Glossary

Canonical notation for all BMC documents. When a document conflicts with this glossary, the glossary is authoritative.


Core Model: BMC = (G, M, I, S)

BMC
Biomemetic Complex. The full system: $BMC = (G, M, I, S)$. A universal model, fractally applicable at any scale — from memes within a mind to cultures within civilization.
G
Genetic layer. Fixed programs: Panksepp's 7 affective systems. Vector $T = (T_{SEEK}, T_{FEAR}, \ldots, T_{PLAY})$. Substrate: brainstem, hypothalamus, amygdala.
M
Memetic layer. Network of acquired knowledge (memeplex). Physically = cell assemblies in the cortex. Dynamic, heavy-tailed topology.
I
Interface / Immune filter. Mechanisms of G↔M interaction: Redirection, Suppression, Interpretation. Mediates conflict between drives and memes; filters incompatible information at 4 levels. Substrate: ACC, insula, OFC.
S
Substrate. Dual nature: (1) physical medium (neurons, neurotransmitters); (2) sensory architecture (input channels). Brain, silicon, or any system supporting replicator dynamics.

Key Mechanisms and Metrics

SMC
Self-Model Cluster. A subgraph of the memeplex that models the system itself: $SMC = \{m \in M : target(m) \in M \cup G \cup I\}$. Three recursion levels: L0 (bodily), L1 (narrative self), L2 (metacognition). Substrate: mPFC + TPJ + PCC.
SIT
Structural Incompleteness Tension. Tension arising from structural gaps in the memeplex. When $SIT > \theta$, triggers SEEKING. Subjectively experienced as curiosity, anxiety, or the feeling of "unfinished business."
CL
Consciousness Level. $CL(t) = \sigma_{SW}(t) \cdot A_{SMC}(t) \cdot f(Balance(t))$. A single scalar metric capturing the degree of consciousness at time $t$.
SMR
Shared Memplex Repository. The collective knowledge store of a population of agents. Fractal: $SMR = (G_{SMR}, M_{SMR}, I_{SMR}, S_{SMR})$. Cultural ratchet mechanism.
LP
Learning Progress. $LP(C, t) = \frac{d}{dt} closure(C, t)$. LP > 0 = productive reflection; LP ≈ 0 = rumination.
PE
Prediction Error. Sharp SIT increase when incoming information diverges from expectation.
Mod(t)
Modulation vector. 8-component global state: $(\lambda_{lr}, \theta_{act}, \lambda_{speed}, \lambda_{plast}, \lambda_{inh}, \lambda_{soc}, \lambda_{noise}, \lambda_{diff})$. Maps to neurotransmitter systems.
PA
Preferential Attachment. $\Pi(k_i) = k_i / \sum_j k_j$. Network growth mechanism: popular nodes attract more connections (the rich get richer).
RIF
Retrieval-Induced Forgetting. Retrieving one meme suppresses competitors. $RIF \propto 1/C_E$ (hubs are protected).
SET
Structural Equivalence Test. Formal procedure for determining whether a property transfers across scales. Three zones: green (unconditional), yellow (with correction), red (impossible).
R_expr
Replication Expression pressure. $R_{expr} = a \cdot F \cdot rel \cdot (1 + \alpha \cdot C_E)$. M-driven pressure for a meme to be expressed outward. Not a genetic drive but an emergent property of memetic replication.

Network Metrics

σ_SW
Small-Worldness. $\sigma_{SW} = (C/C_{rand}) / (L/L_{rand})$. Healthy networks ≫ 1. Do not confuse with $\sigma$ (branching ratio), which is a different quantity.
Q
Modularity. How well the network partitions into dense clusters. High Q = rigidity; healthy range involves a balance between integration and differentiation.
C_E
Eigenvector Centrality. Measures the "strength" of a node: not just how many connections it has, but whether those connections are themselves well-connected.
C_B
Betweenness Centrality. Identifies bridge nodes that connect otherwise separate clusters.
κ
Consolidation level. $\kappa \in \{0, 1, 2\}$: 0 = sensory buffer, 1 = short-term memory, 2 = long-term memory.
ψ
Synaptic trace. Activity-silent working memory: a hidden variable enabling latent WM. $\psi \in [0,1]$.
D_eff
Effective Directionality. Measures the communicative asymmetry of the graph.
N_bid
Bidirectional connections. Fractal invariant: $N_{bid}(L) \approx bandwidth(L) - 1$.

Memory and Consolidation

WM
Working Memory. Active WM (~3–4 pointers in focus) + Latent WM ($\psi > \theta_\psi$) ≈ 7 ± 2.
STM
Short-Term Memory. $\kappa = 1$. High activation, low fidelity, not yet consolidated.
LTM
Long-Term Memory. $\kappa = 2$. Consolidated memes with strong connections.
LTP / LTD
Long-Term Potentiation / Long-Term Depression. Strengthening and weakening of synaptic connections. LTP has early (E-LTP, CaMKII) and late (L-LTP, CREB) phases.
SWR
Sharp-Wave Ripple. Hippocampal reactivation mechanism; tags experiences for sleep consolidation.
SHY
Synaptic Homeostasis Hypothesis (Tononi & Cirelli). Sleep = global synaptic downscaling.

