Mechanisms of Full-Body Intelligence: Interoception, Predictive Inference, Neurovisceral Coupling, and Sensorimotor Grounding
Authors/Creators
Description
Full-body intelligence has been proposed as an organism-level construct in which adaptive cognition and self-regulation depend partly on closed-loop interactions among neural, visceral, autonomic, sensorimotor, neuroendocrine, and environmental processes. This critical narrative review examines four mechanistic literatures most relevant to that proposal: interoception; predictive processing and interoceptive inference; neurovisceral and cardiorespiratory coupling; and grounded or sensorimotor cognition. The aim is not to claim that these literatures already establish a unified faculty of full-body intelligence, but to determine what each mechanism can legitimately contribute to a testable integrative framework. Interoception provides established pathways through which internal physiological signals are sensed and integrated, although its measurement is multidimensional and single-task indices are insufficient. Predictive processing, active inference, EPIC, and allostasis provide theoretical accounts of how internal and external signals may be integrated for regulation, while requiring careful empirical operationalization. Neurovisceral integration and cardiorespiratory resonance provide measurable examples of brain–body coupling, but heart-rate variability and physiological coherence are context-sensitive phenomena rather than general measures of intelligence. Grounded-cognition research supports modality-specific and multimodal recruitment during conceptual processing, whereas strong claims that primary motor cortex is obligatorily necessary for language comprehension remain unsupported. Across these domains, the review identifies a common architecture of sensing, inference, regulation, action, feedback, and learning while explicitly distinguishing established findings from theoretical interpretations and contested extrapolations. The resulting synthesis provides a mechanistic foundation for future construct validation and experimental investigation of full-body intelligence.
Other (English)
This manuscript is Preprint 2 of the Human Embodied Intelligence Research Program. It is a critical narrative review and mechanistic synthesis and has not undergone peer review. No new empirical dataset is reported. The manuscript explicitly distinguishes convergent empirical findings from theoretical models, integrative hypotheses, and contested or overextended interpretations. Full-body intelligence is treated as a proposed construct requiring empirical validation rather than as an established biological faculty.
Methods (English)
This research program was originated and directed by Daphne Garrido. ChatGPT (OpenAI) and Grok (xAI) were used as research-assistance tools during development of the broader program for iterative ideation, literature discovery, conceptual organization, adversarial critique, and drafting assistance. ChatGPT was additionally used for literature verification, critical synthesis, organization, and drafting support for this preprint. AI-generated assertions were not treated as scientific evidence. Scientific claims were grounded in identifiable scholarly literature, and responsibility for the framework, interpretations, and manuscript rests with the author.
Files
Mechanisms of Full Body Intelligence.pdf
Files
(191.8 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:761f632748570798bfa6f97e8f2faf18
|
191.8 kB | Preview Download |
Additional details
Related works
- Is part of
- Preprint: 10.5281/zenodo.22906965 (DOI)
- Is supplemented by
- Preprint: 10.5281/zenodo.22907785 (DOI)
- Preprint: 10.5281/zenodo.22907998 (DOI)
- Preprint: 10.5281/zenodo.22908268 (DOI)
- Preprint: 10.5281/zenodo.22924529 (DOI)