Visualizing and sonifying neurodata (ViSoND) for enhanced observation.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Data Visualization

Background:

• High-dimensional neural and behavioral data present challenges for direct observation and biological interpretation.
• Computational methods reduce data complexity but can obscure biological relevance.
• Existing quantification methods may miss unexpected but important data aspects.

Purpose of the Study:

• To develop an open-source approach, Visualization and Sonification of NeuroData (ViSoND), for enhanced qualitative observation of complex neuroscience data.
• To integrate video display with data sonification for a more intuitive understanding of neural and behavioral dynamics.
• To improve the biological interpretability of complex neuroscience datasets.

Main Methods:

• Developed ViSoND, an open-source tool combining synchronized video and MIDI sonification.
• Mapped multiple physiological data streams to musical notes for auditory representation.
• Applied ViSoND to analyze mouse breathing rhythms and neural activity in the visual cortex.

Main Results:

• ViSoND successfully interpreted behavioral structures from mouse breathing rhythms analyzed by a computational model.
• ViSoND identified patterns of neural activity in mouse visual cortex associated with eye blinks, previously filtered out.
• Demonstrated ViSoND's capability to supplement quantitative analysis with observational insights.

Conclusions:

• ViSoND enhances the observational interpretability of complex neuroscience data.
• The approach provides an accessible method for displaying data, potentially broadening the audience for neuroscience findings.
• ViSoND bridges the gap between computational analysis and biological realism in neuroscience research.