J.Konstapel,Leiden,20-5-2026.
For more than half a century, personalisation systems have promised to treat each person as an individual. In practice, they have delivered something far more modest: assignment to a category.
Whether the Myers-Briggs Type Indicator, the Big Five, the Enneagram, or Holland’s RIASEC typology, every major system maps the infinite diversity of human personality onto a finite set of boxes. A person becomes an INTJ, a Type Five, a High C — and then receives advice, career guidance, and developmental suggestions calibrated to the average of that box.
This has real virtues. It is computationally tractable, easy to communicate, and actionable. But it has a fundamental limitation: it confuses the map for the territory. The space of human individuality is not finite. It is not sixteen types, not five factors, not nine points. The combinatorial explosion that results from crossing even a modest number of continuous dimensions produces a space so vast that no two people occupy the same position.
The Narrative Signature Engine (NSE) is the first principled departure from this finite-typology architecture. Unlike all previous systems, the NSE is not based on theoretical postulates, factor analyses of small samples, or the intuitions of a founder. It is based on 50 years of empirical occupational data — the U.S. Census Bureau (1960–2010) and the complete O*NET database — analysed through a novel emergence engine that extracts the underlying structure of human expectations and failures across all occupations.
The claim is direct and testable: the NSE generates personal development trajectories that are truly unique rather than averaged across type clusters.
Part One: The Empirical Foundation
The Data
The NSE rests on two complementary datasets:
U.S. Census Bureau data (1960–2010). Fifty years of decennial census data capturing occupational distribution, income patterns, educational attainment, and demographic characteristics of the American workforce across half a century of economic transformation.
The O*NET database. The complete Occupational Information Network, containing detailed characteristics of every major occupation in the U.S. economy — required skills, abilities, work activities, work styles, interests, knowledge domains, and contextual variables. Together, these comprise hundreds of thousands of data points spanning the full spectrum of human work.
The Emergence Engine
The emergence engine is the analytical core of the NSE. It processes the Census and O*NET data to extract the latent structure of expectations and failures across occupations. Specifically, it identifies three things:
- Expectation patterns per occupation — what a person in a given occupation is expected to know, do, and be: the implicit and explicit scripts that define successful performance.
- Failure classes per occupation — the characteristic ways in which people fail to meet those expectations. These are not random errors but structured patterns of difficulty specific to the cognitive and relational demands of each occupation.
- Transitions and trajectories — how expectation patterns and failure classes change as a person moves between occupations, gains experience, or shifts domains.
The emergence engine does not assume any pre-existing typology. It allows the structure to emerge from the data. What emerged was unexpected: a four-level structure that corresponds precisely to the Cayley-Dickson chain of normed division algebras — the mathematical sequence ℝ → ℂ → ℍ → 𝕆 (reals, complex numbers, quaternions, octonions).
This is not a theoretical postulate. It is an empirical finding.
Part Two: Why Four Levels? The Mathematics of the Result
The emergence engine discovered that occupational expectations and failures organise themselves into exactly four distinct levels. The question then becomes: why four? Why not three, or five, or seventeen?
The answer lies in a deep theorem of mathematics. The Hurwitz theorem (1898), given its definitive topological form by Adams (1960), establishes that there exist exactly four normed division algebras over the real numbers: ℝ, ℂ, ℍ, and 𝕆. A normed division algebra satisfies three conditions that turn out to characterise coherent expectation structures precisely:
- Every element has an inverse — expectations can be revised.
- The norm is multiplicative — combinations of expectations preserve meaning.
- There are no zero-divisors — two meaningful expectations cannot cancel to nothing.
The Cayley-Dickson construction generates the chain ℝ → ℂ → ℍ → 𝕆, doubling dimension at each step while sacrificing one algebraic property:
- ℂ loses order — you can no longer say one element is definitively “greater” than another.
- ℍ loses commutativity — the order of operations matters: i × j ≠ j × i.
- 𝕆 loses associativity — context matters: (a × b) × c ≠ a × (b × c).
The emergence engine found that occupational failure classes map directly onto these algebraic losses:
| Level | Property Lost | Occupational Failure Class |
|---|---|---|
| ℝ | None | Precision failure — cannot meet formal, rule-based expectations |
| ℂ | Total order | Transformation failure — cannot see patterns or adapt to change |
| ℍ | Commutativity | Sequence failure — cannot handle situations where timing and order matter |
| 𝕆 | Associativity | Synthesis failure — the person’s synthesis consistently outpaces what the environment can absorb |
The 𝕆 failure class deserves special attention. It is structurally invisible to every existing personalisation system — no typology has a category for “person whose characteristic difficulty is that their synthesis is too far ahead of their environment.” Yet the emergence engine found this failure class repeatedly in the data, particularly in creative, strategic, and cross-disciplinary roles. The NSE not only identifies this failure mode; it prescribes the class of challenge that converts it from stagnant repetition into productive learning.
