Research

Inter-areal computation and cortical population dynamics

My active thrust is inter-areal computation, with David J. Freedman’s lab: a geometric and dynamical-systems theory of how the activity of one brain area shapes the computation in another, tested against multi-area recordings, with the FEF preprint as its data anchor. It builds on my established Doiron-lab line — balanced-network theory extended to nonlinear, population-wide correlated activity (Nature Communications 2025); a spiking-network account of how attention rotates a population code, with the Cohen lab; retinal noise resilience with the Wei lab; and the cortical-versus-thalamic origins of behavioral modulation in V1.

Retinal disinhibition and noise resilience

Biologically plausible circuit models of retinal motifs, asking how distinct cell types collectively keep the retina’s output reliable across environments whose noise statistics differ by orders of magnitude.

Dynamic-network construction and sampling over joins

Algorithms and data platforms for the exact, fast construction of dynamic networks from large time series; i.i.d. sampling over unions of joins; fairness-aware data integration.

Perceptual decision-making across the lifespan

How perceptual decisions unfold over time and change with age under varying perceptual noise, and the social effects of crowd gaze on visual search.

Stress neurobiology from imaging, via representation learning

Graph neural networks, dimensionality reduction and representation learning applied to human fMRI and mesoscopic calcium imaging of stress neurobiology.

Abstract-rule learning and visual decision confidence

Semi-supervised learning for Raven’s Progressive Matrices, and computational models of visual decision-making and confidence in OCD from behavioral and fMRI data.

Supervised training algorithms for spiking networks

With Alex Iosevich — supervised methods for training spiking neural networks.

Perception versus cognition, and machine experience

With Alison Peterman — the relationship between computational models of visual perception and visual phenomenal experience.