Hi, I'm Shannon. I am a Ph.D. student working with Prof. David Heeger and
Prof. Jonathan Winawer at New York University.
I am interested in the computational cognitive neuroscience of human internally generated sensory experience, dreaming and consciousness.
My research sits at the intersection of computational neuroscience and perception.
I am interested in how the brain constructs and maintains an internal model of the world that remains coherent in the presence of noise,
uncertainty, and the absence of sensory input. In my current work, I study working memory as a dynamic, noisy, and normalised recurrent system,
asking why continuous representations systematically drift over time and load, and what computational principles can explain this behaviour.
More broadly, I am fascinated by internally generated experience, particularly dreaming, and how the brain can generate vivid, spatially and
temporally structured percepts without external input. By combining theoretical modelling and neural data, I aim to uncover the computational
and neural mechanisms that allow the brain to sustain and generate continuous subjective worlds, and how these mechanisms relate to the foundations
of perception and consciousness.
I completed my Master's degree in Mathematical and Theoretical Physics at the University of Oxford.
My thesis focused on computational models of time cells and place cells in the hippocampus CA3 subfield.
Working with Prof. Vijay Balasubramanian, I developed a recurrent neural network model that treats the hippocampus as a predictive system that reconstructs incomplete sensory experiences.
In this framework, both place cells and time cells emerge from a single recurrent circuit performing pattern completion,
with spatial or temporal representations arising depending on the structure of sensory input and task demands.
This work suggests that spatial and temporal coding in the hippocampus may reflect two regimes of a unified dynamical mechanism rather than separate neural processes.
