Recursive Entropic Time: A Neural Framework for the Informational Construction of Subjective Duration

Abstract The conception of time as a universal, independent parameter is a foundational axiom of physical models, yet it fails to account for the subjective nature of temporal perception and leads to theoretical inconsistencies in complex systems. This paper introduces and provides multifaceted empirical support for the Recursive Entropic Time (RET) framework, a theory positing that subjective time is not a fixed background but an emergent property actively constructed by neural systems engaged in interpretive, associative processing. We hypothesize that the brain employs a functionally segregated system for temporal processing: primary sensory cortices negotiate objective, clock-based time ( t lab ), while higher-order associative cortices construct subjective time ( t RET ) via a mechanism where the rate of temporal flow is inversely modulated by moment-to-moment informational load. We tested this theory through a comprehensive, two-part investigation. Part I utilized a public EEG-fMRI dataset of subjects under the influence of DMT to test the anatomical specificity of RET. This analysis revealed a crucial discovery: the RET model’s efficacy, quantified by the improvement in correlation between neural dynamics and informational load, was significantly greater in associative regions (e.g., Lingual Gyrus) compared to primary sensory regions (V1) (p = 0.0068). This finding refuted a global model of RET and pointed toward a more nuanced, region-specific hypothesis. Part II was designed to probe the unique, non-linear dynamics predicted by this refined hypothesis. We conducted a mechanistic analysis on a public EEG dataset of a temporal reproduction task (Hassall et al., 2020), identifying two behaviorally identical trials (reproduced duration ≈ 0.78s). We demonstrated that their underlying informational load profiles were demonstrably different, a finding inconsistent with simpler linear scaling models. Crucially, the formal RET integral correctly predicted the behavioral outcome, yielding nearly identical total internal durations. Together, these results provide convergent evidence for RET, presenting it as a falsifiable and mechanistic account of how the brain constructs subjective time. We argue that time, as we experience it, is not a perception of an external property but an intrinsic construction of higher-order cognition. This work establishes a computationally tractable and falsifiable method for quantifying subjective time from first principles, with implications for creating dynamic “brain-time” atlases and understanding cognitive pathologies.