A Dopamine Dial for Controlling Neural Variability

The discovery of a new kind of neuron in the bird brain reveals a potential mechanism for controlling variation in zebra finch song.

A novel type of excitatory cell (magenta) target inhibitory pallidal cells (green) in Area X of the zebrafinch brain.

Acquiring a new motor skill — whether it’s learning to talk as a human or learning to sing as a zebra finch — takes practice. Researchers think that certain patterns of brain activity produce the desired action while others don’t, and the patterns that work get stabilized, explains David Perkel, a neuroscientist at the University of Washington, who studies vocal learning in birds.

But how does this variability in the brain circuit arise? In a new study published in the Proceedings of the National Academy of Sciences in May, Perkel and colleagues propose one possible mechanism that generates this variability in zebra finch Area X, a region in the bird brain that is functionally parallel to the basal ganglia circuit in mammals. Based on a combination of experimental work and modeling, they propose that high dopamine levels synchronize different components of the Area X circuit, turning down its variability. But when dopamine levels are low, the firing of these neurons remains unsynchronized, allowing variability to emerge.

Zebra finches learn a stereotyped mating song over a couple months in adolescence, which they later perform for females or ‘practice’ with no specific audience. The latter can vary, but the former much less so. To explore how birds produce both versions, Perkel’s team set out to record from a relatively rare class of neurons in Area X called pallidal neurons.

When targeting the pallidal cells, Perkel and graduate student Agata Budzillo detected a novel type of input onto them, which they propose comes from a new kind of interneuron. While pallidal cells are inhibitory, quieting their downstream targets, the new cells are excitatory — their activity makes their pallidal cell targets more likely to fire. Blocking the activity of these cells made pallidal neurons fire more regularly, suggesting these cells regulate pallidal cell variability. The researchers also found that activating dopamine receptors in these Area X cells increases how often the two types of cells fire in synchrony.

Dopamine is well-known to be involved in regulating zebra finch song — when the animals sing courtship songs to potential mates, the birds’ brains produce more of it. Perkel’s previous work had shown that blocking dopamine in Area X prevented males from switching their practice songs to courtship songs when females fly into the picture.

To understand how dopamine might influence the circuit’s variability, Perkel worked with Adrienne Fairhall, a computational neuroscientist also at the University of Washington. “My student Alison Duffy had a really interesting idea, that the way a neuron is driving a downstream population might control the degree of synchrony of the output,” says Fairhall, an investigator with the Simons Collaboration on the Global Brain.

Fairhall’s team created a simple model that showed how dopamine had the potential to change the coupling between the circuit’s excitatory and inhibitory inputs. A bump in dopamine increases such coupling, effectively synchronizing the firing of the inhibitory cells. This, in turn, controls the variability of the song circuit beyond Area X. “Dopamine is kind of a dial for controlling downstream variability,” Fairhall says. “It’s really the modeling that gave that insight.”