A Generalized Framework of Antisymmetric Polyspectral Indices for Identifying High-Order Neural Interactions
A Generalized Framework of Antisymmetric Polyspectral Indices for Identifying High-Order Neural Interactions
Authors: Alessio Basti, Rikkert Hindriks, Ruggero Freddi, Gian Luca Romani, Vittorio Pizzella, Guido Nolte, Laura Marzetti Date: 2026-05-06 Paper ID: openalex:2605.04636
Summary
This paper proposes a generalized framework of antisymmetric cross-polyspectral indices to characterize high-order neural interactions defined by frequency-sum relationships (fN = sum(fi)). These indices address the limitations of conventional metrics by providing inherent robustness against volume conduction and zero-lag artifacts. The authors validate the framework through cubic nonlinearity simulations and reveal novel high-order interactions in human EEG data with potential applications for personalized TMS modulation.
Key Contributions
- Introduces a generalized family of antisymmetric cross-polyspectral indices to quantify N-order harmonic dependencies in time series.
- Provides theoretical proofs that these indices are intrinsically robust to instantaneous mixing/volume conduction artifacts.
- Demonstrates significant high-order neural interactions in empirical EEG recordings that are not captured by standard spectral metrics.
Open Questions & Future Work
Key Concepts
- antisymmetric-cross-polyspectral-indices: A family of spectral indices that quantify high-order harmonic dependencies in time series while maintaining intrinsic robustness to instantaneous linear mixing artifacts.
Archivist Review
I have approved the core method as a new concept because it addresses a fundamental problem (volume conduction/instantaneous mixing) in multivariate time-series analysis using a formal polyspectral approach. I have approved one open question concerning the multivariate/multidimensional generalization bottleneck, while rejecting the clinical validation request as it represents standard experimental follow-up rather than a structural research question.
Approved Concepts
- Antisymmetric Cross-Polyspectral Indices: This provides a formal, robust mathematical framework for analyzing nonlinear multi-frequency interactions in time series while explicitly solving the volume conduction artifact problem prevalent in neuroimaging and related sensors.
Approved Open Questions
- Multivariate Polyspectral Framework Bottleneck: Advancing from univariate-frequency-centric analysis to full multivariate/multidimensional source-level connectivity is essential for high-resolution neuroimaging and complex sensor array processing.
Rejected Candidates
- [open_question] Stability and Clinical Validation (
validation-and-stability-of-polyspectral-indices) - low_impact: This is a boilerplate request for clinical validation, larger studies, and signal-to-noise ratio optimization, which fails to identify a specific, unresolved theoretical or mechanistic bottleneck.
Links
Metadata & Links
- url
- https://arxiv.org/abs/2605.04636
- paper_id
- 2605.04636
- paper_source
- openalex
- domain
- neuroscience
- tags
-
- architectures
-
- datasets
-
- concept_slugs
- antisymmetric-cross-polyspectral-indices
- dataset_slugs
-
- skill
- TimeSeriesSkill
- created_at
- 2026-05-09T05:58:11Z