A Transceiver Interpretation of Post-Mortem Cortical Plasticity: Information Dynamics Explains Ex-Vivo Associative Learning in Human Brain Explants
Year: 2026
Type: preprint
Abstract: In the framework of Information Dynamics, the brain is modeled not as a classical information processor, but as a transceiver coupled to a non-temporal virtual information space. The virtual space is assumed to have a geometric structure (Gaussian curvature \(K=1/4\), derived from qubit Fubini–Study geometry), and all measurable neural delays arise from finite-time "sample–encode–readout" cycles of the real-space transceiver: \[T_{\mathrm{up}} = \frac{\tau_0}{K} = \frac{10\ \mathrm{ms}}{0.25} = 40\ \mathrm{ms}, \quad T_{\mathrm{cycle}} \approx 150\text{--}200\ \mathrm{ms}.\] This framework has been used to explain perceptual delays (Libet-type experiments, simple reaction times), near-death \(\gamma\)-power surges, and cross-species reaction time scaling. Application to OPAB experiment Applying this transceiver model to OPAB experiment findings, I propose the following interpretations: 1. Post-mortem learning as residual coupling: The explant operates in a "residual" transceiver mode—partially viable hardware maintaining weak but non-zero virtual–real coupling. The improvement from ~31% to ~58% accuracy over three days reflects the gradual self-organization of the coupling matrix \(\widetilde{K}(x)\), driven by repeated pairing of sensory and motor patterns. 2. 17-day retention without rehearsal: This becomes intelligible if the learned pattern is stored not in the explant's synaptic biochemistry (which decays), but in the atemporal virtual space. The explant on Day 17 functions as a degraded but re-tunable receiver, re-coupling to a persistent virtual pattern. This would explain why memory survives without protein synthesis-dependent consolidation. 3. 12-hour PMI cutoff: The sharp boundary reflects the physical lifetime of the transceiver's sampling clock (\(\tau_0\)). Once ATP-dependent ion gradients collapse beyond a critical threshold, population oscillations cease abruptly—the transceiver transitions from "residual" to "silent" mode. 4. Pharmacological abolition (CNQX/APV): Blocking AMPA/NMDA receptors disrupts the transceiver's physical sampling hardware, causing the uplink encoder to fail (\(T_{\mathrm{up}}\) diverges) even though the virtual pattern itself remains unchanged. 5. Spatial electrode-neighbour effects: The spread of peak motor responses to neighbouring electrodes reflects the spatial continuity of the coupling matrix—information in the virtual space exists as continuous field patterns, not pixelated to individual electrodes. Testable predictions The model generates several predictions that could be tested in OPAB experiment platform: - Prediction 1: Baseline \(\gamma\)-oscillation frequency of the explant should positively correlate with learning rate (faster \(\gamma\) → shorter cycle time → more communication cycles per unit time → faster self-organization).- Prediction 2: Information purity \(p_{\mathrm{ID}}\) (operationalised as the ratio of spike-train entropy to stimulus entropy) should increase monotonically over training in learners.- Prediction 3: Two explants from the same donor should show partial cross-transfer—training Explant A should accelerate learning in naive Explant B exposed to the same virtual pattern, if the virtual space is globally coherent.- Prediction 4: Retrieval accuracy on Day 17 should correlate with residual \(\gamma\)-power at retrieval, not with synaptic density.- Prediction 5: An optimal inter-stimulus pairing interval exists (\(\sim100\text{--}200\) ms), with performance dropping at ISI \(<50\) ms or \(>500\) ms.
Authors: Huang, Kai, Liu, Hongkui, Huang, Ziwei
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7204635105