A theory of multineuronal dimensionality, dynamics and measurement

TLDR

Neuroscience experiments record hundreds of neurons across many trials, and dimensionality‑reduction reveals that such high‑dimensional activity can be captured in a low‑dimensional space, prompting fundamental questions about the origins and implications of this simplicity. The authors ask how task complexity shapes neural dimensionality and the reliability of dynamical portraits when only a small fraction of neurons is recorded. They develop a theoretical framework and test it with physiological recordings from reaching‑task monkeys. The theory shows that task complexity limits dimensionality and determines when accurate dynamical portraits can be recovered, offering quantitative guidelines for designing large‑scale experiments.

Abstract

Abstract In many experiments, neuroscientists tightly control behavior, record many trials, and obtain trial-averaged firing rates from hundreds of neurons in circuits containing billions of behaviorally relevant neurons. Di-mensionality reduction methods reveal a striking simplicity underlying such multi-neuronal data: they can be reduced to a low-dimensional space, and the resulting neural trajectories in this space yield a remarkably insightful dynamical portrait of circuit computation. This simplicity raises profound and timely conceptual questions. What are its origins and its implications for the complexity of neural dynamics? How would the situation change if we recorded more neurons? When, if at all, can we trust dynamical portraits obtained from measuring an infinitesimal fraction of task relevant neurons? We present a theory that answers these questions, and test it using physiological recordings from reaching monkeys. This theory reveals conceptual insights into how task complexity governs both neural dimensionality and accurate recovery of dynamic portraits, thereby providing quantitative guidelines for future large-scale experimental design.