Early this month, a team of researchers from the Allen Brain Institute in Seattle published the first mouse connectome, a map detailing how neurons in a mouse brain connect with each other. This essentially serves as a wiring diagram of neural circuits, which some scientists believe could be used to reveal how the brain learns, stores memories, generates actions and performs all the other vital functions it’s responsible for. The connectome has even been thought of as a possible path to immortality.
With huge amounts of dollars invested connectome research, and numerous news stories, TV documentaries and books on the subject, we thought we’d ask Jason MacLean, PhD, assistant professor of neurobiology and expert on neural circuits: What’s the big deal with the connectome?
Jason MacLean: Neuroscientists often rely on computational or statistical models to study neuronal networks, and the connectome provides a lot of ground truth—or hard neuro-anatomical data—which we can use to check our hypotheses against.
What it doesn’t do, unfortunately, is really tell you how the brain works.
Brains are dynamic. They change all the time, as far as which neurons are active and when. A neuron may have 10,000 inputs—it’s receiving information from 10,000 other neurons—but at any given moment in time, only a very small subset of those inputs are active. And all the different combinations and orders of those inputs is what determines if a neuron fires. Having an image of it doesn’t tell you when, in what order and in what combination those inputs are active.
Here’s an analogy. In Times Square, you have the famous scrolling text on the news tickers. If all the lights were lit at the same time, you wouldn’t see any patterns there and it would convey no information. It only becomes meaningful when a subset of lights are active at any given time. The connectome is like a wiring diagram that tells you where all the lights are and how they’re connected. But it wouldn’t tell you which ones are lit relative to other ones to produce a letter, to produce meaning.
That said, there’s a lot of fundamental truths the connectome can tell us about the brain, particularly about neuro-anatomy. For instance, what my lab does is look at the dynamics of local circuits. That’s the spatial-temporal pattern of when neurons are active relative to one another. We are in the process of building a set of inference tools to identify the most likely combination of connections that give rise to a neuron firing. But in the end, when we generate these inference tools, we still need something to confirm that our guesses were correct, and the connectome can absolutely do that. I see both approaches as complimentary.
So in a sense there’s nothing wrong with the connectome. It’s just the promise of it has been potentially oversold. In the end it’s just a wiring diagram, and the wires aren’t all active at the same time. So we need to know which ones are active together. Once we know that, then we really can say this is how information is being routed in the brain.