The human brain is an incredible energy drain. Taking up only about 2 percent of the body's mass, the organ uses more than a fifth of bodily energy. Ever more accurate calculations of its energy budget at the level of the neuron (nerve cell) are important to researchers ranging from functional magnetic resonance imaging (fMRI) analysts to evolutionary biologists.

Fifty-seven years ago, Nobel laureates Alan Hodgkin and Andrew Huxley came up with a model to calculate the power behind electrochemical currents in neurons—a great step forward in understanding how the brain worked and how it divvied up resources. The only problem was that their subject was not a person, or even a rodent, but a squid.* Today, researchers announced that they have found a more accurate model for mammal brains, which elevates some of their transactions to three times more efficient than that of the squid-based equations.

To carry information, chemical signals fire across trillions of brain-cell synapses (connections between neurons). A 50–50 energy use split between action potential propagation and synaptic transmission has been proposed in some energy budgets of the brain (which did assume that the part contributed by action potential propagation is four times the theoretical minimum, as suggested by the Hodgkin-Huxley mode—but only if you know what the contribution of synaptic transmission is you can conclude about a 50–50 split, for example; Hodgkin and Huxley's model does not tell us what the energy cost of synaptic transmission is), Roth notes.**

The new work, reported online today in Science, proposes that in mammalian brains, the split is actually closer to 15 percent for spikes (action potentials) and 85 percent for the subsequent synapse change. Although the total energy usage appears to be the same, the division may have important implications in the understanding of how modern brains developed.

How are mammals able to save so much energy on the front end? Hodgkin and Huxley's squid brain fired more ions that overlapped, requiring more energy to reestablish an electrochemical differential. Mammals' brains, however, appear to space out the movement of the ions, lowering the amount of charge differential restorations needed after the action. "The less charges you move across, the less energy you have to use to reestablish the gradient," says Pierre Magistretti, director of the School of Life Sciences—Brain Mind Institute at the Swiss Federal Institute of Technology in Lausanne, who wrote an accompanying perspectives article in Science.

Scientists who study the brain often depend on energy budgets to estimate allocations of where energy is input and consumed. "It's hard to come up with the numbers for these budgets," says Arnd Roth, a study author and senior researcher at University College London's Wolfson Institute for Biomedical Research. "It's not their fault," he says of researchers who have been working under now-disproved assumptions.

The findings may put to rest the "long-lasting debate about which one of these processes costs the most energy," Magistretti says. There might, however, still be some doubters: The new study was performed on rats, but Roth is reasonably confident that the results would be similar for humans. "We have now another model" to replace the half-a-century-old Hodgkin–Huxley equations, he says, adding, in rats' axons, "shape is similar; the same molecules are involved, so that should suggest that they should behave similarly" in people. Roth and his team's tests were limited to axons in the hippocampus, and he notes that axons elsewhere in the rats should be tested to ensure that energy usage is standard across the brain.

Just how the new ratio will affect analysis of brain images also remains debatable. As Magistretti notes, "All the functional brain imaging techniques that are used to explore brain activity, fMRI or PET (positron emission tomography), measure the use of energy in areas that are more active than others." But those types of imaging measure this sort of energy use only indirectly, so, Roth says, "It's up to the fMRI people to take this on board and decide what it means to the signals they are seeing."

That there is such a vast difference in efficiency between giant squid and small rats raises some big questions for Roth and others interested in how the brain evolved. "It's a curious fact that in the squid [energy use] appears to not be optimal," he says. The difference may be, he speculates, that the squid depends more on its neurons to communicate rapid actions for escape, thereby sacrificing energy efficiency for speed. A bias toward efficiency might, in turn, have allowed mammals to evolve to be brainier—if possibly slower to respond.

Such a proposition raises other questions: What is the optimal efficiency rate? What are some of the other trade-offs, like reliability?

Ultimately, it seems odd to Roth that such a large error should continue to persist for years after testing was possible. But the efficiency figure was so well accepted that it has just "been in the back of people's heads—not at the front," he says. No one had taken the time to investigate: "Does it have to be like that? Is it a law of nature?"

*Correction (9/14/09: This sentence originally referred to the species studied by Hodgkin and Huxley as the giant squid.

**Note (9/14/09): This sentence has been modified to explain more clearly the process of synaptic transmission.