Part Three: The Personal Blueprint
The Personal Blueprint is the complete structural coordinate the NSE computes for each person. It is derived from the emergence engine’s analysis, combined with the person’s birth data — date, time, and place — which provides the initial condition for their developmental trajectory.
The structural coordinate is a five-component tuple:
𝒞 = (q_PoC, ℓ_CD, r_RIASEC, e_Shen, φ_HD)
- q_PoC — a unit quaternion on S³ encoding the person’s Paths of Change orientation across four cognitive modes: Blue (analytical), Red (practical), Green (relational), Yellow (synthetic).
- ℓ_CD — the Cayley-Dickson failure class {ℝ, ℂ, ℍ, 𝕆}: the algebraic level at which this person is most likely to experience characteristic difficulty.
- r_RIASEC — the occupational interest vector, derived from the empirical mapping between the four-level structure and Holland’s well-validated RIASEC typology.
- e_Shen — the primary Wu Xing element {Wood, Fire, Earth, Metal, Water}: a phase factor modulating the expression of the four-level structure.
- φ_HD — the full Human Design configuration. The NSE is not derived from Human Design; rather, Human Design independently arrived at patterns that are projections of the same underlying algebraic structure. The configuration is included because it facilitates communication with audiences familiar with that language.
The Personal Blueprint is not a theoretical construct. It is an empirically derived coordinate that locates the person within the space of possible developmental trajectories, as discovered from 50 years of occupational data. No two people occupy the same point.
Part Four: Structure and State — The Two-Layer Architecture
The NSE’s most important architectural innovation is the strict separation between a permanent structural layer (the Personal Blueprint) and a dynamic state layer.
The Structural Layer — Invariant
Computed once from birth data and never updated. It is the bedrock: the deep pattern that does not change across the lifespan.
It contains:
- The Paths of Change quaternion
- The Cayley-Dickson failure class
- The structural Fiske vector per domain (work, learning, politics, family)
The State Layer — Dynamic
Updated continuously from observed behaviour. It contains:
- The current Fiske vector per domain — how the person actually organises social relations right now
- The coherence score — the angular similarity between the structural vector and the current state vector
State updates follow the Free Energy Principle (Friston, 2010): the system logs each behavioural observation and updates the state vector via a precision-weighted moving average. The person is modelled as a system that continuously minimises long-term surprise by maintaining and revising a generative model of its environment.
The Coherence Score
$$\text{coherence} = \frac{\mathbf{f}{\text{current}} \cdot \mathbf{f}{\text{structural}}}{|\mathbf{f}{\text{current}}||\mathbf{f}{\text{structural}}|}$$
- Near 1.0 — current behaviour is well-aligned with structural disposition. The person is living in accordance with their Personal Blueprint.
- Below 0.7 — accumulated drift. The person is consistently acting in a mode that is not their natural one, typically under external pressure.
- Below 0.5 — explicit intervention warranted. The Coherence Mirror surface in the SWARP platform displays the drift and invites recalibration.
This two-layer separation resolves a contradiction that has plagued personalisation systems for decades. If a profile is stable, it cannot reflect real developmental change. If it is responsive, it becomes a mirror of recent behaviour rather than a guide to deeper pattern. The NSE avoids this entirely: the structural layer is invariant, the state layer drifts, and the drift itself is informative.
Part Five: The Challenge Table — Engineering Productive Failure
The most actionable component of the NSE is the challenge table. For each Cayley-Dickson failure class and each domain, it specifies the structural features of the productive next challenge.
The logic follows directly from the failure class:
| CD Level | Failure Class | Challenge Structure |
|---|---|---|
| ℝ | Precision failure | Present a situation demanding precision without supplying it |
| ℂ | Transformation failure | Force cross-domain pattern recognition |
| ℍ | Sequence failure | Disrupt the person’s habitual sequence |
| 𝕆 | Synthesis failure | Seek an environment already at the right level — rather than pushing the existing environment upward |
This is the architecturally critical distinction between the NSE and conventional recommender systems. A recommender surfaces what the user already tends toward, optimising for engagement by increasing similarity between profile and content. The NSE surfaces what productively violates the user’s existing scripts — optimising for growth by targeting the specific failure class that triggers learning rather than flight.
Challenge selection is grounded in Case-Based Reasoning (Schank, 1982): a productive challenge is one whose structural features are close enough to past experience to be recognisable as a problem, yet distant enough from past resolutions to require genuine script revision.
Part Six: A Worked Example
Birth data: 22 April 1951, 01:02, Leiden, Netherlands.
