BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Carina Curto (The Pennsylvania State University - USA) DTSTART:20220401T160000Z DTEND:20220401T170000Z DTSTAMP:20260423T041508Z UID:GEOTOP-A/16 DESCRIPTION:Title: Dynamically relevant motifs in inhibition-dominated networks\nby Car ina Curto (The Pennsylvania State University - USA) as part of GEOTOP-A se minar\n\n\nAbstract\nMany networks in the brain possess an abundance of in hibition\, which serves to shape and stabilize neural dynamics. The neuron s in such networks exhibit intricate patterns of connectivity whose struct ure controls the allowed patterns of neural activity. In this work\, we ex amine inhibitory threshold-linear networks (TLNs) whose dynamics are const rained by an underlying directed graph. We develop a set of parameter-inde pendent graph rules that enable us to predict features of the dynamics\, s uch as emergent sequences and dynamic attractors\, from properties of the graph. These rules provide a direct link between the structure and functio n of inhibition-dominated networks\, yielding new insights into how connec tivity shapes dynamics in real neural circuits. Recently\, we have used th ese ideas to classify dynamic attractors in a two-parameter family of TLNs spanning all 9608 directed graphs of size n=5. Remarkably\, we find a str iking modularity in the dynamic attractors\, with identical or near-identi cal attractors arising in networks that are otherwise dynamically inequiva lent. This suggests that\, just as one can store multiple static patterns as stable fixed points in a Hopfield model\, a variety of dynamic attracto rs can also be embedded in TLNs in a modular fashion.\n LOCATION:https://researchseminars.org/talk/GEOTOP-A/16/ END:VEVENT END:VCALENDAR