Elections in the Brain
How decision-making by nerve cells compares to democracies and dictatorships
Posted February 26, 2016
If you live in the U.S. and have not been barricaded in a cave for the past several months, you have been inundated with political advertisements and talk of polling, voting, caucusing, and how the world will end if candidate X is elected. Perhaps you wonder if our chaotic (though entertaining) approach to collective decision-making is the best way to go about this. Perhaps you wonder if money and advertising distort the process.
You may also wonder what goes on in your own brain when you make decisions. In some ways, this process is analogous to U.S. primaries and elections.
The voters in our brains are nerve cells or neurons. Each neuron votes by either firing off an action potential (also called a spike) at a given moment or remaining silent. A spike is a brief electrical signal that is conveyed to a group of neurons on its individual recipient list. Each neuron at each moment adds together votes from its own input neurons to decide whether it should spike or not. Ultimately, behavior (movement) is implemented by the spikes of a subset of neurons, the motor neurons, which are connected to muscles and cause them to contract.
In many cases, a large number of brain neurons collectively determine your next move by voting. For example, as you are reading this, your eyes dart from one spot to another on the screen. These eye movements are called saccades.
Saccades are triggered by neurons in the midbrain that are arranged in a map of movement space, where neurons in each part of the map trigger saccades with a particular angle and distance. If you temporarily prevent the neurons in one subregion from voting (voter suppression?) by anesthetizing them, then the saccade predictably shifts in angle and/or distance, at least in monkeys. So in the absence of voter suppression, the votes of the whole set of neurons in the midbrain collectively determine where your eyes move, through something like a pure democracy.
Not all neurons vote equally, however. Some are more equal than others. If a neuron generates more spikes per second, it votes more often. (I am from Chicago, where there is a history of this kind of thing….) Fast-firing neurons influence the behavioral outcome more than neurons in the relatively silent majority. This also has a counterpart in U.S. politics: those who can speak more loudly (or pay for more advertising) are more likely to influence message recipients. This may also be important in the type of caucus (such as Democrats hold in Iowa) in which members of candidate groups that are too small to count are persuaded to switch to another group.
In the midbrain map of saccades, if you activate two different subsets of neurons, you could actually trigger a saccade that is in between the two stimulated saccades in angle and distance. Call it a compromise. (You remember what that is, right?) For some behaviors, however, we (or at least monkeys) use a different kind of election, called winner-take-all, in which competing sets of neurons duke it out to determine the behavioral outcome. This happens when we try to judge the direction of movement of slightly moving objects we see. If you stimulate two areas within the part of the brain that controls this visual perception, you typically trigger one movement perception or the other, not a compromise between them. This process is similar to how most U.S. states award delegates in primary elections.
Not all decision-making in the brain is democratic. In some cases, a single neuron calls the shots, like a dictator. Neurons like this are called command neurons. There may not be any command neurons in the human brain (though it is almost impossible to know for sure), but they certainly exist in many invertebrates. There is at least one example in vertebrates.
Fish and amphibians have one giant neuron on each side called the Mauthner cell (the pair of large red cells in the photo). A single spike in a Mauthner cell triggers an entire escape movement, in which the animal rapidly turns away from danger and swims. When this neuron speaks, the whole body responds. Because it is so large (particularly its axon, which is like a long wire going from the brain down to the spinal cord), it can send its command very quickly, which is important if you want to avoid being eaten.
But if the Mauthner neuron is killed, a small group of other (large, but not giant) neurons (the smaller red neurons in the photo) immediately take over, a kind of oligarchy that is waiting in the wings in case the dictator is assassinated. Command neurons and neuronal oligarchies may also act together to control movements. It is probably the rule, not the exception, that multiple forms of decision-making occur simultaneously to control many of our behaviors.