The Personal Blueprint
- Paths of Change quaternion (normalised): Blue 0.24 · Red 0.08 · Green 0.40 · Yellow 0.88
- Dominant colour: Yellow → Cayley-Dickson level 𝕆 (synthesis failure class)
- RIASEC dominant triad: Investigative–Artistic–Social
- Lifecycle phase: Age 74, Role Model — the phase of overview and transmission
Domain Projections
Work domain Structural Fiske vector: CS 0.44 · AR 0.10 · EM 0.30 · MP 0.16 Dominant mode: Communal Sharing — shared mission, not institutional position. Challenge: “Find two people who already understand your synthesis; build with them, not alone.”
Learning domain Structural Fiske vector: CS 0.28 · AR 0.10 · EM 0.42 · MP 0.20 Dominant mode: Equality Matching — peer exchange, not instruction. Challenge: “Write your synthesis for someone who does not share your framework.”
Politics domain Structural Fiske vector: CS 0.49 · AR 0.10 · EM 0.35 · MP 0.06 Dominant mode: Communal Sharing — values-based alignment. Challenge: “Find one politician or official who operates from your values; make contact.”
The Generated Narrative Signature
You are someone who sees systems that others have not yet seen, and who works through trust and direct connection — not through position or institutional power. You learn most through peer exchange with people who are equally far along or further, not in classroom settings where knowledge flows in one direction. Your recurring pattern is to offer a synthesis that the surrounding system cannot yet absorb — not because the synthesis is wrong, but because it is early. After a lifetime of experience, you now work from overview: no longer in the arena, but as someone who sees what is actually at stake and transmits that understanding to those who are ready to receive it.
Next step: Write one page this week for someone who does not share your framework — not to persuade them, but to test whether you can say it in their language.
Coherence Tracking
If the user subsequently engages primarily with structured analytical content (Blue-dominant) and avoids collaborative activities, the state vector drifts. The coherence score falls to 0.71. The Coherence Mirror flags mild drift:
You are currently moving in a more structured direction than your natural pattern. This is common under external pressure. Your foundational pattern is 𝕆-level synthesis. The challenge appropriate to your structure is: seek the people who already understand your synthesis.
Part Seven: What Makes the NSE Different
The NSE differs from existing personalisation systems in five respects that are architectural rather than merely parametric.
1. Empirically derived, not theoretically assumed. The four-level structure was extracted from 50 years of occupational data — not assumed from a theoretical framework and then fitted to data.
2. Structurally permanent. The Personal Blueprint does not drift with platform engagement or recent questionnaire responses. It provides a stable reference point that systems chasing recent behaviour cannot offer.
3. Failure-class specific. The NSE is the first system to specify not just who a person is, but what class of cognitive failure they are most likely to experience and what structural features a challenge must have to convert that failure into productive learning. This is directly actionable in ways that type descriptions are not.
4. Coherence as a measurable quantity. The coherence score operationalises a distinction that every practising coach and therapist knows to be real but that no existing system has formalised: the difference between the person’s deep structure and their current state. A person who has been living out of alignment for years is in a different situation from a person facing a novel challenge from a position of alignment. The NSE distinguishes these and responds differently to each.
5. Continuous, not categorical. The coordinate is a point in a continuous space. Two people with the same dominant colour but different secondary colours, or different lifecycle phases, receive different coordinates and different narratives. Personalisation here is genuine: calibrated to the individual point, not to the centre of a cluster.
Conclusion
The Narrative Signature Engine represents a departure from the finite-typology approach that has dominated personalisation for fifty years. Its four-level algebraic structure — corresponding to the Cayley-Dickson chain ℝ → ℂ → ℍ → 𝕆 — is not a theoretical postulate. It is an empirical discovery about the deep structure of human work and development, extracted from U.S. Census Bureau data (1960–2010) and the O*NET database.
The convergences with Paths of Change, Fiske’s relational models, and Human Design are not assumptions built into the model. They are independent traditions and taxonomies that arrived at the same structure through different means — which is precisely what one would expect if the structure is real.
The two-layer architecture — permanent Personal Blueprint, dynamic coherence score — resolves the stability-responsiveness contradiction that has plagued personalisation since its inception. The challenge table converts the framework from description into prescription.
The NSE does not claim to be a complete theory of human individuality. It claims to be a better computational architecture for representing and acting on individual difference than the finite-typology approaches currently in use. The SWARP platform is now instrumented to answer the empirical question: whether it produces better developmental outcomes. That question is answerable, and the answer is being collected.
Annotated References
Empirical Foundation
U.S. Census Bureau. (1960–2010). Decennial Census of Population and Housing. The primary empirical foundation of the NSE. These decennial censuses provide the longitudinal occupational data — spanning 50 years — from which the emergence engine extracted the underlying structure of expectations and failures. Publicly available through the Census Bureau’s data portal.
