Uncovering motifs of concurrent signaling across multiple neuronal populations

Authors: Evren Gokcen, Anna Jasper, Alison Xu, Adam Kohn, Christian K. Machens, Byron M Yu

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We illustrate these features in simulation, and further validate the method by applying it to previously studied recordings from neuronal populations in macaque visual areas V1 and V2. Then we study interactions across select laminar compartments of areas V1, V2, and V3d, recorded simultaneously with multiple Neuropixels probes.
Researcher Affiliation Academia 1Dept. of Electrical and Computer Engineering, Carnegie Mellon University 2Dominick Purpura Dept. of Neuroscience, 3Dept. of Ophthalmology and Visual Sciences, 4Dept. of Systems and Computational Biology, Albert Einstein College of Medicine 5Champalimaud Neuroscience Programme, Champalimaud Foundation 6Dept. of Biomedical Engineering, Carnegie Mellon University
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes A MATLAB (Math Works) implementation of m DLAG is available on Git Hub at http://github. com/egokcen/m DLAG and on Zenodo at https://doi.org/10.5281/zenodo.10048163 [41].
Open Datasets Yes We turned to neural recordings from two areas in the macaque visual cortex, V1 and V2 [24]. ... Zandvakili, A. & Kohn, A. Paired V1-V2 neuronal spiking responses in anesthetized macaque monkey. https://doi.org/10.6080/K0B27SHN (2019).
Dataset Splits No Each dataset included 400 trials: we allocated at random 300 trials as a training set and 100 trials as a test set on which to measure model performance (we employed a leave-group-out prediction metric; see Supplementary Section S4). ... For each dataset, we allocated at random 225 trials as a training set and 75 trials as a test set on which to measure model performance. A separate explicit validation split is not mentioned.
Hardware Specification No The paper mentions Neuropixels probes for data recording but does not specify any hardware used for computation or model training (e.g., CPU, GPU models, memory).
Software Dependencies No The paper mentions that the implementation is in MATLAB but does not provide specific version numbers for MATLAB or any other software libraries/dependencies.
Experiment Setup Yes We set the initial number of latents (p = 10) to be greater than the ground truth number (p = 7)... Each trial was 500 ms in length, comprising T = 25 time points with a sampling period of 20 ms... all hyperparameters were fixed to a very small value, β, aϕ, bϕ, aα, bα = 10 12