Anuma Research · Intelligence Architecture

Transactional–Relational
Intelligence

A research framework for distinguishing intelligence expressed through immediate transactions from intelligence emerging through relationships, memory, context, continuity and adaptation.

Central Principle

Transactions execute actions;
relationships construct intelligence.

01 — Research Question

Many computational and organizational systems are designed primarily around transactions: an input is received, an operation is performed, and an output is produced.

The Transactional–Relational Intelligence framework asks a different question:

What changes when intelligence is understood not only as successful transaction, but as the development of relationships across time?

The framework therefore investigates transaction, relationship, memory and context as distinguishable components of intelligent behaviour.

02 — Transactional Mode

Transactional intelligence concerns exchange, execution and immediate output.

Dimension	Transactional Mode
Primary function	Exchange or execution
Temporal orientation	Immediate result
Context	May be limited to the current interaction
Memory	May be minimal or external to the transaction
Relationship	Not necessarily required for successful execution
Output	Action, exchange or response

Transactional intelligence is not treated as inferior or unnecessary. It is fundamental to many systems. The research distinction concerns what additional properties emerge when transactions occur within persistent relational structures.

03 — Relational Mode

Relational intelligence concerns the capacity of a system to preserve, interpret and adapt to relationships across time.

I_R = f(R, M, C)

R — Relationship The structure connecting entities, states or participants.

M — Memory Information that permits continuity across interactions.

C — Context Conditions that give an interaction meaning or relevance.

I_R — Relational Intelligence A proposed representation of intelligence arising from the interaction of relationship, memory and context.

In this framework, relational intelligence is therefore not simply the ability to complete another transaction. It concerns what the system can develop from the continuity between transactions.

04 — From Interaction to Continuity

Consider two interactions:

T₁ → T₂

In a purely transactional interpretation, the second transaction can be treated independently from the first.

A relational system instead permits:

T₁ → M → C → R → T₂

The first interaction can therefore modify memory, context or relational structure, which subsequently changes the conditions under which the next interaction occurs.

This creates a feedback structure in which interaction can alter the future state of the system.

05 — Memory as Relational Continuity

Within the framework, memory is not considered only as stored information. It can also function as a mechanism of continuity between states and relationships.


          Stateₜ → Interaction → Memoryₜ₊₁ → Future Interaction

This allows the history of interaction to become relevant to future behaviour.

The research implication is that a system may require more than computational capacity to exhibit increasingly contextual behaviour. It may also require structures through which previous relationships influence future states.

06 — Combined Intelligence Model

The framework proposes a combined representation of transactional and relational intelligence:


          I = I_T + λI_R

where:

I_T Transactional intelligence.

I_R Relational intelligence.

λ A model parameter representing the contribution assigned to relational intelligence.

I A proposed combined intelligence representation.

This equation is a modelling hypothesis rather than an established law. Its usefulness depends on whether the quantities can be operationalised and whether the resulting model provides explanatory or predictive value.

07 — Nonlinear Extension

A relational contribution does not necessarily need to be additive. A more general formulation may be represented as:


          I = Φ(I_T, I_R, I_E)

where Φ represents a potentially nonlinear relationship between transactional, relational and existential components.

This formulation allows future experiments to investigate interactions in which the combined system behaviour cannot be adequately represented by a simple sum.

08 — Relational Breakdown

The framework also considers the possibility of relational breakdown: situations in which transactions continue to occur while continuity, context or meaningful relationships become degraded.


          Transaction ↑
            while  
          Relationship ↓

The distinction is important because a system can remain highly active at the transactional level while losing contextual coherence at the relational level.

This creates a research problem rather than a predetermined conclusion: can measures of relational structure identify forms of system degradation that transaction-level metrics fail to capture?

09 — Data Extraction vs. Intelligence Development

The framework distinguishes between extracting information from a system and developing intelligence within a relational system.

Information Extraction	Relational Intelligence Development
Focuses on obtaining information	Focuses on developing contextual understanding
Can be transaction-oriented	Requires continuity across interactions
May treat participants as data sources	Considers participants as components of a relationship
Optimises information acquisition	Investigates adaptation and meaning within context

This distinction forms part of the broader Anuma investigation into how intelligence can be developed across living, technological and social systems without reducing intelligence to data extraction alone.

10 — Experimental Structure

The framework proposes comparing systems with different levels of relational information.

Transactional baseline

A model receives the current state and produces a prediction or action.


          Xₜ → Model → X̂ₜ₊₁

Relational model

A model receives current state together with relational structure, memory and context.


          (Xₜ, Rₜ, Mₜ, Cₜ) → Model → X̂ₜ₊₁

Their predictive performance can then be compared:


          ΔP =
          P_relational
          −
          P_{transactional}

This provides a concrete experimental pathway for determining whether relational variables provide measurable predictive information beyond a transactional baseline.

11 — Information-Theoretic Formulation

One proposed test asks whether relationships and memory contain information about a future state after conditioning on the current state.


          I(Xₜ₊₁ ; Rₜ, Mₜ | Xₜ) > 0

If the conditional mutual information is greater than zero under an appropriately designed experiment, relational and memory variables contain information about the future state that is not completely represented by the current state alone.

The statistical interpretation, controls and experimental design would need to be specified for any empirical test.

12 — From Individual Systems to Living Systems

The Transactional–Relational framework is designed to extend beyond individual computational systems.

A network may be represented as:


          G = (V, E, W)

where nodes, relationships and relationship weights form an evolving structure.

The broader living-intelligence formulation is:


          I_system =
          ΣI_i
          +
          I_relations
          +
          I_emergence

This creates a research bridge between relational intelligence, network science, distributed cognition, complex adaptive systems and multi-agent intelligence.

13 — Relationship to Resonance Transformation Theory

Transactional–Relational Intelligence provides the relational foundation for Resonance Transformation Theory.


          Context → Relationship → Resonance → Transformation

RTT introduces resonance as a proposed transformation variable, while the Transactional–Relational framework establishes why relationships, memory and context may need to be represented alongside immediate transactions.

Together, the frameworks form part of the larger Anuma research architecture.

14 — Scientific Positioning

Established Foundations

The framework draws upon established areas including information theory, network science, dynamical systems and complex systems.

Proposed Framework

Transactional–Relational Intelligence proposes a conceptual and mathematical distinction between transaction and relationship.

Testable Questions

The framework generates measurable questions concerning prediction, memory, relational structure and system transformation.

The framework should therefore be read as a research programme: its proposed mathematical relationships require operational definitions, experimental design, reproducible measurements and comparison with appropriate baselines.

15 — Anuma Research Core

Transactional–Relational Intelligence is one of the foundational research domains within the Anuma Research Core.


          Information → Intelligence → Meaning → Coordination

The research programme investigates how these transitions may emerge when systems are understood as relational, contextual and evolving structures.

← Return to ANUMA Research Core