Abstract:The brain produces diverse functions, from perceiving sounds to producing arm reaches, through the collective activity of populations of many neurons. Determining if and how the features of these exogenous variables (e.g., sound frequency, reach angle) are reflected in population neural activity is important for understanding how the brain operates. Often, high-dimensional neural population activity is confined to low-dimensional latent spaces. However, many current methods fail to extract latent spaces that are clearly structured by exogenous variables. This has contributed to a debate about whether or not brains should be thought of as dynamical systems or representational systems. Here, we developed a new latent process Bayesian regression framework, the orthogonal stochastic linear mixing model (OSLMM) which introduces an orthogonality constraint amongst time-varying mixture coefficients, and provide Markov chain Monte Carlo inference procedures. We demonstrate superior performance of OSLMM on latent trajectory recovery in synthetic experiments and show superior computational efficiency and prediction performance on several real-world benchmark data sets. We primarily focus on demonstrating the utility of OSLMM in two neural data sets: μ ECoG recordings from rat auditory cortex during presentation of pure tones and multi-single unit recordings form monkey motor cortex during complex arm reaching. We show that OSLMM achieves superior or comparable predictive accuracy of neural data and decoding of external variables (e.g., reach velocity). Most importantly, in both experimental contexts, we demonstrate that OSLMM latent trajectories directly reflect features of the sounds and reaches, demonstrating that neural dynamics are structured by neural representations. Together, these results demonstrate that OSLMM will be useful for the analysis of diverse, large-scale biological time-series datasets. Extracting insight from data into the dynamic processes that produce observed phenomena is a ubiquitous challenge in biology. For example, brain functions are generated by the collective activity of many neurons. Extracting insight from biological time-series data can be challenging because the number of observations can be large, activities can be noisy, and collective dynamics can be complex. Here, we developed a new latent processes Bayesian regression model, the orthogonal stochastic linear mixing model (OSLMM) to address these challenges. Compared to competing methods, we demonstrate that OSLMM has superior recovery performance of latent trajectories in synthetic experiments and is computationally efficient. In two diverse neural data sets, we find that the latent spaces extracted from OSLMM are directly structured by task parameters. Our results demonstrate that latent dynamics can be structured by representations and suggest that OSLMM will be useful for data-driven discovery in large-scale biological time-series data.

Characterizing the nonlinear structure of shared variability in cortical neuron populations using latent variable models

Linear dynamical neural population models through nonlinear embeddings

Modeling dynamic neural activity by combining naturalistic video stimuli and stimulus-independent latent factors

Bayesian inference of structured latent spaces from neural population activity with the orthogonal stochastic linear mixing model

Nonlinear Modeling of Neural Interaction for Spike Prediction Using the Staged Point-Process Model

Multi-Modal Latent Variables for Cross-Individual Primary Visual Cortex Modeling and Analysis

Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics

Time-varying generalized linear models: characterizing and decoding neuronal dynamics in higher visual areas

Disparate nonlinear neural dynamics measured with different techniques in macaque and human V1

Analysis and modelling of variability and covariability of population spike trains across multiple time scales

Deciphering neuronal variability across states reveals dynamic sensory encoding

Recurrent Switching Dynamical Systems Models for Multiple Interacting Neural Populations

Latent dynamics of primary sensory cortical population activity is structured by fluctuations in the local field potential

Hierarchical models for neural population dynamics in the presence of non-stationarity

Learning identifiable and interpretable latent models of high-dimensional neural activity using pi-VAE

Cortical state contributions to response variability in the early visual cortex: A system identification approach

Modeling Latent Neural Dynamics with Gaussian Process Switching Linear Dynamical Systems

A flow-based latent state generative model of neural population responses to natural images

Mesoscopic modeling of hidden spiking neurons

Mesoscale neuronal granular trial variability in vivo illustrated by nonlinear recurrent network in silico