Role-Reversal Simulation¶
Facilitated simulation — instantiates Other-Agent State Model Calibration
Steps through the situation from the other agent's information, constraints, and incentives — arguing their case as they would — to expose where the actor's model is really just projection.
Role-Reversal Simulation has the actor deliberately occupy the other agent's position and reason forward from their information set — their knowledge, constraints, incentives, and stakes — rather than the actor's own. Where a perspective-taking interview asks the agent and a red team attacks the model with rival explanations, role reversal inhabits the perspective, ideally out loud or in writing so the reasoning is audible. That is both its power and its hazard: because it is imaginative, it is uniquely good at surfacing the actor's own hidden assumptions — but it is only trustworthy when fed the agent's real information instead of the actor's guesses about it.
Example¶
An internal developer-tools team keeps hearing that their setup guide is "obvious," yet new hires keep stalling. In a role-reversal simulation, two engineers are stripped down to only what a brand-new user actually has — no shell aliases, no tribal knowledge, just the literal written steps — and must complete onboarding while narrating every assumption aloud, arguing the newcomer's case: "I don't know what a 'token scope' is; the page says 'paste your token' and I'm holding three of them." Constrained to the newcomer's information set, they hit the exact wall real users hit — a step that silently assumes an environment variable nobody was told to set. The team's model shifts from "users don't read" to "the guide encodes knowledge only we have." The projection became visible the moment the actors were forced to reason from a subtracted information set instead of their own.
How it works¶
- The actor takes the agent's seat — their information, constraints, incentives, and stakes — and pointedly not the actor's own knowledge.
- Reasons or acts forward from that position, out loud or on paper, arguing the agent's case as the agent would put it, not as the actor would caricature it.
- The engine is subtraction — deliberately stripping away what the actor knows but the agent cannot — which is what makes buried assumptions audible.
- Checks the simulated perspective against real evidence, to catch the failure mode where reversal merely re-runs the actor's own assumptions in costume.
Tuning parameters¶
- Information fidelity — how strictly the actor is held to only the agent's real information. Loose reversal drifts back into projection; strict reversal (props, restricted access) reveals more but is harder to stage.
- Evidence grounding — how far the reversal is anchored in real data about the agent versus pure imagination. Imagination scales but recreates the actor's blind spots; grounding costs preparation.
- Adversarial versus empathetic stance — whether the actor argues the agent's best case or their likely one. Best-case surfaces steelman constraints; likely-case predicts behavior.
- Number of seats — reverse into one agent or several in turn (each party to a conflict). More seats expose asymmetries but dilute depth.
- Facilitation — solo, paired, or refereed by a challenger who flags the moment the actor smuggles their own knowledge back in.
When it helps, and when it misleads¶
Its strength is that it is the sharpest tool for surfacing the false-consensus effect — the tendency to overestimate how much others share your knowledge, views, and likely reactions — because inhabiting a constrained perspective makes your own hidden assumptions suddenly audible.[n1] It is fast, needs no access to the agent in the moment, and is uniquely good at finding curse-of-knowledge gaps in explanations, interfaces, and instructions.
Its core weakness is built in: a reversal run from imagination tends to recreate the actor's own assumptions in the other's voice, producing a confident model that is really a self-portrait, with stereotypes filling the gaps where real information is missing. It can also breed a false confidence — "I've walked in their shoes, so I know" — that forecloses actually asking. The discipline is to constrain the reversal to the agent's real information, ground it in evidence from inquiry rather than imagination, and treat its output as a hypothesis to test, not a conclusion to trust.
How it implements the components¶
Role-Reversal Simulation realizes the projection-exposure side of the archetype — it stress-tests the actor's model against the actor's own blind spots:
counterfactual_role_reversal— it is the reversal: reasoning the situation forward from the other agent's seat rather than the actor's.actor_state_separation— the deliberate subtraction of the actor's private knowledge is the engine of the exercise; it forces the what-I-know / what-they-can-know split into the open.projection_and_stereotype_filter— by making assumptions audible, it exposes exactly where the model was projection (the agent imagined as too much like me) rather than an accurate read.