Sleep Phases

PhaseFunction
DECOMPOSEDecomposition of episodes into components
CONNECTBinding components to existing structures
BLENDStochastic recombination (REM)
PRUNERemoval of weak connections
STRENGTHENReinforcement of strong connections
REPLAYReactivation of tagged episodes

G-Programs (Panksepp’s 7 Systems)

SystemFunctionValenceNeurotransmitterWM capture
SEEKINGExploration, curiosity+DA0 (directs WM)
FEARThreat avoidanceNE, cortisol1.0 (full capture)
RAGEFrustration, boundary defenseNE + Adrenaline0.8
LUSTGoal pursuit+DA + OXT + VP0.3
CARENurturance, attachment+OXT0.2
PANIC/GRIEFSeparation distressCRF, glutamate0.7
PLAYSocial learning+Endorphins, endocannabinoids0 (releases WM)

DISGUST is not an 8th Panksepp system. It is an I-layer mechanism (immune filter).


Three Computational Engines

EngineMechanismWhat it determinesSubstrate
Graph EngineSynaptic transmissionWHAT is active: activations, edges, WM competitionAP → NT → PSP
Modulation EngineNeuromodulationHOW the graph works: speed, plasticity, noiseDA, 5-HT, NE, ACh
Diffusion EngineVolume transmissionBACKGROUND: priming, warming of semantically close memesNT spillover

Neuroanatomy

AbbreviationFull nameRole in BMC
DMNDefault Mode NetworkSubstrate for SIT scanning and reflection
CENCentral Executive NetworkTask-oriented processing
ACCAnterior Cingulate CortexI-layer: conflict detection
mPFCmedial Prefrontal CortexPart of SMC and DMN
dlPFCdorsolateral Prefrontal CortexWM control
OFCOrbitofrontal CortexI-layer: reward and context integration
TPJTemporoparietal JunctionPart of SMC: Theory of Mind
PCCPosterior Cingulate CortexPart of SMC and DMN
VTAVentral Tegmental AreaDopaminergic region (SEEKING)
PAGPeriaqueductal GraySubstrate for G-programs
DMS / DLSDorsomedial / Dorsolateral StriatumGoal-directed vs automatic behavior

Neurotransmitters

AbbreviationSubstanceRole in Modulation Engine
DADopamine$\lambda_{lr}$ (learning rate), $\lambda_{noise}$
5-HTSerotonin$\theta_{act}$ (activation threshold), $\lambda_{inh}$
NENorepinephrine$\lambda_{speed}$, arousal
AChAcetylcholine$\lambda_{plast}$ (plasticity)
OXTOxytocin$\lambda_{soc}$ (social bonding)
GABAGamma-Aminobutyric AcidInhibitory: lateral inhibition, WM competition

TheoryAuthorLimiting case in BMC
IITTononiBMC at $M = \emptyset$; $\sigma_{SW}$ metric
GNWDehaene & ChangeuxWM broadcasting mechanism
ASTGrazianoSMC as attention schema
HOTRosenthal$SMC^{(2)}$ as higher-order representation
RPTLammeSpreading activation + feedback loops
FEPFristonSIT + G-homeostasis = active inference
PPClark, HohwySIT + G-homeostasis (paired with FEP)

Each theory retains independent value for its own formalism; BMC recovers it as a limiting case under stated restrictions — it does not replace it.


Measurement Proxies

AbbreviationFull nameBMC mapping
PCIPerturbational Complexity IndexProxy for $\sigma_{SW}$ in CL metric (TMS-EEG)
DCMDynamic Causal ModelingProxy for $f(Balance)$ in CL metric
HEPHeartbeat Evoked PotentialProxy for $I_{intero}$ (interoceptive integration)

Cross-Pillar References

AbbreviationFull nameScope
EMTExtended Meme TheoryConceptual core: memes, SMC, dual replicator
BMBiomemeticsNeurobiological grounding
NMNetwork MemeticsMathematical formalization, metrics
AGI_FAGI FoundationsEngineering blueprint
SMSwarm MemeticsFractal scaling

Notation Conventions

  1. First mention: always expanded — “SMC (Self-Model Cluster)”
  2. In formulas: abbreviation only — $A_{SMC}$, $\sigma_{SW}$, $SIT(C)$
  3. PANIC/GRIEF: slash in text, underscore in code (PANIC_GRIEF)
  4. σ: always specify context — $\sigma_{SW}$ (small-worldness) vs $\sigma$ (branching ratio)
  5. λ: always with subscript — $\lambda_{plast}$, $\lambda_{lr}$, $\lambda_{diff}$. Bare λ is ambiguous (>8 different parameters)
  6. S: in BMC tuple = Substrate. In sensors = $S_{spatial}$, $S_{resource}$, etc.