Cortical Dynamics and Mouse Behavior: Insights from Widefield Calcium Imaging

The mammalian cortex comprises multiple specialized regions that work in concert to perceive, plan, and execute behavior. In mice, as in other species, these regions are tightly integrated and dynamically adapt their interactions based on the behavioral context. Understanding how behavior modulates cortical dynamics is central to uncovering the brain’s organizational principles.

A Brain-Wide Approach to Studying Behavior

Widefield calcium imaging enables mesoscale recordings across the mouse neocortex. The high spatial coverage of this approach makes it well-suited to study how large-scale neural circuits interact during behavior. In this experiment, a transgenic mouse expressing the red calcium indicator jRGECO was imaged using the Light Track OiS200 system, with lime LED illumination at 567 nm and a frame rate of 20 Hz.

Spontaneous behavior was recorded for 30 minutes while the animal was awake and head-fixed on a low-friction belt treadmill. No explicit stimuli were applied. Behavior was concurrently captured via an infrared behavioral camera and the treadmill’s encoder signal.

Behavioral States and Their Classification

Four primary behavioral states were identified:

• Locomotion (determined from treadmill velocity > 5 mm/s)
• Whisking
• Grooming
• Eye movements

For non-locomotor behaviors, events were extracted using image-based features. For whisking and grooming, pixel intensity gradients within a region of interest (ROI) around the nose were used. Whisking was defined as changes >1 standard deviation (SD), grooming >2 SD. Eye movement was tracked by detecting shifts in pupil position, quantified as >2 SD displacement from baseline.

These behaviors were classified only during treadmill “quiet” periods to reduce confounding motor effects.

Calcium Imaging Analysis Pipeline

Fluorescence traces were processed via:

1. Global signal regression to remove shared variance.
2. Normalization to ΔF/F.
3. Segmentation into trials centered on each behavioral event.

Each trial consisted of a 0.5s baseline and a 3s post-onset window. Activity was averaged across trials for each behavior, revealing consistent cortical activation maps.

Distinct Cortical Signatures of Behavior

Despite the spontaneity and partial overlap of some behaviors, clear spatiotemporal patterns emerged:

Retrosplenial Cortex Is Consistently Recruited

Across all behavioral states—locomotion, whisking, grooming, and eye movements—the Retrosplenial cortex (RS) showed early activation. RS has been implicated in spatial learning and head direction encoding (Vann et al. 2009; van der Goes et al. 2024), suggesting it plays a central role in contextual navigation, even in the absence of external stimulation.

Behavior-Specific Cortical Activity

• Locomotion: Broad activation across motor and sensory areas, with prominent RS engagement.
• Whisking: Increased activity in somatosensory cortex and RS.
• Grooming: Distributed and variable activation, reflecting its complex motor sequence.
• Eye movements: Activation of posterior-medial (PM) cortex, a putative higher-order visual area possibly involved in visual tracking (Andermann et al. 2011).

These patterns reflect the neural substrates supporting spontaneous actions and suggest a modular yet integrated organization of the cortex.

Conclusion

This dataset demonstrates how widefield calcium imaging can uncover behaviorally relevant cortical dynamics in awake mice. Even in spontaneous, unstructured paradigms, consistent neural signatures align with specific behaviors. The robust activation of the retrosplenial cortex across conditions reinforces its proposed role in spatial cognition and internal mapping.

In the second part of this series, we will explore functional connectivity across cortical regions to investigate how networks reorganize in response to different behavioral states.

Products Used in This Study