Arousal Neurons’ Activity Explains Brain’s Blood Flow Dynamics in Mice

Why It Matters

The discovery refines the biological basis of the BOLD signal, implying that brain‑state fluctuations can confound functional imaging results. Accounting for arousal‑linked neurons will improve the accuracy of neuroimaging diagnostics and research.

Key Takeaways

• Arousal‑plus and arousal‑minus neurons predict blood flow better than bulk firing
• Neuropixels + functional ultrasound revealed distinct neurovascular coupling across brain states
• Thalamus contains more arousal‑plus neurons; somatosensory cortex more arousal‑minus
• Findings suggest fMRI must account for arousal‑related blood flow signals

Pulse Analysis

Neurovascular coupling—the relationship between neuronal activity and cerebral blood flow—has long underpinned functional imaging techniques such as fMRI. Traditional models assumed that aggregate firing across a region directly dictated blood volume changes, treating the brain’s vasculature as a uniform readout of overall activity. Recent advances, however, have hinted that this simplification may overlook critical nuances, especially when brain state fluctuates between wakefulness and sleep. Understanding these subtleties is essential for researchers who rely on blood‑oxygen‑level‑dependent (BOLD) signals to infer neural processes.

In the new Nature study, investigators combined high‑density Neuropixels electrophysiology with functional ultrasound imaging to monitor nearly 19,000 neurons in freely moving mice. They identified two opposing populations: arousal‑plus neurons that increase firing during wakeful, whisker‑driven exploration, and arousal‑minus neurons that fire more during sleep. Remarkably, the combined activity of these groups explained blood‑volume dynamics across thalamic and cortical regions far better than bulk firing rates. The distribution of the two types varied—thalamus favored arousal‑plus cells, while secondary somatosensory cortex housed more arousal‑minus neurons—yet the underlying coupling rule remained consistent throughout the brain.

The implications for human neuroimaging are profound. If similar arousal‑linked neuronal subsets exist in people, BOLD signals may reflect not just generic neural activation but also the brain’s vigilance state. Consequently, fMRI studies must incorporate behavioral or physiological markers of arousal to avoid conflating true task‑related activity with background state‑driven blood flow. This refined perspective promises more precise mapping of brain function, better interpretation of clinical scans, and a clearer path toward integrating multimodal data in neuroscience research.