Abstract
What are the spatial and temporal scales of brainwide neuronal activity? We used swept, confocally-aligned planar excitation (SCAPE) microscopy to image all cells in a large volume of the brain of adult Drosophila with high spatiotemporal resolution while flies engaged in a variety of spontaneous behaviors. This revealed neural representations of behavior on multiple spatial and temporal scales. The activity of most neurons correlated (or anticorrelated) with running and flailing over timescales that ranged from seconds to a minute. Grooming elicited a weaker global response. Significant residual activity not directly correlated with behavior was high dimensional and reflected the activity of small clusters of spatially organized neurons that may correspond to genetically defined cell types. These clusters participate in the global dynamics, indicating that neural activity reflects a combination of local and broadly distributed components. This suggests that microcircuits with highly specified functions are provided with knowledge of the larger context in which they operate.
Introduction
Neuroscience has long grappled with two seemingly contradictory views of the brain. One perspective emphasizes specialized, independent microcircuits dedicated to specific tasks like grooming or walking. The other, supported by recent wide-field imaging, suggests that behavior triggers massive, "global" waves of activity across the entire brain. This study addresses this dichotomy by imaging a large volume of the Drosophila brain at single-neuron resolution, investigating how local microcircuits interact with global brain states during spontaneous behavior.
Methods
To bridge the gap between whole-brain dynamics and single-cell resolution, the researchers employed SCAPE 2.0 microscopy, a light-sheet technique capable of volumetric imaging at over 10 volumes per second. By using nuclear-localized GCaMP6s indicators, neurons were visualized as discrete points, enabling the tracking of 1,400–1,600 individual cells per fly. Behavioral monitoring was achieved using Deep Graph Pose (DGP) to track limb movements, which were then categorized into states such as running, grooming, and quiescence using semi-supervised mapping. A regression model with cell-specific lags was used to account for varying neural response speeds relative to behavior.
Global Dynamics and the "Flygenvector"
A primary finding of the study is that most neurons in the dorsal fly brain are either correlated or anti-correlated with vigorous movement, such as running or flailing. This brain-wide activity, termed the "global mode" or "Flygenvector," appears to be driven by the intensity (vigor) of the behavior rather than the specific motor modality. For instance, while running involves all limbs and triggers a massive global wave, grooming involves fewer limbs and elicits a significantly weaker global response.
These correlations occur on two distinct timescales: a fast component (seconds) reflecting immediate motor output and a slow component (up to a minute) likely representing internal states like arousal. Approximately 41% of cells exhibited fast dynamics, while 31% were associated with slower shifts in activity.
Local Microcircuits and Residual Activity
When the activity explained by simple behavioral vigor is mathematically "subtracted," the remaining "residual" activity is not random noise. Instead, it is high-dimensional (explained by 40–60 PCA modes) and organized into small, spatially symmetric clusters. These clusters, often consisting of as few as 20 cells, likely represent genetically defined cell types performing specialized local computations.
These local clusters are not isolated; they participate in the broader global dynamics. This suggests that specialized microcircuits perform their functions while being continuously informed of the animal's overall behavioral and arousal context.
Conclusion
By combining high-speed volumetric imaging with precise behavioral tracking, this research demonstrates that the fly brain simultaneously maintains global representations of state and high-dimensional local computations. The integration of these signals ensures that microcircuits with highly specified functions operate within the context of the larger organism's activity. Future work will focus on expanding these observations to the entire central nervous system and using optogenetic manipulations to establish causal links between these residual clusters and global brain states.
Summary updated