The energy cost of the brain

In order for an animal to survive, its brain must interpret incoming sensory information to determine the animal’s next course of action, such as hunting a prey or escaping from a predator. Whilst animals with simpler behaviours are able to get by with a smaller number of neurons, for example the humble sea slug has only approximately 10,000 neurons (Moroz, 2011), whereas animals that utilise more intricate behaviours usually have orders of magnitude more neurons, such as ravens (~2.1 million), humans (~8.6 billion), and elephants (~25 billion). However, the advantages of having more neurons come at a price: neural tissue is very expensive in terms of metabolic energy (Laughlin et al. 1998), with the human brain accounting for 20% of a person’s energy consumption, despite representing only 2% of its body weight (Hofman, 1983). This high metabolic demand means there is a selective evolutionary pressure to find efficient mechanisms which reduce the amount of metabolic energy used, without sacrificing the behavioural capabilities necessary for the animal to survive.

However, as Prof. Laughlin explained, it is not simply the number of neurons which dictates the amount of energy a brain will consume. The amount of information a neuron conveys also determines its energetic cost. But what exactly does information mean in this context? Information can be thought of as a combination of both the reliability and rate at which a neuron conveys its message. For example, to track small, fast moving prey, a predator needs neurons with the capability of transmitting high information content that can reliably communicate where the prey is as it tries to escape. Whilst it may seem intuitive that a brain would want to maximise the amount of information each neuron conveys, neurons that are capable of conveying more information require much more energy; increasing the information capacity of a neuron 5 times requires 25 times more energy, placing a large energetic cost on information (Niven et al., 2007). This is because to increase the information capacity of a neuron requires an increase in the expression of receptor proteins and ion channels. Whilst an increase in receptors or channels acts to increase the reliability of the neuron, with every additional activated receptor or opened ion channel comes a price of increased energy consumption.

What strategies can be employed to make the brain more energy efficient? One such method the brain seems to utilise is optimising the placement of its neurons. Neurons connect with one another via wire-like processes, known as axons and dendrites. With increasing lengths of wire, there is an increasing energetic cost, of both maintaining these connections and communicating information over longer distances. Therefore, an efficient brain should aim to minimise the overall length of its wires, a process known as wiring economy. Indeed, a study which modelled the efficiency of the placement of neurons within a neural circuit of the Drosophila brain, demonstrated that the configuration of the wiring was very close to the optimal configuration to minimize the total length of wire (Rivera-Alba et al., 2011). It turns out that models of wiring economy can be used to explain the structure of a number of micro-circuits in the brain, demonstrating that the brain has developed efficient strategies to reduce metabolic cost. Further strategies Prof. Laughlin has unearthed include sending the lowest amount of information necessary, send it as slowly as possible and spread the information cost over many neurons, each of which should transmit information at a low rate (Sterling and Laughlin, 2005).

The work presented by Prof. Laughlin is remarkable in that it goes beyond simply describing how the brain works, but delves into the realms of trying to understand why brains are designed the way they are. These basic principles will hopefully act as a guiding light for future neuroscience researchers to elucidate the theoretical principles underlying neurobiology.

References

Hofman M.A (1983) Energy metabolism, brain size and longevity in mammals. The Quarterly Review of Biology. 58(4):495-512

Laughlin S, Anderson J.C, O’Carroll, D.C and De Ruyter van Steveninck, R.R (2000) Coding efficiency and the metabolic cost of sensory and neural information. In Information Theory and the Brain. Cambridge University Press.

Rivera-Alba M, Vitaladevuni S.N, Mischenko Y, Zhiyuan L, Takemura S, Scheffer L, Meinertzhagen I.A, Chklovskii D.B, de Polavieja G.G (2011) Wiring economy and volume exclusion determine neuronal placement in the Drosophila brain. Current Biology 21(23): 2000-2005

Moroz L (2011) Aplysia. Current Biology 21(2):R60-R61

Niven J.E, Anderson J.C, Laughlin S.B (2007) Fly photoreceptors demonstrate energy-information trade-offs in neural coding. PLoS Biol 5(4):e116

Niven J.E and Laughlin S.B (2008) Energy limitation as a selective pressure on the evolution of sensory systems. Journal of Experimental Biology 211(11):1792-1804

Sterling P and Laughlin S.B (2015) Principles of Neural Design (The MIT Press)

Scientifica supports the NEUReka! Seminar Series to help them book speakers from top research institutions around the world.

Sign up to receive our latest news

Find out about Scientifica's latest product releases, company news, and developments through a range of news articles, customer interviews and product demonstration videos.