MadSci Network: General Biology

Re: What are the energy requirements of thought?

Date: Thu Sep 22 05:51:30 2005
Posted By: Alex Goddard, Grad student, Neuroscience, Harvard Medical School
Area of science: General Biology
ID: 1126722155.Gb

Hi Teddy,

     That’s a good question! I think the short answer is that it’s less energy than you might expect. I’ve tried to explain why I think so below.

     First, I’d say that the basic energy expenditure per unit area of brain is probably the same between different types of animals. A human brain uses more total energy than a rat brain because it's bigger and more complex. This idea probably applies to heat generation, blood flow and caloric requirement as well.

     I saw somewhere that the brain uses 20% of your caloric input for the day. I don’t think that number is absolute, but it seems reasonable. So, assuming you eat 2000 calories per day, your brain uses about 400 calories per day. Then, we need to ask, how much is spent on ‘thinking’ or cognitive abilities? (I’m leaving the idea of consciousness out of this, because it has no definition that has been agreed upon)

     To start out, we need to realize that the energy expenditure of the brain at rest is really quite high. This is because maintaining a neuron's baseline is very metabolically "expensive;” before neurons can transmit information, they have to be properly connected and maintained. A neuron's ability to use electricity to transfer information comes at the cost of a large amount of energy; it needs to constantly pump different ions into and out of the cell (ions carry the electric charge). Maintaining the long nerves that may extend a meter (such as the ones from your brain to your spinal cord) is also metabolically taxing as new proteins have to be made constantly and shipped down the nerve. All these processes require energy, and we haven’t even sent any information down axons yet! I would contend that the amount of energy spent maintaining the cell is far more than is actually spent on sending an electrical impulse down an axon.

     Ok, so cell maintenance requires a lot of energy. One could then ask the question, "What about the cognitive functions? How much energy do they require?" I think the best way to answer this would be to first ask, 'How much of the brain is devoted to cognitive function?" If we could say that 50% of the brain was devoted to cognitive function, then we could say 50% of the energy is devoted towards maintaining cognitive function.

     The answer to that question is a hard one to pin down. Most of the brain is involved in non-cognitive processing. The brain must recreate visual space from a bunch of pixels in your eye. It computes the location of a sound based on when the sound hits your right ear versus your left ear (within 2 degrees, I think). It maintains your balance, triggers fight-or-flight responses, and decides how your fingers must move to type on a keyboard. Lots of non-cognitive processes are going on.

     Furthermore, cognitive abilities are distributed all over the brain. (For those not familiar with brain anatomy, take a peek here for reference for the following descriptions: )

     For instance, a brain scan (functional MRI) of people doing arithmetic showed activity in frontal lobe, temporal lobe, and occipital lobe (although activation of the occipital lobe, the site of vision processing, may have had to do with the subjects seeing the math problem). Language processing and generation is located mainly in the temporal lobe and frontal lobe. Decision-making and determining the consequences of actions is thought to occur in the frontal lobe. Generally, the frontal lobe is thought to be a hot seat of ‘cognitive ability,’ but it is not a pure ‘cognitive center.’ It is thought to be involved in non-cognitive decision making (i.e. determining unconscious preference and value). So to ask how much of the human brain is solely dedicated to cognitive tasks, I’d have to hazard a very hand-waving guess of 5%. And that’s a pretty liberal guess. And most of the energy used is spent on maintaining the cells and connections in these areas, not the actual cognitive processing itself…

     How does that relate to other animals, primate and non-primate? Primates have a very similar level of cognitive power. They do appear to have a simple form of language, and may even be able to read! ( ). They can make complicated decisions. Non-primate animals definitely have cognitive abilities as well, such as decision making, risk assessment, and memory formation, though the degree to which these decisions are 'conscious' as in humans is quite debatable. I don’t know if I could put a number on how much more brain space and energy is spent on cognition by humans, because it could be the type and amount of connectivity between cells that is more important than the percentage of cells in the brain.

     Lastly, I’ll just briefly mention one interesting tidbit of info. The process of being aware of your surroundings apparently requires a lot of energy, which is not surprising. The basic idea is that as you transmit information, it uses some energy. According to one report, "The high resting brain activity is proposed to include the global interactions constituting the subjective aspects of consciousness. Anesthesia by lowering the total firing rates correlates with the loss of consciousness." That is to say, low conciousness = lower firing rates = less total energy consumed. fMRI really uses a measure of metabolism to determine firing rates…

     I hope that provides some insights – I don’t know if we know enough to calculate an absolute number. If you have more questions, please do submit them!

-Alex G

     Reference to math computation in the brain studied by fMRI:

Current Queue | Current Queue for General Biology | General Biology archives

Try the links in the MadSci Library for more information on General Biology.

MadSci Home | Information | Search | Random Knowledge Generator | MadSci Archives | Mad Library | MAD Labs | MAD FAQs | Ask a ? | Join Us! | Help Support MadSci

MadSci Network,
© 1995-2005. All rights reserved.