National Center for O*NET Development. (2024). O*NET Database. U.S. Department of Labor. The complete Occupational Information Network database. Every major occupation in the U.S. economy is characterised across skills, abilities, work activities, work styles, interests, knowledge domains, and context. Combined with the Census data, this is the empirical substrate from which the NSE’s structure was derived. Freely accessible at onetonline.org.
The Mathematics of Four
Baez, J. C. (2002). The octonions. Bulletin of the American Mathematical Society, 39(2), 145–205. The most accessible survey of the algebraic structures underlying the NSE’s four-level classification. Sections 1–2 cover the Cayley-Dickson construction and the properties of ℝ, ℂ, ℍ, and 𝕆. Open access. Start here for understanding why exactly four levels emerge from the mathematics.
Adams, J. F. (1960). On the non-existence of elements of Hopf invariant one. Annals of Mathematics, 72(1), 20–104. The definitive topological proof that normed division algebras exist only in dimensions 1, 2, 4, and 8. Technically demanding; the non-specialist should begin with Baez (2002).
Cognitive Dynamics and Learning
Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138. The canonical short statement of the Free Energy Principle, which governs the NSE’s state layer. The key claim — that agents minimise long-run surprise by continuously updating a generative model of their environment — is developed on p. 129. Readable by a non-specialist with patience.
Parr, T., Pezzulo, G., & Friston, K. (2022). Active Inference: The Free Energy Principle in Mind, Brain, and Behavior. MIT Press. The accessible book-length development of the FEP. Chapters 1–4 are conceptual and require no advanced mathematics. Open access. Recommended for understanding how the NSE’s state update rule works in practice.
Schank, R. C. (1982). Dynamic Memory: A Theory of Reminding and Learning in Computers and People. Cambridge University Press. The foundational text for the NSE’s challenge-selection logic. Schank’s central insight — that learning is triggered by expectation failure, not by repetition — is the direct basis for why the challenge table targets productive violations rather than familiar content.
Relational Structure
Fiske, A. P. (1992). The four elementary forms of sociality: Framework for a unified theory of social relations. Psychological Review, 99(4), 689–723. The foundational paper for Relational Models Theory. The four modes — Communal Sharing (CS), Authority Ranking (AR), Equality Matching (EM), and Market Pricing (MP) — provide the NSE’s primary descriptive vocabulary for both the structural and state layers. Essential reading for understanding what the Fiske vector actually measures.
Convergent Validations
McWhinney, W. (1997). Paths of Change: Strategic Choices for Organizations and Society. Sage. The empirical origin of the four-orientation taxonomy (Unitary, Sensory, Social, Mythic) that independently converged on the same structure the emergence engine discovered from Census and O*NET data. The convergence is not assumed; it was found.
Human Design literature (various sources). Not a foundation of the NSE, but a convergent validation. Human Design independently arrived at patterns that are projections of the same underlying algebraic structure. The NSE references Human Design to facilitate communication with audiences familiar with that vocabulary — not as a source of its architecture.
Hamacher, D. W. (2022). The First Astronomers. Allen & Unwin. The best single-volume introduction to Aboriginal and Torres Strait Islander astronomy and knowledge systems. Demonstrates that the same coherence structures the NSE derived from 50 years of occupational data have been encoded in Indigenous knowledge systems for 65,000 years. Recommended for understanding the deep-time convergent validation.
Occupational Psychology
Holland, J. L. (1997). Making Vocational Choices (3rd ed.). Psychological Assessment Resources. The standard reference for RIASEC (Realistic, Investigative, Artistic, Social, Enterprising, Conventional) — the best-validated occupational interest taxonomy. The NSE incorporates RIASEC through the empirical mapping between Holland’s typology and the four-level algebraic structure discovered from the data.
Coherence Intelligence Framework
U.S. House Oversight Committee. (2023). Hearing: Unidentified Anomalous Phenomena — Exposing the Truth. 118th Congress. Official transcript of sworn congressional testimony on non-human phenomena. Provided as institutional context for the broader Coherence Intelligence Framework within which the NSE sits — specifically the hypothesis that coherence structures are not exclusively biological in origin.
Suggested Reading Order
For a reader new to the NSE’s foundations:
- This essay — architecture and empirical foundation
- Baez (2002) — why exactly four levels
- Fiske (1992) — the relational vocabulary
- Parr, Pezzulo & Friston (2022), chapters 1–4 — the dynamic state layer
- Hamacher (2022) — deep-time convergent validation
The Census and O*NET data are the empirical bedrock; they are cited for transparency but are not required reading for understanding the NSE’s architecture.