Ahola, O., Haxel, L., Ziemann, U. et al. · neuroscience
Methods 95%
Abstract
Evoked and induced responses are fundamental neurophysiological readouts in the development and assessment of treatments for disorders with impaired neural function. Distinguishing these responses is essential for interpreting single-trial variability, identifying the determinants of these responses, and enabling real-time probing of cortical responsiveness. However, because of dynamic overlap, separating evoked and induced activity remains unsolved at the single-trial level when instantaneous fluctuations of cortical responsiveness are of interest. We developed the Evoked-Ongoing/Induced Separator (EOIS), a framework for separating event-evoked, ongoing, and event-modulated induced activity from single-trial data. EOIS recovers evoked and ongoing/induced components by leveraging their distinct phase-consistency features in MEG and real-time EEG, enabling subject-agnostic single-trial time-series analysis and attribution of cortical response variability to specific components. By delivering component-resolved time series, EOIS enables biomarker refinement, pharmacodynamic readouts, and closed-loop neuromodulation, facilitating translational deployment across individuals and care settings.
Tootoonian, S., Kollo, M., Schaefer, A. T. · neuroscience
Computational models 95%Neural coding 78%
Abstract
Whole-cell recordings from olfactory bulb mitral/tufted cells in awake mice revealed an inverse relationship between baseline activation and odour responses wherein cells that were spontaneously active tended to be inhibited by odours, while `silent' cells, quiet at baseline, tended to have vigorous odour responses. In this work we provide a normative explanation for these findings by extending previous sparse coding models of the olfactory system to include `common feature subtraction,' an odour-triggered variant of background subtraction in which the receptor contribution of features common to the odours being discriminated is subtracted. Using a circuit-level formulation of this computation consistent with the anatomy of the olfactory bulb we show how baseline activity cancels the effect of the common olfactory features on receptor activations, promoting inference of the target odours. We provide simulation results showing qualitative agreement with experimental findings. The model also predicts in closed form, without fitting, the slopes of the relationships between a cell's baseline and its common input and between its baseline and its odour response, and these predictions agree with simulations. Our model makes the experimentally testable prediction that a cell's baseline activation is not an intrinsic property, but reflects the negative of its recent activity.
Chengcheng Wu, Zichen Shi, Jiayang Xu et al. · q-bio.NC
Methods 94%Computational models 39%
Abstract
Network neuroscience has traditionally relied on node-centric approaches, which are inherently limited in capturing the brain's overlapping functional architectures and higher-order edge-to-edge interactions. The emerging edge-centric paradigm addresses this gap by shifting the analytical unit from nodes to edges. However, while edge-centric models have been established for DTI and fMRI, comparable EEG methodologies have yet to be developed. In this study, we proposed a Phase-based Edge-centric Coupling (PEC) algorithm for EEG network modelling, where edge time series are derived from instantaneous phase differences of EEG signals and edge-to-edge coupling is quantified via normalized inner product. Simulation tests demonstrated the robustness of PEC algorithm, achieving high accuracy (<10% relative error) in coupling estimation and discriminative capacity (AUROC>0.7) against spurious coupling.When provided sufficient data length(>=16 cycles), PEC outperformed traditional Phase Node-centric Coupling (PNC) in both individual identifiability and cognitive-state identifiability, enabling superior characterization of stable individual neural signatures and dynamic task-dependent functional configurations. Furthermore, unlike PNC's hard-partitioned modularity, PEC captured highly overlapping yet functionally specialized modular structures and connector hubs, aligning with task-engaged cognitive processes. Applied to dyadic interactions, PEC identified distinct cross-brain modules encoding symmetric and asymmetric neural coordination. These findings establish PEC as a valid and powerful tool for EEG network modelling, providing new insights into the brain's complex, multifunctional organization at both intra-brain and inter-brain networks.
The analysis of transient oscillatory activity in neural recordings remains challenging for linear methods, particularly when signals are nonlinear, noisy, and distributed across multiple channels. This paper introduces a Reduced-Observable Extended Dynamic Mode Decomposition (RO-eDMD) algorithm using a Random Fourier Features dictionary to characterize high-dimensional data in electroencephalography (EEG) recordings. The method extends the Koopman operator framework by incorporating rank reduction directly in the observable space and employing a randomized feature map that approximates the Gaussian kernel, enabling the capture of transient components without explicit dictionary optimization. To assess the method and demonstrate applicability, a logistic regression model is trained using a working memory paradigm as a test case, using dominant power values extracted via RO-eDMD in the theta band as a predictor and the type of event (encoding vs.\ fixation) as the dependent variable. We find correlations between oscillatory power and memory encoding events in the analyzed bands. Additionally, we find an increase in correlation associated with theta power and memory load, replicating similar results in the literature. The presented algorithm offers a computationally efficient, interpretable framework applied to a working memory paradigm dataset, but is relevant to the analysis of neural time series from diverse experimental contexts.