Grateful for a chance to celebrate the retirement of my advisor Don Johnson. What a joy to celebrate research impact (including signal processing & neuro), education, and countless people like me who had the trajectory of their lives changed by his selflessness and generosity.

I was grateful for the chance today to meet with the congressional offices that represent me to emphasize how important it is to have a stable long term investment in biomedical research funding.

While it's just a small case study, we examined human LFP data during the viewing of emotional videos. This model had less noisy switching between different states, which made it possible to identify the periods of emotional video viewing *and* the emotional content!

The result is a rich dynamical model that better identifies and captures the motifs in complex and time varying dynamical systems. We demonstrate this on synthetic test cases, as well as simulated neuroscience experiments.

The is a probabilistic generative model that represents dynamical systems as a fluctuating combination of dynamical primitives. It uses simple LDS primitives for interpretable analysis similar to switched LDS, but instead uses time-varying softer combinations of these primitives.

Get a free NIH BRAIN Initiative art calendar to prep for 2025: catalog.ninds.nih.gov/publications... .

I'm proud to say that the cover is a piece we created from the DTI imaging of a subject in the BRAIN funded depression DBS study we reported last year: go.nature.com/48lmlzC

🧪🧠