Two Personalities, One Relationship: Structural Geometry of Dyadic Interaction

Executive Summary

• Cross-band Coherence alignment is the single strongest predictor of relationship-level integration, explaining 40% of dyadic Coherence variance (r = 0.63, R² = .395, N = 1,000 dyads). How partners’ integration levels interact across the Relational column matters more than any other single structural feature.
• The interaction tensor captures relationship structure across five genuine dimensions, with channel-level alignment explaining 35% of dyadic Coherence (r = 0.59, R² = .350). Relationships aren’t one-dimensional; collapsing them to a single compatibility score discards a third of the available structural information.
• Reinforcing and catalytic Trap interactions serve fundamentally different functions (t = 43.10, d = 1.928). Mutual amplification of dysfunction is the dominant mode of cross-partner Trap interaction, outnumbering growth-promoting perturbation by nearly two standard deviations.
• Reinforcing Trap accumulation degrades relational integration (r = –.39, R² = .151), accounting for 15% of dyadic Coherence variance. When both partners’ stuck points feed each other, the relationship pays a measurable structural cost.
• Gateway compatibility predicts dyadic Coherence at a medium effect (r = 0.43, R² = .185). Which structurally critical centers align between partners matters far more than how many channels are open or blocked.
• Bond-Capacity convergence operates through five fully independent Domain channels (PCA: 5 effective dimensions). Emotional convergence tells you nothing about relational convergence; each Domain requires separate clinical attention.
• 97% of hypotheses across this family were null (29 of 30). You cannot predict relationship quality from individual personality features alone. The constructs that work (cross-band alignment, tensor structure, interaction types) require both profiles to compute. This boundary is clinically critical.
• Relational Trap count matters more than Trap type: pursue-withdraw, the most clinically prominent pattern, explained less than 1% of collision risk independently. Total Trap load is the reliable signal.
• Enmeshment Basin scores specifically predict identity erosion (r = –.14), targeting the exact centers governing self-other boundaries, while the “healthy relationship” Basin and adversarial Basin showed negligible or null associations with their predicted outcomes.
• Dyadic pattern counts don’t predict relational outcomes: total active pattern count bore zero relationship to dyadic Coherence (r = .02). Patterns describe real configurations but their clinical meaning is configural, not additive.

Research Overview

Dyadic interaction operates at the level of specific centers, specific Gateways, and specific feedback loops that lock two people into patterns neither created alone. This family of 10 computational studies investigated the structural geometry of dyadic interaction from every angle the Icosa model provides: how partners’ integration levels combine, how their channels align and conflict, how their dysfunction patterns interlock, how their integrative Capacities converge or diverge across experiential Domains, and whether the model’s relational constructs predict the outcomes they’re supposed to predict.

The intellectual agenda was a single question examined through 10 lenses: does the relationship have its own structural architecture that’s distinct from either partner’s individual profile? The answer, across 10,000 dyadic profiles and 30 formal hypotheses, is a qualified yes. The constructs that predict relational outcomes are relational; they require both Icosaglyphs to compute and can’t be derived from either partner’s profile alone. Cross-band Coherence alignment, tensor alignment, reinforcing Trap interactions, Gateway compatibility, and bond merge all carry measurable predictive weight. But the qualification matters just as much: 29 of those 30 hypotheses were null or negligible, and the three that produced large or medium effects all share a common property. They capture how two profiles interact structurally, not what either profile contains individually.

That boundary (between individual structure and relational emergence) is the central finding of this family. It constrains what clinicians should and shouldn’t infer from personality assessment in couples work, and it defines where the Icosa dyadic engine adds information that individual profiling can’t provide.

Key Findings

The Relational Column Drives Cross-Band Integration

When partners sit in different Coherence bands (one Steady, one Struggling, or one Thriving and one in Crisis) the clinical question is how those different integration levels combine at the relationship level. The intuitive model says the struggling partner drags the relationship down, or the stronger partner lifts it up, or the gap itself is the problem. The data point somewhere simpler and more useful.

The relational grid weighted mean, the average alignment score across Intimacy (Open × Relational), Attunement (Focus × Relational), Belonging (Bond × Relational), and Voice (Move × Relational), correlated with dyadic Coherence at r = 0.63, explaining 39.5% of the variance. That’s a large effect, and it’s the single strongest predictor of relationship-level integration identified across all 10 studies. It held up despite non-normality in one variable and across the full range of Coherence band pairings in the 1,000-dyad sample.

