Kevin Franks
Duke Institute for Brain Sciences, Duke University School of Medicine
Respiration coordinates the cortical olfactory code
11:00 am, Tuesday 07 October 2025
Location: Florence Buchanan Lecture Theatre, Sherrington Building
Abstract: Neurons in early sensory cortices typically respond to simple stimulus features that vary along continuous axes. In vision and hearing, neurons exhibit simple feature tuning, such as selectivity for edges or tones. These basic responses provide the building blocks upon which richer representations are built. In olfaction, however, no such simple feature tuning has been observed, precluding a clear logic for cortical odour coding. I will show that respiration provides the organizing scaffold. Individual piriform neurons are tuned to specific phases of the sniff cycle, and their phase preferences are consistent across inputs. Phase‑locked ensembles organize population activity along a low‑dimensional cylindrical manifold, supporting flexible and enriched odour representations. I will also discuss the circuit processes that support this coding scheme. More broadly, these findings identify respiration as an organizing axis of cortical computation and show how rhythmic physiological signals structure neural population dynamics.
Biography: Kevin Franks was born in South Africa in the 1970s, moved to Canada in the 1980s, and then to California in the 1990s. He is not, however, a fascist or a billionaire. He earned his Ph.D. at UCSD with Terry Sejnowski, focusing on simulations of synaptic transmission. In his postdoc work, first with Jeffry Isaacson at UCSD, and then with Richard Axel and Steve Siegelbaum at Columbia, he studied the synaptic organization of the piriform cortex. This work provided the foundation for his current research into the mechanisms underlying cortical coding. In 2013, he established his lab at Duke University, where he is now an Associate Professor. His research uses the mouse olfactory system to investigate the neural mechanisms underlying sensory perception and cortical computation.