A state-space model of cross-region dynamic connectivity in MEG/EEG

Authors: Ying Yang, Elissa Aminoff, Michael Tarr, Robert E. Kass

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Compared with a two-step method, which first obtains the commonly used minimum-norm estimates of source activity, and then fits the auto-regressive model, our state-space model yielded smaller estimation errors on simulated data where the model assumptions held. When applied on empirical MEG data from one participant in a scene-processing experiment, our state-space model also demonstrated intriguing preliminary results
Researcher Affiliation Academia Ying Yang Elissa M. Aminoff Michael J. Tarr Robert E. Kass Carnegie Mellon University, Fordham University
Pseudocode No The paper describes the EM algorithm using mathematical formulas and textual descriptions but does not include structured pseudocode or algorithm blocks.
Open Source Code Yes The implementation of the model and the E-M algorithm in Python is available at github.com/Ying Yang/MEEG_connectivity.
Open Datasets No We simulated MEG sensor data according to our model assumptions. The 306-channel MEG data were recorded while a human participant was viewing 362 photos of various scenes.
Dataset Splits No Each simulation had q = 200 trials, and 5 independent simulations for each a value were generated. Given the data, we estimated the dynamic connectivity between the neural responses to the 362 images in the two ROIs (EVC and PPA), using our state-space model and the two-step MNE method. We also bootstrapped the 362 observations 27 times to obtain standard deviations of entries.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The implementation of the model and the E-M algorithm in Python is available at github.com/Ying Yang/MEEG_connectivity. (Only specifies 'Python' without a version or any specific library versions).
Experiment Setup Yes In the prior of {At}T t=1, we set λ0 = 0 and λ1 = 0.1. ... Each simulation had q = 200 trials, and 5 independent simulations for each a value were generated. ... In the prior of {At}T t=1, we set λ0 = 0 and λ1 = 1.0; ... The data was down-sampled from a sampling rate of 1 k Hz to 100 Hz, and cropped within 100 700 ms, where 0 ms marked the stimulus onset. Together, we had T + 1 = 80 time points