Two competing models were tested alongside it and came back weak. Relational variance (how unevenly alignment is distributed across those four centers) predicted structural safety at r = –0.09, less than 1% of variance. The “weakest link” model, where the single worst relational center constrains the whole system, predicted transmission efficiency at rₛ = 0.26, about 7%. Both effects were in the predicted direction but neither approached clinical decision-making magnitude. The model behaves additively for cross-band pairings: what matters is the average level of relational alignment, not its shape or its floor.

This simplifies the assessment picture for couples where partners are at different integration levels. The clinician’s first structural question isn’t “where’s the worst mismatch?” or “how uneven is the relational column?” It’s “what’s the overall relational alignment level, and what’s pulling it down?” The Centering Plan can then sequence interventions that lift the relational average rather than drilling into the single weakest center, because the data say the average does the predictive work.

The remaining 60% of dyadic Coherence comes from other structural sources: Emotional Domain contagion (the most contagious channel between partners), the Move → Open cross-partner pathway (the strongest transmission channel), Gateway interdependencies, and the interaction dynamics captured by the tensor and Trap analyses described below. The relational column is the best single window, but it’s one window into a multi-room structure.

The Relationship Has Five Structural Dimensions

If the relational column captures the most visible layer of how a couple functions, the interaction tensor captures the full wiring diagram. The tensor maps transmission, alignment, complementarity, and conflict across all 20 Capacity-Domain channels for each dyadic pair: a channel-by-channel account of how each partner’s personality state influences the other’s at every center.

Mean tensor alignment (the degree to which both partners’ Capacity flows point in compatible directions across all 20 channels) correlated with dyadic Coherence at r = 0.59, explaining 35% of the variance. That’s a large effect from a different structural angle than the relational column finding, confirming that relational integration depends substantially on directional compatibility across the full personality structure, not just the four Relational-Domain centers. Tensor conflict told the inverse story: higher conflict predicted lower transmission efficiency (r = –.30, R² = .090), a medium effect confirming that cross-partner conflict doesn’t just create friction; it measurably degrades the system’s Capacity to relay psychological resources between partners.

Principal component analysis on 10 tensor metrics revealed five effective dimensions explaining 97.8% of variance. The relationship isn’t one-dimensional. It has at least five independent axes of variation, each potentially responsive to different interventions. These dimensions didn’t map neatly onto simple labels (they’re structural properties of the tensor, not yet interpretable as named factors) but their existence confirms what experienced couples therapists sense: a couple can be emotionally intimate and physically disconnected, mentally in sync and relationally adrift. The tensor preserves that granularity instead of averaging it away.

This finding converges with the bond merge analysis, which tested how partners’ Bond-Capacity functioning aligns across the five experiential Domains. Mean bond merge predicted dyadic Coherence at r = .37 (R² = .134, medium effect), and PCA on the five Domain-specific scores found complete independence; each Domain contributed uniquely, with no latent dimensionality. Emotional convergence doesn’t predict Relational convergence. Physical convergence doesn’t predict Spiritual convergence. The co-regulation composite (TMRC) similarly predicted dyadic Coherence at r = .32 (R² = .103, medium effect) with five fully independent co-regulation modes. The convergence across these analyses is striking: whether you look at the tensor, the bond merge, or the co-regulation composite, the relationship decomposes into independent Domain-level channels that don’t transfer gains between them.

For clinical practice, this means that progress in one Domain doesn’t automatically lift another. A couple that achieves emotional breakthroughs in session may terminate prematurely, believing the relationship is repaired, while structural divergence in the Relational or Physical Domain persists beneath the surface. Domain-specific tracking catches this before it leads to relapse.

Mutual Amplification Is the Default; Disruption Isn’t Automatically Therapeutic

Across 1,000 dyadic pairs, the dominant interaction pattern at points of dysfunction was consistent: mutual amplification dominates. Reinforcing Trap interactions (where both partners’ stuck points cycle in the same direction, each sustaining the other’s) outnumbered catalytic interactions by nearly two standard deviations (t = 43.10, d = 1.928). That effect is massive, and it describes a structural default: when two people’s Traps meet, the overwhelming tendency is for each person’s dysfunction to maintain the other’s rather than perturb it toward change.

