What brain resources are needed per limb?

Question

I have previously asked this question about the plausibility of large six-legged creatures, but I decided to split it into several more specific questions because of the advice here.

I want to first ask about the issue of how much brain (and nervous system power) it takes to control various limbs. I have rarely seen much discussion about this issue at all, except in this article, where the author notes that there are some who claim that large creatures cannot have more than four limbs because it would be too much of a challenge for a brain to handle. He quickly dismisses these objections, but I do not wish to rely on one source only, and I cannot access the reference he uses because a subscription is necessary.

The big problem is that I have tried to look in every resource I have found, and I have found barely any information about how many neurons and neuron interconnections, or how much of the brain, it takes to control a single limb. There do not seem to be any numbers or even much qualitative information I can see. Therefore, what I want to know is, for each individual limb in an average animal, how many neurons and what percentage of the brain does it take to control that limb?

If it scales by size, or there is a "law of diminishing returns" in which the more limbs are added the more it takes to control each, such information would be particularly useful to find out. Also useful would be data for creatures such as humans (because they're one of the few creatures with fine motor control of arms and hands, and I've heard that arms need more of the brain than legs do), elephants (they have a quite dextrous trunk with far more muscles in it than found in the whole human body), and insects (they're small, but they do have six legs). But really, considering the utter lack of information about this topic, any numbers on this subject would do.

---

Clarification

Note that I have tagged this question hard-science. I believe that in this case, the tag is appropriate as I am currently asking for information and numerical data about an aspect of real-life creatures.

Answer 1

For each individual limb in an average animal, how many neurons and what percentage of the brain does it take to control that limb?

I think there's a false assumption in this question similar to "humans only use 10% of their brain". This has a view of the brain like some a computer with a generic central processing unit that can do a fixed amount of processing and fixed amount of memory. The brain doesn't work like that.

Note: given that the question is based on a false assumption, I don't think it's possible to meet the normal standards of the hard-science tag. I'll do my best.

Instead, the brain is made up of very specialized areas which control certain things. Saying "you only use X% of your brain" is like saying "you only use X% of your house". Having someone watching TV in the living room doesn't help you cook dinner in the kitchen.

It's not the size of the brain, or some sort of total brain power, that matters. But whether or not the portion of the brain for limb control is developed to handle coordinating six limbs. To use the house analogy again, it doesn't matter how big it is, if there's only one toilet then only one person can use it at a time.

So the answer is: if you need to control six limbs, evolution will have made sure the portion of the brain for controlling six limbs is well developed. This might be to the detriment of other parts of the brain... or you might just have a more densely packed brain.

This is why brain size doesn't matter as much as brain density and specialization. Brains consume a lot of energy. A human brain is just 2% of your body weight, but consumes 20% of your energy. A more complex brain means more energy. Human evolution has made that work for a little while, but its usually not a good evolutionary trade-off. For example, a Koala has one of the lowest brain-to-body-weight ratio of any mammal, and it's very smooth reducing its surface area further reducing the density of neurons. This is, in part, because its diet of eucalyptus leaves is very low in calories and nutrition. A Koala doesn't need a lot of brain power, so it doesn't waste the energy on it.

In short, there is no answer to "percentage of the brain per limb" because that isn't how the brain works. Instead, the areas for controlling limbs would be... not necessarily more, but differently developed: the primary motor cortex, the premotor cortex, the supplementary motor area, and others.

---

The real reason humans don't have six limbs is because we've descended from a body plan set in place at least 500 million years ago shared by all vertebrates. A spinal column, head, and four limbs. Even snakes, with apparently no limbs, have four vestigial limbs. This basic body plan does not change, evolution has to build incrementally with what it's got. It can adapt, and remove, but rarely does it add, or make radical changes.

All this adaptation is also why it's difficult to answer "how many neurons per limb". Your brain, like your body, is adaptation piled on adaptation piled on adaptation. And your brain, like your body, is very efficient; it does with about 20 Watts what a traditional computer, with its very organized structure, would need 10 Megawatts. But that efficiency means its very, very, very complex, and the same neuron might be involved in many functions. Its doubtful you can look at a neuron and say "this is only for controlling a limb" because its probably involved in doing something else as well. We don't fully understand how the brain coordinates movement, or does most things.

Answer 2

A surprising amount of calculations are not done in the brain at all, but in the spinal column! A large amount of our movement capabilities are actually managed within the neural network of our spine.

As an astonishing example, consider the central pattern generators found in the spine for controlling our gait. If you are walking, and your right hand brushes up against something, you will actually adjust the movement of the left leg to compensate before the signal has even reached the brain.

A particular study in 2007 used decerebate cats. These are cats whose cerebellum was removed in the name of science. If that idea bothers you, this may be a good point to stop reading and just accept that the spinal column is responsible for a remarkable portion of the control of our limbs.

The particular details are not for the squeamish, but fortunately are well obscured by the precise scientific jargon so can be reproduced here to specify just how little of the brain remained after the surgery:

Next, we performed a pre-mammillary decerebration. The brainstem was transected rostral to the superior colliculus, at approximately a 45° angle, in order to preserve the mammillary bodies and sub-thalamic nucleus. All brain matter rostral and lateral to the transection was removed.

These cats were then put on a treadmill, which caused the cats to walk, despite having no higher functions. They then adjusted the angle of the neck of the cat to simulate walking up and down while keeping the head level.

