Researchers at NYU Langone Health, led by associate professor of neuroscience Katherine Nagel, confirmed a long-standing hypothesis about how the brain encodes short-term memories, using the fruit fly as a model. The study was published in Nature on October 7 and describes a neural circuit that allows the fly to temporarily remember a direction and move toward an odor it wants to follow.
Unlike the human brain, which has tens of billions of neurons, the fly's brain has fewer than 200,000. Despite its relative simplicity, however, it is strikingly similar to the human brain in its organization and the way neurons communicate with one another. The fruit fly is a valuable model for neuroscientists because all the connections between its neurons—the so-called “connectome”—have been fully mapped, allowing researchers to directly study how different types of neurons interact to produce specific behaviors.
Working memory is the ability to quickly activate temporary information—for example, briefly holding the digits of a security code in mind. It would make little sense, however, to waste energy retaining unnecessary information, such as memorizing every sequence of numbers we encounter during the day. The study examined how neural interactions can be stable—meaning they persist for some time—while also being switched on or off quickly.
When the researchers exposed the flies to a strong smell of apple cider vinegar, they moved toward the odor, even for a few seconds after it disappeared. Monitoring the flies' brains during the process, the researchers found that two different types of neurons—PFG and hΔK—responded to the odor with similar patterns of electrical activity. This led them to believe that the cells work together to control the fly's movement in response to the odor.
The two types of neurons form what is known as an attractor network: a type of neural circuit in which a group of neurons “talks” to itself until a stable signal emerges. In this case, however, there is a distinctive feature. PFG and hΔK do not communicate continuously: most of the time, hΔK activity is blocked.
When communication between them is interrupted, PFG neurons track the fly's orientation in space by receiving information from its “compass” system. But when the block is lifted and the two types communicate, the fly can “lock onto” a specific starting point, such as the source of an odor, and move toward it. The researchers call this system a “split attractor network”: PFG receives the content of the memory, hΔK controls when it forms, and the block on communication acts as a gate. This combination provides the flexibility and signal stability that working memory requires.
As Nagel explains, scientists had long hypothesized that such an arrangement powers working memory, but the experiments now confirm that it exists and show how it works in a specific context. “Right now, one of the frontiers of neuroscience is understanding what specific networks do, and the fruit fly is one of the best models for studying this,” Nagel says. “The fly has a remarkable track record of revealing how human biology works in a clear and simple way. I hope it will give us insight into processes such as working memory, which we have not yet had the tools to study in depth.”
As its next steps, Nagel's laboratory wants to learn how this circuit is controlled over different timescales and to characterize the kinds of information tracked by other types of neurons. The team also aims to understand how and why different regions of the brain can control similar functions at the same time.
Funding came from the US National Institutes of Health and the National Science Foundation, while Bard Ermentrout from the University of Pittsburgh also took part in the study.
Comments