Predictive Bodily Sensations: A Functional and Algorithmic Framework for Temporal Modeling under Uncertainty with Data-Informed Analysis

This article examines predictive bodily sensations (PBS) with respect to their potential relationship to future, emotionally relevant states. While previous work has provided initial statistical indications of such correlations, the present study focuses on the methodological and conceptual structuring required for their systematic analysis and practical use. A central challenge lies in the subjective nature of PBS, as their perception, parameterization, and recording depend entirely on the individual. This introduces systematic uncertainties, particularly in the estimation of duration and intensity of PBS, which are reflected in the dispersion of experimental data. Based on initial observations, a functional relationship between the PBS–Emotion Time Interval (PETI) and the PBS parameters duration (PD) and intensity (PI) is proposed in the form PETI = f(PD, PI). Although the available data are not sufficient for a statistically rigorous determination of this function, they exhibit a consistent structure that allows for a pragmatic mathematical approximation describing the central tendency of the data. This assessment is supported by an exploratory analysis of selected subject-based data, which reveals characteristic variability patterns of PETI depending on its temporal scale. At the same time, the calculated PETI values are inherently uncertain, as they rely on subjectively estimated input parameters. The resulting uncertainty overlaps with the experimental dispersion and prevents the interpretation of PETI as a precise point value. This leads to a key methodological consequence: PETI must be treated as a range rather than a single value. This consideration motivates the introduction of the Attention Time Window (ATW), which captures both the variability of the underlying relationship and the uncertainty of parameter estimation, enabling a robust temporal framework for the analysis and application of PBS. The proposed approach provides a foundation for algorithmic implementations, including personalized systems for real-time processing, while maintaining consistency with the limitations imposed by subjective data and limited statistical grounding.