In comparison to mammals, neuron densities in the avian brain are high, opening up the question of how birds can metabolically support their large neuron numbers. Von Eugen et al 2022 show below that the neuronal energy budget of pigeons is about 3× lower compared to mammals, possibly indicating a more efficient neuronal processing in the avian clade.

Brains are among the most energetically costly tissues in the mammalian body.¹ This is predominantly caused by expensive neurons with high glucose demands.² Across mammals, the neuronal energy budget appears to be fixed, possibly posing an evolutionary constraint on brain growth.^{3, 4, 5, 6} Compared to similarly sized mammals, birds have higher numbers of neurons, and this advantage conceivably contributes to their cognitive prowess.⁷

We set out to determine the neuronal energy budget of birds to elucidate how they can metabolically support such high numbers of neurons. We estimated glucose metabolism using positron emission tomography (PET) and 2-[¹⁸F]fluoro-2-deoxyglucose ([¹⁸F]FDG) as the radiotracer in awake and anesthetized pigeons. Combined with kinetic modeling, this is the gold standard to quantify cerebral metabolic rate of glucose consumption (CMR_glc).⁸

We found that neural tissue in the pigeon consumes 27.29 ± 1.57 μmol glucose per 100g per min in an awake state, which translates into a surprisingly low neuronal energy budget of 1.86 × 10⁻⁹ ± 0.2 × 10⁻⁹ μmol glucose per neuron per minute. This is ~3× lower than the rate in the average mammalian neuron.³

The remarkably low neuronal energy budget explains how pigeons, and possibly other avian species, can support such high numbers of neurons [almost twice as many neurons as a similarly sized mammal] without associated metabolic costs or compromising neuronal signaling. The advantage in neuronal processing of information at a higher efficiency possibly emerged during the distinct evolution of the avian brain.

…Potential factors explaining the reduced neuronal energy budget: There is ample evidence that any increase in size of parts of or the whole nervous system evolved under a selection pressure to reduce energy consumption.^6,38 This can be achieved by either reducing the costs of signaling itself, with alterations in the biophysical properties of cells and circuits, or with alternative coding strategies.³⁹ Understanding how energy efficiency is obtained within the brain starts with collecting detailed neuroanatomical and neurophysiological data such as neuron size, firing rates, ion channel kinetics, membrane capacitance, etc. Based on these data, computational models can test theoretical predictions of optimal energy efficiency under specific conditions.³⁹

Currently, most of the crucial data are lacking for birds. This highly complicates identifying any underlying mechanism for the low neuronal energy costs. Based on what is known, we can identify two potential contributing biophysical properties that differ in birds compared with mammals: namely, neuron size and brain temperature.

Cell size is a key factor in the metabolism of any neuron and correlates positively with energy consumption.^30,40 Though no systematic survey has ever been conducted in birds, there is support for the idea that the average avian neuron size is smaller than in mammals. As mentioned above, neuron densities in the avian brain are much higher compared to similarly sized mammals, and this likely translates into small neuronal sizes and short inter-neural distances.^7,41 Indeed, a mathematical approach demonstrated that neuron and non-neuronal cell density could be used as an indicator of cell size, where higher density corresponds to smaller neuron sizes.⁴² Moreover, from comparison between the macaque and mouse, it is known that some specific neuron types indeed scale positively with increasing brain size.⁴³

A smaller neuron is more energy efficient in a variety of ways.^44,45 For example, in line with the reduced membrane surface area and cytoplasmic volume, smaller neurons accommodate fewer receptors, ion channels, and mitochondria. They are also characterized by a lower membrane capacitance. Lastly, the overall lower number of components translates into reduced maintenance and housekeeping costs. Thus, smaller neuron sizes could explain, at least in part, the lower metabolic costs we observe in the pigeon brain.

Next to neuron size, another stark difference between birds and mammals is the higher body, and thus brain, temperature in birds.⁴⁶ In pigeons, the core brain temperature measures 40℃–42℃⁴⁷ compared with 36ºC–37ºC in the rat brain.⁴⁸ The effect of temperature on behavior and neural activity is a widespread phenomenon, and it influences multiple cellular components and dynamics, including the resting membrane potential, generation of action potentials, synaptic transmission, and axon conduction velocity.⁴⁹ For example, increases in temperature of 1.5° have been recorded in rat hippocampus during active exploration, in concordance with changes in the waveform of action potentials.⁵⁰ These interactions have also been found in birds. In zebra finches, the cooling of specific song-related brain nuclei had a causal decreasing effect on song tempo.⁵¹ This is not just an effect of experimental manipulation, because zebra finches show natural temperature fluctuations related to a diurnal cycle and social context; in correlation to either a decrease or increase in brain temperature, the song tempo decreased or increased, respectively.⁵²

At the base of many of these processes is the direct effect of temperature on ion channel kinetics, maximum conductance, and gating kinetics.⁵³ The most prominent example of this effect is its influence on the time of overlap between Na⁺ and K⁺ currents that flow across the membrane during an action potential. Time of overlap and costs are positively correlated, since higher numbers of ions need to be pumped across the membrane by the Na⁺/K⁺-ATPase after the action potential. Higher temperatures were found to exponentially and strongly decrease the time constants of Na⁺ and K⁺ channel activation and inactivation, thereby lowering the energy costs.⁵⁴ This type of modulation of ion channel kinetics not only reduces costs of individual neurons but also seems to improve signal transmission.

This was demonstrated in the swordfish (Xiphias gladius), where regional warming of the eye and brain improved temporal resolution of visual processing by 10×.⁵⁵ Thus, the higher brain temperature present in birds could contribute to reducing energy consumption of neurons by both making ion channel kinetics more efficient and improving information rates.

…Evolutionary perspective: It is important to keep in mind that what is “economical” compared to a mammal can still be costly for a bird. Indeed, across different bird species, we observe many examples indicating that neurons are not a free-for-all commodity. This is clearly apparent in songbirds, which demonstrate some of the most impressive types of neuroplasticity among adult vertebrates. The most extreme example comes from the spotted towhee (Pipilo maculatus), whose HVC (a song nucleus important for learning of song) in males can increase 300% in size during a breeding season to facilitate complex song repertoires to attract mates and defend territory.⁵⁶ Of course, it is unlikely that any organism wastes neuron numbers or other costly tissue.⁵⁷ These extreme examples of neuroplasticity do show that avian species are under particularly high pressure to reduce (cerebral) energy consumption. In addition, the higher body temperature and costly capacity of flight might also be crucial contributing factors.^46,58

As mentioned, despite this extreme pressure, birds attain higher neuron numbers compared to similarly sized mammalian species;⁷ like mammals,⁵⁹ these numbers positively correlate with cognitive performance.⁶⁰ Our findings from pigeons suggest that the ceiling on neuron numbers might be raised by attaining a 3× lower neuronal energy budget. Future studies will have to verify whether this is a class-wide phenomenon, but the high neuron densities observed across almost all avian species studied so far can be taken as a first indicator that this is the case.⁷ Importantly, the lower budget does not seem to compromise neuronal computations, since birds are considered perceptually and cognitively on par with mammals.¹⁰ The last common ancestor of birds and mammals existed ~312 mya,⁶¹ and in the long parallel evolution of both lineages, birds ended up with tiny brains comprising high numbers of small neurons organized in a distinct cerebral layout^7,10 and situated in a warmer physiology.⁴⁶ The combined effect of these distinct elements on neuronal dynamics generated a possible advantage in neuronal processing of information at a higher efficiency: cheap neurons with advanced processing capacity.

See Also: