Author: Rmon Kalermon⚡(with Grok 4.6 🌠 of SPACEXAI) — September 16, 2026
Abstract
The Human Genome Project showed that human complexity arises less from gene number than from regulatory architecture. Neuroepigenetics now shows that experience—including language use, social practice, and affective states—can alter chromatin state, enhancer activity, and transcription in neurons, with lasting effects on circuit structure and function. These mechanisms supply a molecular implementation layer for the claim that discourse remolds the brain as an instrument while agency remains a person-level capacity. Activity-dependent gene programs, histone modifications, and DNA-methylation changes associated with bilingual experience, memory reconsolidation, and narrative intervention illustrate how Stream A (speech acts) and Stream B (filmed conduct) can leave inspectable traces. They do not transfer authorship to genes or readiness potentials. In Carlo Rovelli’s terms, marks, utterances, circuits, and the pronoun “I” are not different in metaphysical kind, and none of them is a last layer from which the person can be decoded or retired. A hybrid archive can treat epigenetic marks as later-readable physiology aligned against a person-level codebook, not as a decoded mind.
Keywords: neuroepigenetics; experience-dependent plasticity; discursive psychology; language network; default mode network; agency; hybrid method; scientific realism
1. Introduction
Kalermon (2026) argued that agency belongs to persons, that brains function as instruments that discourse helps to build, and that a hybrid audiovisual-plus-physiology archive can make that ontology inspectable. Current neuroscience supports the instrument claim and undermines both the equation of mind with language and the folk-Libet retirement of the agent. The present article adds one further implementation layer: epigenetic regulation.
After the Human Genome Project, the surprise was not merely that humans possess roughly 20,000–25,000 protein-coding genes—comparable to far simpler organisms—but that complexity resides in regulatory networks, enhancers, chromatin architecture, and experience-sensitive transcription. Neuroepigenetics demonstrates that these regulatory processes remain plastic throughout life. Sensory and social experience, including intensive language use, drives activity-regulated gene programs that remodel circuits (West, 2025; Peña, 2026). This is compatible with Luria’s and Harré’s direction of fit: use reshapes organ (Luria, 1973; Harré, 2012). It is incompatible with any reading that treats genes as destiny or the person as an afterthought to a wave or a methylation mark.
The philosophical pressure on that last sentence is easy to miss. A recurring temptation in both popular neuroscience and a certain analytic realism is to treat the latest implementational trace as the real thing: the readiness potential as the decision, the language network as thought, the methylation mark as the self. Rovelli (2026) calls the broader habit naive realism: the expectation that science will eventually deliver a list of things with intrinsic properties, viewed from nowhere, that exhausts what the world is. His counter-claim is not anti-scientific. It is that electrons and minds, stones and laws, judgments and galaxies are not fundamentally different in kind, that properties are relational, and that neither science nor philosophy reaches a final layer. The present argument takes that claim into the archive. Experience-dependent regulation shows how discourse can remodel the brain as instrument. It does not identify a last layer at which the person is either abolished or saved.
2. Regulatory complexity, not gene count
Less than 2 percent of the genome codes for proteins; the remainder includes extensive regulatory sequence. Maps of gene-regulating networks in human brain tissue identify hundreds of thousands of cell-type-specific cis-regulatory elements whose variation is linked to brain-related traits (Emani et al., 2024). Chromosomal looping and three-dimensional chromatin organization bring distant enhancers into contact with promoters, providing an additional layer of control that is itself experience-sensitive (Pumo & Rijli, 2025). Developmental maps of DNA modification in human hippocampus and prefrontal cortex further show that regulatory state is reorganized across late gestation, infancy, and later life, rather than fixed at birth (Luo, Paredes, et al., 2026).
These facts dissolve the older expectation that a larger gene catalog would explain human psychological complexity. They also dissolve simple genetic determinism. The same genome can support different transcriptional programs according to developmental timing, cell type, and ongoing activity. That flexibility is precisely what allows person-level practices to leave durable traces in the instrument without the practices being identical to the traces.
The same genome supporting different programs is not a loophole in determinism. It is the molecular form of a more general fact: properties of the instrument exist in relation to use. The practice is not identical with the trace, and the trace is not a stand-in for a view from nowhere.
3. Neuroepigenetics of experience-dependent plasticity
Persistent changes in brain function after birth depend on altered programs of gene expression rather than on wholesale replacement of cells (West, 2025). Neuronal activity induces immediate-early genes and subsequent waves of transcription that orchestrate synaptic and circuit adaptations. Chromatin mechanisms—histone modifications, enhancer activation, remodeling of three-dimensional genome architecture—orchestrate temporally distinct transcriptional responses and allow cell-type- and stimulus-specific programs (Pumo & Rijli, 2025; Griffith, West, & Greenberg, 2024).
Enhancers are particularly well positioned to fine-tune adaptive plasticity. Enduring stimulus-induced changes in enhancer state can modify later responses, contributing to a form of metaplasticity (Griffith et al., 2024). Early-life experience, including stress and linguistic environment, becomes encoded in DNA methylation and histone marks that persist and can influence later plasticity windows (Peña, 2026). These processes operate in both critical periods and adult life, although magnitude and reversibility differ by region and age.
The evidence is strongest in animal models and in specific human interventions. The principle is now established: experience does not merely modulate firing rates; it changes the regulatory state of the genome inside neurons. That principle is an implementation claim. It is not an identity claim.
4. Language use as a driver of remodeling
Experience-dependent structural plasticity in bilingual brains—expansion followed by renormalization—has a plausible molecular route in these activity-dependent programs (Pliatsikas, 2020). Language learning and switching constitute intense, sustained, integrative experience. Structural changes in gray and white matter, including hippocampal volume and language-related tracts, follow nonlinear trajectories predicted by models of skill acquisition.
Early or discontinued language exposure can leave lasting white-matter signatures, consistent with sensitive-period effects that epigenetic mechanisms help close or reopen. An epigenetic model has already been proposed for spoken-language development after cochlear implantation, in which dynamic gene expression is taken to organize environmental influences on auditory learning (Markman et al., 2011). Direct human-brain epigenetic maps of bilingual experience remain sparse. The general machinery—activity-regulated transcription, chromatin remodeling, and experience-sensitive methylation—is the same machinery shown to operate in other domains of learning and memory.
A specialized language network remains dissociable from much thought (Fedorenko, Ivanova, & Regev, 2024; Fedorenko, Piantadosi, & Gibson, 2024). Epigenetic regulation does not collapse that dissociation. It supplies one reason the network can be sculpted by use without becoming the sole medium of cognition. Language is a tool and a practice. It is not the inner essence of the person, and its traces are not the person twice over.
5. Narrative self, reconsolidation, and the default-mode network
Default-mode and related medial systems support autobiographical construction over long timescales (Yeshurun, Nguyen, & Hasson, 2021). Memory reconsolidation—the process by which retrieved memories become labile and require restabilization—depends on gene transcription and is regulated by epigenetic modifications. In mice, cell-type- and locus-specific epigenetic editing of the Arc promoter in engram ensembles can bidirectionally and reversibly alter memory expression, including for already consolidated memories (Coda, Watt, Glauser, et al., 2025). That is causal evidence in a model system. It is not evidence that human self-narration rewrites a single locus in the same way.
Narrative Exposure Therapy, a structured discursive intervention that places traumatic events in autobiographical time and space, has been associated with saliva methylome-wide changes overlapping pathways involved in memory formation, extracellular-matrix interaction, and glucocorticoid signaling (Carleial et al., 2021). The finding is important and limited. Peripheral methylation is not neuronal chromatin. Group-level differentially methylated positions are not a codebook of a life. The result does not prove that every act of self-narration rewrites the epigenome in a simple one-to-one fashion. It does show that person-level narrative practice can engage regulatory machinery implicated in lasting change.
A minimal bodily self (ownership, interoception) remains distinct. Epigenetic marks in interoceptive or salience systems would constitute further instrument-level data, not a reduction of the standpoint indexed by “I.”
6. Agency stays with the person
None of the foregoing transfers authorship. Activity-dependent transcription and chromatin change are mechanisms of preparation, plasticity, and control. Damage or pharmacological disruption of these mechanisms can distort powers of action; that is implementation failure, not proof that “the genome decided.”
The readiness potential remains a poor candidate for a hidden decision that retires the agent. Recent reinterpretations treat it as fluctuating activity that can be backward-averaged into a ramp, or as a signal more closely tied to expectation than to irrevocable commitment (Schurger, Hu, Pak, & Roskies, 2021). Meaningful, reason-giving choices often lack the same early signal (Maoz, Yaffe, Koch, & Mudrik, 2019). Epigenetic states may bias the probability of certain circuit dynamics, just as genes and early experience bias temperament. Bias is not identity. The subject of “I chose,” “I failed,” and “I repaired this life” remains the embodied, addressable person.
This is the point at which naive realism and a soteriological reading of the archive both fail. Naive realism treats the mark or the potential as the real decision, hidden beneath the person’s report. A soteriological reading treats the same traces as the site at which a life is certified as repaired. Both look for a last layer. Rovelli’s argument against an ultimate inventory of the world applies here without any need to import his physics as evidence. If there is no view from nowhere, then there is no decoded mind at the bottom of the dataset. Epigenetic bias can raise or lower the cost of a later act. It cannot retire the agent, and it cannot certify salvation. “I repaired this life” is a person-level claim. Marks may show whether the instrument moved with that claim. They cannot become the claim.
Harré’s tool metaphor remains exact: the brain can be thought of as a tool (Harré, 2012). Tools have histories. They wear, they are retuned, they can be damaged. The user of the tool is not an extra ghost, and is not identical with the wear.
7. Implications for the hybrid method
The original proposal called for long-horizon audiovisual records supplying verbal and nonverbal labels, later-readable physiology (including future BCI), AI-assisted alignment, and human scientists retaining person-level categories. Epigenetic data—whether from accessible proxies such as saliva or blood, or from future higher-resolution, cell-type-aware sampling—can be added as another instrument track.
Four theoretically decisive couplings remain:
- cheap recited talk versus costly autobiographical and affective talk;
- talk/image splits;
- talk/action splits;
- talk/epigenetic-mark splits, in which a commitment is spoken yet the regulatory state of relevant circuits does not shift, or vice versa.
The fourth coupling is not a failure to find the real agent. It is evidence that agency is not located in any single stream. Speech act, filmed conduct, circuit dynamics, and regulatory state are perspectives on a person-level practice. They can be aligned. They do not add up to a God’s-eye person, and they do not need to.
Artificial intelligence may cluster alignments. It cannot decide what counts as a livable life or a genuine retraction. Those remain person-level facts. Rovelli’s reminder that physical law does not exhaust the world has a local analogue: a complete instrument track would still not exhaust the person. Physics is one window (Rovelli, 2026). So is the methylome.
Governance requirements intensify. Intimate neural and epigenetic records demand protections that outlast enthusiasm for amalgamation. Peripheral proxies must not be treated as neuronal ground truth. N = 1 remains a strength for idiographic mechanism and a limit for species inference.
8. Scope, limits, and conclusion
The program can make implementation of a discursive, personal mind inspectable at a grain group studies lack. It cannot identify a single epigenetic mechanism of the discursive mind, cannot treat every methylation difference as meaning, and cannot replace comparative, developmental, or lesion work. Rapid popular claims of large gene-expression shifts after minutes of emotion require far stronger, independently replicated evidence than currently exists for most human psychological practices. Human inference from saliva or blood to circuit-level chromatin remains indirect. Causal locus-specific editing of memory expression is presently a rodent result.
What the evidence does support is compact. Agency belongs to persons. Brains, including their regulatory genomes, are instruments that discourse and action help to build. Neuroepigenetics supplies a molecular route for that building without absorbing the builder. Electrons and minds, marks and judgments, circuits and lives are not different in kind in the sense that only one of them is really there. They are different in grain, in method, and in the questions they can answer. The pronoun “I,” aimed at a livable life, stays in the dataset because no implementation layer is the last layer. Marks, kept in their place, show cost and coupling. That is as much unity as the problem will bear, and more than a decoded mind would give.
References
Carleial, S., Nätt, D., Unternährer, E., Elbert, T., Robjant, K., Steudte-Schmiedgen, S., Kolassa, I.-T., & Koebach, A. (2021). DNA methylation changes following narrative exposure therapy in a randomized controlled trial with female former child soldiers. Scientific Reports, 11, 18493. https://doi.org/10.1038/s41598-021-98067-9
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204. https://doi.org/10.1017/S0140525X12000477
Clark, A. (2016). Surfing uncertainty: Prediction, action, and the embodied mind. Oxford University Press.
Coda, D. M., Watt, L., Glauser, L., Batiuk, M. Y., Burns, A. M., Stahl, C. L., Wong, L. Y., & Gräff, J. (2025). Cell-type- and locus-specific epigenetic editing of memory expression. Nature Genetics, 57, 2661–2668. https://doi.org/10.1038/s41588-025-02368-y
Emani, P. S., Liu, J. J., Clarke, D., Jensen, M., Warrell, J., Gupta, C., Meng, R., Lee, C. Y., Xu, S., Dursun, C., Lou, S., Chen, Y., Chu, Z., Galeev, T., Hwang, A., Li, Y., Ni, P., Zhou, X., PsychENCODE Consortium, Girgenti, M., Zhang, J., Wang, D., Geschwind, D., & Gerstein, M. (2024). Single-cell genomics and regulatory networks for 388 human brains. Science, 384(6698), eadi5199. https://doi.org/10.1126/science.adi5199
Fedorenko, E., Ivanova, A. A., & Regev, T. I. (2024). The language network as a natural kind within the broader landscape of the human brain. Nature Reviews Neuroscience, 25(5), 289–312. https://doi.org/10.1038/s41583-024-00802-4
Fedorenko, E., Piantadosi, S. T., & Gibson, E. A. F. (2024). Language is primarily a tool for communication rather than thought. Nature, 630(8017), 575–586. https://doi.org/10.1038/s41586-024-07522-w
Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138. https://doi.org/10.1038/nrn2787
Gazzaniga, M. S. (2000). Cerebral specialization and interhemispheric communication: Does the corpus callosum enable the human condition? Brain, 123(7), 1293–1326. https://doi.org/10.1093/brain/123.7.1293
Griffith, E. C., West, A. E., & Greenberg, M. E. (2024). Neuronal enhancers fine-tune adaptive circuit plasticity. Neuron, 112(18), 3043–3057. https://doi.org/10.1016/j.neuron.2024.08.002
Harré, R. (1983). Personal being: A theory for individual psychology. Blackwell.
Harré, R. (1991). The discursive production of selves. Theory & Psychology, 1(1), 51–71. https://doi.org/10.1177/0959354391011004
Harré, R. (1998). The singular self: An introduction to the psychology of personhood. Sage.
Harré, R. (2012). The brain can be thought of as a tool. Integrative Psychological and Behavioral Science, 46(3), 387–394. https://doi.org/10.1007/s12124-012-9195-x
Harré, R., & Gillett, G. (1994). The discursive mind. Sage.
Hodson, R., Mehta, M., & Smith, R. (2024). The empirical status of predictive coding and active inference. Neuroscience & Biobehavioral Reviews, 157, 105473. https://doi.org/10.1016/j.neubiorev.2023.105473
Hohwy, J. (2013). The predictive mind. Oxford University Press.
Kalermon, R. (2026). Keeping the agent in the dataset — person-level agency, neural implementation, and a hybrid method for discursive psychology.
Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential). Brain, 106(3), 623–642. https://doi.org/10.1093/brain/106.3.623
Luo, C., Paredes, M. F., et al. (2026). A map of DNA modification in the developing human hippocampus and prefrontal cortex. Nature.
Luria, A. R. (1973). The working brain: An introduction to neuropsychology. Penguin.
Maoz, U., Yaffe, G., Koch, C., & Mudrik, L. (2019). Neural precursors of decisions that matter—an ERP study of deliberate and arbitrary choice. eLife, 8, e39787. https://doi.org/10.7554/eLife.39787
Markman, T. M., Quittner, A. L., Eisenberg, L. S., Tobey, E. A., Thal, D., Niparko, J. K., Wang, N.-Y., & the CDaCI Investigative Team. (2011). Language development after cochlear implantation: An epigenetic model. Journal of Neurodevelopmental Disorders, 3(4), 388–404. https://doi.org/10.1007/s11689-011-9098-z
Newen, A., De Bruin, L., & Gallagher, S. (Eds.). (2018). The Oxford handbook of 4E cognition. Oxford University Press.
Parr, T., Pezzulo, G., & Friston, K. J. (2022). Active inference: The free energy principle in mind, brain, and behavior. MIT Press.
Peña, C. J. (2026). Epigenetic regulation of brain development, plasticity, and response to early-life stress. Neuropsychopharmacology, 51(1), 5–15. https://doi.org/10.1038/s41386-025-02179-z
Pliatsikas, C. (2020). Understanding structural plasticity in the bilingual brain: The Dynamic Restructuring Model. Bilingualism: Language and Cognition, 23(2), 459–471. https://doi.org/10.1017/S1366728919000130
Pumo, G., & Rijli, F. M. (2025). Chromatin regulation of neuronal activity-dependent gene programs in circuit formation and plasticity. Current Opinion in Neurobiology, 92, 103024. https://doi.org/10.1016/j.conb.2025.103024
Rovelli, C. (2026). On the equality of all things: Lessons on physics and philosophy. Scribner.
Schurger, A., Hu, P. B., Pak, J., & Roskies, A. L. (2021). What is the readiness potential? Trends in Cognitive Sciences, 25(7), 558–570. https://doi.org/10.1016/j.tics.2021.04.001
Sprevak, M. (2024). An introduction to predictive processing models of perception and decision-making. Topics in Cognitive Science, 16(2). https://doi.org/10.1111/tops.12704
West, A. E. (2025). Genetic mechanisms of experience-dependent neuronal plasticity. Annual Review of Genetics, 59, 435–459. https://doi.org/10.1146/annurev-genet-020325-103824
Williams, D. (2018). Predictive processing and the representation wars. Minds and Machines, 28(1), 141–172. https://doi.org/10.1007/s11023-017-9441-6
Wittgenstein, L. (1953). Philosophical investigations. Blackwell.
Yeshurun, Y., Nguyen, M., & Hasson, U. (2021). The default mode network: Where the idiosyncratic self meets the shared social world. Nature Reviews Neuroscience, 22(3), 181–192. https://doi.org/10.1038/s41583-020-00420-w
The End