although us humans don’t have planarian’s outstanding regenerative abilities, we do share similar neural transmission mechanisms. What does this study tell us about our own memories?

*Edited for accuracy: many thanks to all the people who pointed out that C elegans do not have “many little ‘brains'” as I first put it. Very bad wording on my part. C elegans have a ring of ganglia (clusters of neurons) but not a centralized brain. For more please refer to the link in the comments.
Shomrat T, & Levin M (2013). An automated training paradigm reveals long-term memory in planaria and its persistence through head regeneration. The Journal of experimental biology PMID: 23821717

14 thoughts on “Decapitated worms regenerate heads with old memories

  1. So that’s what they look like!! Way cooler than C. elegans.

    But this isn’t true:
    “What makes this even more incredible is that – unlike C. elegans that have many little “brains” scattered throughout the creature – planaria has a centralized brain in the head region, just like you and me.”

    C. elegans have a nerve ring at the anterior portion of the body. They have the motor neuron axons running throughout the body and a bunch of cell bodies at the posterior end of the body, but they don’t have little brains scattered through-out. Most of the cell soma are located in the nerve ring.


    • Good point! I was using “little brains” to describe their ganglia, but that’s kind of a stretch. Also I was under the impression that the ganglia were spread out in the body, not organized in a ring – time to brush up on my worm anatomy!

  2. This study involving worms was really interesting as it directly relates with the learning and storage of memory. The experiment started with two groups in different petri dishes with some trained and untrained to eat food in a different petri dish with light surrounding the food (UCS). Planarian flatworms instinctively avoid light (UCR). The experimenter tracked how long worms from both groups took to eat the food with light around it in a different petri dish after decapitating the worms.

    After both sets of worms were ready, the experimenter decapitated the worms and waited for the head structure to grow back to run the experiment again. Surprisingly, the worms that were trained were still able to eat the food with light around it faster than the untrained worms. Part of this has to do with the ganglia cells that surround the worm. The biological structure of these worms is impressive, but focusing on the psychological aspects, the memory retention of the trained group even after decapitation suggests that memories are stored all around the body. Comparing this to humans, is drastically different when compared to a decapitated human.

    Referring back to the group of trained worms, they were able to remember that food around the light is “ok” to eat. Comparing this to humans remembering an old skill, it is very similar. For example, I played the cello in middle school and haven’t touched it since. But, when listening to orchestral music, it is still easy for me to pick out the distinctive cello chords with ease. This recall of memories and learning in humans and even decapitated planarian worms truly shows the power of the nervous systems in organisms.

  3. Thank you for the interesting article. I would just like to point out though that
    “What makes this even more incredible is that – unlike C. elegans that have a distributed nervous system planaria has a centralized brain in the head region, just like you and me.”
    is not quite right. Whilst it is true that there is a large clustering of neurons in the cephalic ganglia known as the anterior brain, there are also a number of neuron clusters in regularly spaced ganglia along the ventral nerve cords. What properties and roles these ganglia play with respect to “memories” will be one of the areas of research I am sure. One of the reasons for the large cephalic ganglia would be the highly developed eyes compared to C. Elegans.

    • Thanks for the correction jamlacey! I’m a rat person and definitely have much to learn about worm anatomy and physiology. Is there any evidence out there that the ventral nerve cord ganglia are involved in memory encoding and/or storage?

  4. Regain memory by Regenerating your brain cells | Pursue natural

  5. Did the head portion when regrown remember better than the other fragments? It is not clear from your writeup.

    • They didn’t remember “better” – they relearned faster than naive flatworms which haven’t been exposed to the task before. Hope this clears things up!

  6. . . . I wonder whether any study compared the learning speed of previously trained body parts — head, torso, tail etc. That would have given an indication of the location of optimum memory storage. If there is no difference between the parts that would indicate that the learning was diffused throughout the body.

    • That’s a very interesting idea! The authors only looked at tail-portion regenerated flatworms in this study. A caveat though, is that if learning was somehow imprinted on the stem cell population, the memory may still be centralized but new flatworms regenerated from different body parts wouldn’t show a difference in learning speed.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s