The cost of that amplification is measurable. Reinforcing Trap count correlated negatively with dyadic Coherence at r = –.39 (R² = .151), a medium effect. Roughly 15% of what determines relational integration traces to mutual amplification of trapped states. This isn’t about the content of arguments or the surface-level dynamic. It’s about how underlying personality structures reinforce each other’s feedback loops at specific centers: both partners’ Emotional Rumination cycling in lockstep, or one partner’s Emotional Explosion feeding the other’s Empathic Overwhelm through the Move → Open channel.

The catalytic finding ran opposite to prediction and carries the most important clinical implication. Catalytic Trap count correlated negatively with growth alignment (rₛ = –.18, R² = .031); more catalytic interactions between partners meant slightly less alignment in their developmental trajectories, not more. Disruption without sequencing creates noise, not growth. A partner whose configuration catalytically perturbs the other’s Identity center while the recipient’s Centering Path prioritizes the Body Gate and Feeling Gate first introduces premature destabilization. The structural data challenges the common therapeutic assumption that partner differences are inherently growth-promoting. Sometimes they are. But the data say unsequenced perturbation is more common than sequenced growth, and the distinction between the two is whether the disruption lands at a structural site the recipient is ready to integrate.

This reframes intervention sequencing for couples. The clinical priority is identifying reinforcing Trap loops between partners first, then targeting shared escape Gateways to break the most mutually sustaining cycles. Catalytic interactions should be assessed against both partners’ Centering Path priorities: encouraged where aligned with sequencing, contained where premature.

Gateway Architecture Constrains What’s Possible

Not all personality centers carry equal structural weight. The nine Gateways (structurally critical centers that unlock or constrain entire regions of the system) serve as escape routes for the 42 individual Traps. The Body Gate (Open × Physical) and Choice Gate (Focus × Mental) each serve 10 Traps; the Voice Gate (Move × Relational) serves none. When two partners’ Gateway states are compared, the alignment of these high-leverage centers predicts relational integration at r = 0.43 (R² = .185, medium effect), while simple counts of open or blocked channels showed negligible or null effects.

This finding is architectural. A couple with three aligned high-leverage Gateways can have far better relational integration than a couple with seven aligned low-leverage ones, because what matters is the cascade effect through each partner’s system, not the raw count. Counting open channels treats the Body Gate and Voice Gate as interchangeable; the structural data says they’re not. Configuration, not quantity, drives the relationship.

The Gateway finding connects directly to the Trap dynamics. When a couple carries multiple reinforcing Trap loops, the escape Gateways for those Traps become shared leverage points. If both partners’ Emotional Rumination routes through the Feeling Gate (Bond × Emotional), and both partners’ Cognitive Paralysis routes through the Body Gate (Open × Physical), the clinician has two shared targets that could break multiple reinforcing loops simultaneously. The Centering Plan sequences these by accessibility (which Gateway is closest to opening) rather than by which Trap appears most dramatic in session.

The enmeshment Basin finding adds specificity to this architectural picture. Enmeshment Basin scores correlated negatively with identity differentiation at r = –.14 (R² = .019, small effect), targeting the exact centers governing self-other boundaries: over-activated Belonging and Embrace alongside suppressed Identity and Discernment. The intervention priority for enmeshed couples isn’t the enmeshment itself; it’s the Identity Gate (Bond × Mental) and Discernment Gate (Focus × Emotional) work that restores each partner’s differentiation. You build the structural floor that lets differentiation hold, rather than trying to pull enmeshed partners apart without giving them the internal scaffolding to stand separately.

Trap Count Matters More Than Trap Identity

Across the relational Trap analyses, a consistent pattern emerged: the total number of active relational Traps predicted dyadic Coherence (r = –.22, R² = .048), while individual Trap types showed negligible independent effects. Pursue-withdraw (the most clinically prominent pattern in couples therapy) accounted for less than 1% of collision risk variance (rₛ = .09, R² = .008). The emotional caretaker Trap showed no variation across resonance coupling levels (F = 1.41, p = .237). Dysregulated lock score predicted collision risk at rₛ = .11 (R² = .012), small but reliable, and operating at the dyadic level rather than within either partner alone.

The clinical implication is direct: count the Traps before sequencing the work. A dyad carrying four simultaneous relational Traps has a measurably different structural prognosis than a dyad carrying one, regardless of which specific Traps those happen to be. Intervention sequencing should prioritize reducing total Trap load through the most structurally accessible escape Gateway rather than targeting whichever Trap appears most symptomatic. Each resolved Trap lightens the structural load on the remaining Gateways through shared dependencies in the Bond and Focus rows.