They found that, when walking flat, the EMG readings showing muscle activity matched the patterns associated with a normal cat walking with extraordinary precision. The brain was, in fact, not required at all for this motion. They then tilted the neck up and down and found that doing so lead to remarkably similar results to normal cats walking up and down hill while keeping their head level. The propreceptors in the neck were actually being integrated into the data processed by the spinal column and the gait was adjusted accordingly.

So I'd say more limbs is totally valid, because a surprisingly large amount of what we do with them is actually a distribute capacity found in the spinal column, not the brain. Juggling 7 balls with 6 arms may still be a difficult task, but merely operating those arms would not be.

Answer 3

There isn't really an "average" animal, but I suspect the median animal with limbs is krill, which has a dozen or so limbs and a tiny nervous system.

One of the largest animals in the world is the giant squid, which has ten limbs which can move in much more complex ways than jointed skeletal limbs, so putting a limit on size of multi-limbed creatures has obvious counterexamples.

Squid brains are rather small compared with the size of them. Instead, cephalopod limbs are controlled by a more distributed network of ganglia. If there is an advantage for a large animal to have many limbs, then evolution will find a way to achieve it.

Answer 4

I can only answer about humans. You ask what is needed to control a limb... I assume you are only talking about conscious motor functions.

1. "Conscious" means that we must exclude everything related to the peripheral nerves that inervate the limbs, as well as reflexes (these are located in the spinal cord)
2. "Motor" means that we must exclude everything related to the "sensory" neurological processes

These exclusions are artificial, since the limb movements inherently integrate these neurological processes. For example, without the peripheral nerves, there would be no way to transmit the instructions from the brain into the muscles. Also, without sensory processes, movement would be more difficult... so, for instance, proprioception allows the brain to know where each part of the limb is located in space, and therefore, how the limb can move in order to achieve the desired effect.

However, in order to simplify my answer, and since this seems to be what you're asking, I'll restrict myself to said conscious motor functions.

---

Schwern is correct in answering that the brain simply doesn't function like that. It is not so much of a matter of percentage of brain, or of number of neurons... what is important is the number of neurological connections (synapses) involved in a function, and the complexity of those connections.

But, even so, I'll try to answer the question "What percentage of the brain does it take to control that limb?

---

Our brain is covered by a superficial layer, that is called the cerebral cortex. It is in this layer that our neurological connections are the most complex... and therefore, it is in this layer that the most superior brain functions are located.

So where in the cerebral cortex are the motor functions located?

If you watch a brain from above, you'll see a fissure that divides the brain in half - a left half and a right half. Those are the hemispheres. The left hemisphere of the brain controls the motor functions in the right half of the body and the right hemisphere controls the motor functions in the left half of the body.

If you watch the same brain from the sideways, you'll see another groove that divides the brain in two parts - an anterior part and a posterior part. This groove is called the central sulcus. The motor functions are located on the front part, the frontal lobe, right adjacent to that central sulcus... this is the motor cortex.

If you take this part of the brain and slice it on a right-left direction, you'll be able to see the cortex enveloping the brain. Now, if you superimpose on each part of the motor cortex the body parts that it controls, you'll be able to draw a "cortical homunculus".

As you can plainly see, the motor cortex that controls the hand is much greater than the motor cortex that controls the rest of the arm. Also, the motor cortex that controls the leg is much smaller than the motor cortex that controls the face. This is because the hand has much more nuanced and fine movements than the arm... and all our facial expressions need a much finer motor control than the simple movements needed to move a leg (which consist almost exclusively in moving it forward or backward). So, the amount of neural connections needed to control all the hand and face movements needs more brain space than the leg.

So, it is not a matter of how much limbs your being has... but of how fine the movements needed for said limbs are.

---

Please note that this is an oversimplified answer... To be more precise, I would need to detail a lot of other brain centers, namely responsible for motor cohordination. The motor cortex that I detailled above is only responsible for moving the limbs, not to cohordinate those movements with the other parts of the body. So, if you electrically stimulate the "leg part" of the motor cortex, your leg will jerk accordingly, but that movement may be extremely imprecise for any objective (vg: kicking a football).

PS: Here's another image of a cortical homunculus, ie, a representation of the body in which each bodypart has a size directly proportional to the respective amount of motor cortex.

Answer 5

Can't give a definitive answer, but I'd like to point out some experiments with brain-computer interfaces in monkeys and even humans show that the brain can learn to control an additional limb (real, virtual or even just a cursor on a screen).

Human trials (Mainly the "Braingate" chip) have been limited to people unable to move their real limbs to begin with, so it is hard to claim that the artificial limb counts as "additional" instead of just a replacement.

But experiments in monkeys (Mainly the work of Miguel Nicolelis at Duke University) have showed control of an additional limb independently of the monkey's real limbs. Even though the setup starts with the monkey controlling a joystick and the BCI being programmed to react to patterns from that activity, eventually the monkey is able to control the BCI without moving the joystick or its real arms.

https://www.sciencedaily.com/releases/2005/05/050511073108.htm

My point is that while I can not quantify the resources needed for motor control, I'm convinced that what we already have is easily sufficient for controlling multiple additional limbs. The amount of dexterity and coordination probably depends mostly on practice and not so much on brain capacity. The plasticity of the brain is quite incredible, it can adapt to very unexpected conditions.