PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 5, 20260 citationsOpen Access

Multi-Channel Compact-Object Inference in Time-Scalar Field Theory: Expanded Populations, Correlation Structure, and Predictive Validation of the Λ Stability Functional

View Full Paper
JFJordan Gabriel Farrell

Key Points

  • The aim is to evaluate the Λ stability functional's predictive power across various compact astrophysical objects.
  • Expanded analysis beyond previous studies to include pulsars, magnetars, CCOs, BHXBs, and white dwarfs.
  • Utilized curated datasets for correlation, regression, and classification analyses without class-specific tuning.
  • Performed statistical tests to assess the performance of Λ-based models against baseline correlations.
  • Found that Λ-based models achieved a multi-class AUC of 0.981, indicating strong predictive capability.
  • Observed significant improvements in explained variance for radiative outputs, with R2 = 0.55 for BHXB peak X-ray flux compared to 0.32 for baseline.
  • Identified non-random residual structure that is class-dependent, suggesting meaningful interpretations of channel transitions.

Abstract

Time-Scalar Field Theory (TSFT) models compact astrophysical objects as solutions to coherence routing under compression, rather than as manifestations of distinct fundamental forces. In prior work, this framework was shown to organize magnetars and pulsars through a stability functional Λ derived from timing observables. Here we extend that analysis across a broader compact-object ladder—including canonical pulsars, magnetars, Central Compact Objects (CCOs), black-hole X-ray binaries (BHXBs), and white dwarfs—testing whether a single scalar functional retains explanatory and predictive power across regimes of increasing compactness and channel geometry. Using curated datasets from ATNF, the McGill Magnetar Catalog, De Luca (2017) CCO compilations, the BlackCAT BHXB catalog, and SDSS DR7 white dwarfs, we perform correlation, regression, and classification analyses without class-specific tuning. We find that Λ-based models outperform baseline correlations across multiple targets, achieving a multi-class AUC of 0.981 and statistically significant improvements in explained variance for radiative outputs (e.g., R2 = 0.55 vs. 0.32 for BHXB peak X-ray flux). Residual structure is shown to be non-random and class-dependent, suggesting physically meaningful channel misalignment rather than noise. These results elevate Λ from a descriptive magnetar diagnostic to a falsifiable, cross-population inference tool. We interpret residuals as signatures of channel transitions and outline explicit observational predictions, establishing a framework for survey-guided discovery without post hoc fitting.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Jordan Gabriel Farrell (2025) studied this question.

synapsesocial.com/papers/698434ebf1d9ada3c1fb3adchttps://doi.org/10.5281/zenodo.18458124
Ask AI
Helpful
Bookmark
Share
View Full Paper