Boundaries of the Evidence

Twenty-nine of 30 hypotheses across this family were null or negligible, a 97% null rate. This is the most important null finding in the entire Icosa validation program, and it establishes a boundary that should govern how personality assessment is used in couples work.

The nulls cluster in a specific pattern. Individual-level metrics don’t predict relational outcomes. Counting things (open channels, active patterns, blocked pathways) doesn’t capture how relational dynamics work. Simple linear associations between protective factors and safety outcomes don’t hold in randomly generated profiles. Pattern labels don’t predict integration when treated as additive counts. Cascade asymmetry doesn’t predict structural safety. Harmonic locking doesn’t predict resonance coupling. The functional dyad Basin doesn’t independently predict structural safety. Repair Capacity shows only a weak link to structural safety (rₛ = .13, R² = .017). These nulls aren’t failures of the model; they’re the model demonstrating that it doesn’t produce spurious correlations. When a construct shouldn’t predict an outcome, it doesn’t.

The boundary this establishes is clinically critical: don’t use individual profiles to predict relational outcomes. A partner’s Coherence score, Trap configuration, or Gateway status tells you about that person’s individual structure. It doesn’t tell you how the relationship will function, because the relationship’s structural properties emerge from the interaction between two profiles, properties that can’t be computed from either profile alone. The constructs that do predict relational outcomes (cross-band alignment, tensor alignment, reinforcing Trap interactions, Gateway compatibility, bond merge) all require both Icosaglyphs to compute. They’re relational. The 97% null rate confirms that the model distinguishes between individual and relational levels of analysis rather than conflating them.

Several specific nulls carry their own clinical weight. The Formation compatibility study found that total active pattern count bore zero relationship to dyadic Coherence (r = .02, p = .569); pattern labels describe real geometric configurations but counting them discards the configural information that matters. The risk-protection study found that cascade asymmetry showed no association with structural safety (r = –.01, p = .782), likely because structural safety is a threshold phenomenon that barely varies in randomly generated profiles. Protective factors appear to require specific structural conditions to emerge, conditions that real couples may develop through co-regulation over time but that random pairing doesn’t produce. Risk, by contrast, is structurally accessible: dysregulated locking appears readily in random pairings. That asymmetry (risk is easy to produce, protection is hard) may reflect something real about how relationships work.

Clinical Use

The combined findings from this family reshape couples assessment workflow in three specific ways: what to measure at intake, how to sequence intervention, and what to track across treatment.

At intake, both partners complete an individual assessment. Standard tier (32 questions, ~5 minutes each) is sufficient for treatment planning. The Icosa Atlas dyadic engine maps both Icosaglyphs, computes the center-by-center interaction types (Reinforcing, Complementary, Catalytic, or Neutral at each of the 20 Harmony pairs), assigns a Formation Family classification (Resonant, Complementary, Asymmetric, Stagnant, Distressed, Mirrored, Transitional, or Crisis), calculates dyadic Coherence, and generates the couple screening classification (GREEN/YELLOW/RED).

Intervention sequencing follows from the structural map rather than from clinical salience. The findings converge on a clear priority order: identify reinforcing Trap loops between partners first (they account for 15% of Coherence variance), then target shared escape Gateways to break the most mutually sustaining cycles (Gateway compatibility accounts for 18.5%), while monitoring the relational column average as the primary integration metric (40% of variance). The Centering Plan sequences these by structural accessibility (which Gateway is closest to opening) not which Trap appears most dramatic. If the Feeling Gate is Partial while the Belonging Gate is Closed, the work starts at the Feeling Gate even if the Belonging-related pattern is what the couple talks about every session. Each resolved Trap lightens the load on remaining Gateways through shared structural dependencies.

The Domain independence finding (confirmed across bond merge, co-regulation, and tensor analyses) carries a specific clinical caution: gains in one experiential Domain don’t transfer to others. A couple showing emotional breakthroughs needs continued monitoring of Relational, Physical, Mental, and Spiritual convergence. The Timeline lets clinicians track Domain-specific trajectories through incremental assessment updates without full readministration. It displays these trajectories visually, and the therapeutic valley prediction anticipates the temporary dips that often occur when stable (if dysfunctional) configurations are disrupted. This is particularly relevant for enmeshed couples beginning differentiation work, where the relationship can feel worse before the structural reorganization takes hold.