It does not collect the real information the reversal must be grounded in — that comes from Perspective-Taking Interview — nor does it keep a running accuracy record; that is Prediction and Surprise Log. Its adversarial cousin, Counterparty Model Red Team, generates rival models rather than inhabiting one.
Related¶
- Instantiates: Other-Agent State Model Calibration — role reversal is the appraisal's projection check, run by inhabiting the other's position.
- Consumes: Perspective-Taking Interview supplies the agent's real information, so the reversal isn't pure imagination.
- Sibling mechanisms: Perspective-Taking Interview · Counterparty Model Red Team · Empathy Map with Evidence Marks · Belief-Desire-Knowledge Map · False-Belief Check · Active Listening Loop · Interaction After-Action Review · Prediction and Surprise Log · Consent and Privacy Boundary Checklist · Stakeholder Hidden-Constraint Board
Editorial Notes¶
Form Classification¶
Form family: Experiment, Test & Rehearsal
Rationale: Role-Reversal Simulation operates as an active test, trial, simulation, drill, or rehearsal that generates evidence through a deliberate attempt or perturbation because it steps through the situation from the other agent's information, constraints, and incentives — arguing their case as they would — to expose where the actor's model is really just projection.
Independent corroboration: The frozen evidence defines Role-Reversal Simulation as 'Steps through the situation from the other agent's information, constraints, and incentives — arguing their case as they would — to expose where the actor's model is really just projection', so its operative form is Experiment, Test & Rehearsal.
Nearest alternative: Communication, Facilitation & Learning — Role-Reversal Simulation includes features of a designed message, facilitated interaction, ritual, or learning activity that changes shared understanding, but its defining operation is an active test, trial, simulation, drill, or rehearsal that generates evidence through a deliberate attempt or perturbation.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Psychology
Origin pattern: Convergent development
Present-day reach: Universal
Rationale: Enacting another person's constraints and perspective is rooted in psychodrama and perspective-taking psychology.
Related originating lineages:
- Behavioral Economics — Incentive-aware analysis materially sharpens simulation of the other actor's choices.
- Cognitive Science — Cognitive-science research on representation, learning, and recall supplies a parallel or contributing lineage for the mechanism's defining operation: steps through the situation from the other agent's information, constraints, and incentives — arguing their case as they would — to expose where the actor's model is really just….
- Organizational & Management Science — Negotiation and leadership training independently apply role reversal.
Review resolution: Both blind reviewers agree that psychology is the primary historical origin. Explicit reconciliation of alternate origin disagreement, origin mode disagreement, domain reach disagreement, encyclopedia synthesis disagreement starts from reviewer_a’s mechanism-specific evidence: Enacting another person's constraints and perspective is rooted in psychodrama and perspective-taking psychology. Reviewer A proposed alternates=behavioral_economics, organizational_management, origin_mode=convergent, domain_reach=universal, and encyclopedia_synthesis=false; reviewer B proposed alternates=cognitive_science, origin_mode=single_lineage, domain_reach=multi_domain, and encyclopedia_synthesis=true. The final record retains every independently supported alternate from either review (behavioral_economics, organizational_management, cognitive_science) without an arbitrary cap, selects origin_mode=convergent to represent the combined lineage evidence, and keeps domain_reach=universal and encyclopedia_synthesis=false from the more mechanism-specific assessment. Present-day transfer is recorded as reach and is not treated as proof of historical origin.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
The headline benefit — "I walked in their shoes" — is also the trap. Imagined perspective-taking that isn't disciplined by real information can increase confidence without increasing accuracy, and can lean harder on stereotypes precisely when the actor feels most empathetic. Pair it with inquiry; never let it stand alone as the model.
[n1] The false-consensus effect — the tendency to overestimate how widely others share one's own beliefs, knowledge, and likely choices. Role reversal counters it by forcing the actor to reason from the other agent's information rather than assuming it matches their own. ↩