The catalytic finding adds a necessary clinical guardrail. When the dyadic profile shows catalytic interactions at centers that don’t align with either partner’s Centering Path priorities, those sites should be contained rather than activated. Premature catalytic disruption introduces instability without direction. The clinician checks catalytic sites against both partners’ centering priorities and distinguishes between catalytic interactions to encourage (aligned with sequencing) and those to hold in reserve (premature for the current treatment phase).

Applied Example

A couple presents with what both describe as chronic disconnection despite genuine affection. They’ve tried communication workshops and attachment-based therapy. They can name their attachment styles and identify their triggers. The disconnection persists. Both complete the Standard assessment through Icosa Atlas.

The dyadic profile reveals a Stagnant Formation Family with dyadic Coherence of 54 (Struggling band). The relational column tells the first structural story: Intimacy and Attunement show Reinforcing interactions with both partners centered; they can receive and attend to each other. But Belonging is Complementary (Partner A slightly over-activated, Partner B under-activated), and Voice is Neutral (both suppressed, both in a Freezing state on the Move Capacity). The relational weighted mean sits at a moderate level, which the cross-band study says is the primary target for improving dyadic Coherence.

Looking deeper, the tensor reveals high alignment in the Emotional Domain but conflict in the Physical and Relational Domains; they feel each other’s emotions accurately but their bodies are out of sync and their relational negotiation creates friction. Bond merge confirms the picture: Embrace merge is 68 (strong emotional convergence) while Belonging merge is 31 (significant relational divergence). The Domain independence finding tells the clinician that strengthening the already-strong emotional connection won’t fix the relational gap. These are separate rooms.

Three reinforcing Trap interactions cluster in the Emotional and Relational Domains. Both partners show matched Capacity-flow displacements at Passion and Belonging, the centers governing emotional expression and relational integration. The reinforcing Trap analysis says these mutual amplification loops account for a measurable share of the couple’s Coherence deficit. The escape Gateways for these Traps converge on the Feeling Gate (Bond × Emotional, Partial for both partners) and the Belonging Gate (Bond × Relational, Closed for Partner B). The Centering Plan sequences the Feeling Gate first (it’s Partial, meaning there’s existing structure to build on) before tackling the Belonging Gate.

Two catalytic interactions sit at Mental Domain centers. Checking against both partners’ Centering Paths, neither path prioritizes these sites in the first three steps. The clinician holds these in reserve. They represent real growth opportunities, but only after the Emotional and Relational reinforcing loops have been disrupted. Activating them prematurely (exploring identity questions or meaning-making before the emotional infrastructure is in place) would introduce the kind of unsequenced perturbation the catalytic study identified as counterproductive.

As sessions progress, the Timeline tracks the specific centers under intervention. After eight sessions, the updated profile shows the Feeling Gate has moved from Partial to Open in both partners, and one of the three reinforcing Trap interactions has resolved. Belonging merge has shifted from 31 to 39, modest but measurable movement in the right Domain. The relational weighted mean has climbed, and dyadic Coherence has moved from 54 to 59. The therapeutic valley prediction had flagged a dip around session five as the Stagnant Formation began destabilizing, and the clinician had prepared both partners for the temporary increase in tension that accompanied the structural shift. The couple’s Formation Family classification has shifted from Stagnant to Transitional, a transition that carries its own clinical challenges but represents genuine structural movement.

The plain-language summary gives both partners an accessible version of this progress: where their relationship was, what shifted, and what the next phase of work targets. The Clinician Map retains the full structural detail for formulation and documentation. The couple can see why Voice work comes next (the Belonging Gate is now Partial, making it structurally accessible) and why the identity-related catalytic interactions are being held for a later phase. That transparency, grounded in structural data rather than clinical authority alone, builds the kind of collaborative alliance that sustains couples through the harder phases of treatment.

Connections Across the Research

The dyadic family’s central finding (that relational outcomes can’t be predicted from individual profiles alone) converges directly with the Clinical family’s couples-therapy-indicators study, which confirmed this boundary from the clinical side: individual metrics predicted relational outcomes at r = 0.11 and r = 0.07, negligible effects that independently establish the same constraint. The 97% null rate in the dyadic family and the Clinical family’s null results on individual-to-relational prediction form a consistent boundary across different analytical approaches.

The Constructs family’s findings about Trap-Basin relationships (r = 0.39 shared variance at the individual level) don’t directly transfer to the relational level, a distinction the dyadic family’s data makes explicit. Individual Traps and Basins share structural territory within a single Icosaglyph, but dyadic Traps and Basins operate through cross-partner channels that introduce interaction effects absent from individual profiles. The enmeshment Basin’s selective association with identity differentiation (r = –.14) illustrates this: the Basin exists between two profiles, and its predictive specificity (targeting identity-related centers while showing no association with structural safety or collision risk) demonstrates that relational constructs carry their own structural logic rather than simply extending individual-level dynamics to the dyadic context.

Operational Impact

For practices doing couples work, the combined evidence base from this family provides three measurable advantages: faster structural formulation, more precise intervention sequencing, and trackable Domain-specific progress metrics.

The assessment-to-intervention timeline compresses when the structural map is available from intake. Instead of spending four to six sessions mapping the couple’s dynamics through observation and report, the dyadic profile provides the channel-level picture computationally: which Domains align, which conflict, where reinforcing Traps lock, which Gateways are accessible. The Centering Plan sequences interventions based on structural dependency rather than clinical intuition about where to start. For practices billing couples sessions, moving from assessment interpretation to targeted intervention one or two sessions earlier across a caseload of 20 couples is a meaningful operational gain. The evidence base supports specific claims: relational column alignment predicts 40% of dyadic Coherence, Gateway compatibility predicts 18.5%, and reinforcing Trap count predicts 15%. These are structural metrics with known effect sizes that can be documented, communicated to referral sources, and tracked across treatment.

The differentiation angle is concrete. Couples therapists who can show clients their dyadic profile (Formation Family classification, Gateway alignment map, Domain-specific bond merge scores, reinforcing Trap locations with escape Gateway sequencing) are offering structural formulation that goes beyond attachment categories or communication assessments. The plain-language summary provides accessible versions; the Clinician Map retains full structural detail. Multi-reporter capability catches blind spots in how each partner perceives the relationship. The GREEN/YELLOW/RED couple screening provides immediate viability assessment. And because the 97% null rate confirms that the model doesn’t produce spurious correlations (that the constructs which predict relational outcomes are relational rather than artifacts of individual-level features) the clinical reasoning is structurally grounded in ways that build credibility with both clients and referring providers.

Summary

For clinical practices doing couples work, the evidence from this family of studies provides a measurable operational advantage: the ability to formulate relationally from intake, sequence interventions structurally rather than intuitively, and track progress across independent Domain channels with known effect sizes. The relational column weighted mean predicting 40% of Dyadic Coherence, Gateway compatibility predicting 18.5%, reinforcing Trap count predicting 15%: these aren’t theoretical constructs awaiting validation. They’re structural metrics available computationally from Standard-tier assessments, ready to inform treatment planning before the first couples session begins. The dyadic profile shows you which Domains align and which conflict, where reinforcing loops lock both partners into mutual amplification, which Gateways are structurally accessible, and how the relationship’s integration level distributes across experiential channels that don’t transfer gains between them.

The 97% null rate across individual-to-relational predictions establishes the clinical boundary that matters: you cannot use individual profiles to predict relationship outcomes, and the constructs that do predict relational functioning are emergent, properties of the interaction between two structures that can’t be computed from either profile alone. That distinction reshapes couples assessment workflow by clarifying what information individual personality testing provides and what it leaves out. The differentiator for practices adopting this approach is precision: knowing where to intervene based on structural dependencies rather than clinical salience, tracking Domain-specific trajectories that reveal when emotional progress masks relational stagnation, and sequencing Gateway work against both partners’ Centering Paths to distinguish growth-promoting catalytic interactions from premature destabilization.

What becomes possible is couples therapy grounded in the relationship’s actual architecture. Not in attachment categories, not in communication skills alone, but in the channel-level map of how two personality structures transmit, align, conflict, and lock at specific centers with specific escape routes. The plain-language summary provides formulation clients can understand and collaborate with. The Clinician Map retains the structural detail that builds credibility with referral sources and supports clinical documentation. The couple leaves intake knowing their Formation Family classification, their Dyadic Coherence score, which reinforcing Trap loops are sustaining their presenting complaints, and what the sequenced intervention priorities are, not because the clinician spent six sessions mapping dynamics through observation, but because the structural formulation was available computationally from the assessment. That’s the operational shift this research enables, and it’s available now.