Adaptimist Lab

Conversationalist

A branching rehearsal tool for difficult conversations. Set the context, try a move, inspect the coaching signals, and see how the exchange changes depending on what you say next.

Interactive conversation rehearsal

This lab is for rehearsal, not prediction. It simplifies complex interpersonal dynamics and should not be treated as a guarantee of real-world conversational outcomes.

Conversation Lab

Conversationalist

Orientation

Conversationalist is designed for moments when you need to practise a hard conversation before having it for real. You can set the stakes, relationship, calibration, and counterpart stance, then step through the exchange one move at a time.

Use it as a rehearsal space, not a verdict engine. The goal is to notice how different choices shift agency, psychological safety, alliance, and escalation risk, and to discover lines that feel more workable before the real conversation begins.

Set the situation

Add just enough context to make the rehearsal believable.

Try a line

Send a move, inspect the shift, then reveal their reply.

Read the shift

Use the right-hand panels to see what changed after you spoke, what happened after they responded, and which branch is worth trying next.

Step 1

Set up the conversation

Start with the example or swap in your own situation. You do not need to fill every box perfectly.

Fastest way in

Use the example to see how the rehearsal works, or clear it and sketch your own situation.

The three most important things are: what you want, who this is with, and enough backstory to make the moment believable.

Loaded example

A coworker keeps dropping last-minute requests near the end of the day. You want to stay collaborative without silently accepting the pattern.

Minimum setup

Objective

Relationship

Counterpart stance

Backstory

Fine-tune the rehearsal

Calibration

Power

Medium

Stakes

Style mode

You can start as-is, then replace the example later with your own situation.

Step 2

Try a line

Click a node on the right to jump in time. Send tests your line. Release reveals their reply.

Start a session, then type the first line you want to try.

What Happens Next

Once you start, the lab will do three things for you

You try a line

Type what you want to say as if the conversation were already happening.

The app scores the shift

You’ll see how your move affects safety, alliance, escalation, and other signals before their reply appears.

You compare branches

Release their reply, then branch from any moment to test a better version.

Read the shift

What your move changes

Two snapshots per node: after you speak, then after they respond.

After You Spoke

Coach (after you)

—

After They Responded

Coach (after them)

—

Try this instead

Choose a released node to see an alternate line worth testing.

Signals

User signals

Counterpart signals

Step 3

Compare branches

Hollow node = evaluated (reply hidden). Solid node = released.

👉 Click a node to continue from that moment. Then type your next line on the left.

Selected node: —

Working…

Warming up the simulation…

What you may have noticed

Look back not only at the wording that “worked,” but at the tradeoffs it created. A move can increase clarity while reducing warmth. Another can preserve the relationship while weakening the boundary. The value of the lab is in making those tensions visible.

If a branch surprises you, that is useful information. Often the rehearsal becomes most valuable at the point where your first instinct stops feeling obviously sufficient.

How this works

This lab is a branching conversation rehearsal. You begin by setting a simple context for the exchange, such as the objective, relationship, power dynamic, stakes, and counterpart stance. Then you type a line, send it for evaluation, and inspect two snapshots: what the move seems to do immediately after you speak, and what happens after the simulated counterpart responds.

What to notice as you use it

As you move through the branches, notice how the same goal can be pursued in very different ways. Some lines raise agency but lower warmth. Some reduce escalation risk while also making the point blurrier. Others preserve the relationship but fail to move the conversation where you need it to go.

Notice which moves change the tone of the exchange most dramatically.
Notice which tradeoffs feel acceptable to you and which do not.
Notice whether your strongest lines are direct, careful, warm, boundaried, or unexpectedly simple.

Why the experience is designed this way

The app separates Send from Release reply so you can inspect your move before seeing the counterpart's response. That pause makes the rehearsal more informative. It slows down the natural conversational rush and helps you see how your own line shifts the state of the interaction.

The branch map exists because difficult conversations rarely have one correct script. Different choices create different futures. The tree view lets you compare those futures rather than pretending there is only one optimal path.

The coaching panels and slider snapshots make hidden dynamics more visible. Instead of only asking “Did it work?”, the lab asks what happened to agency, communion, safety, alliance, and escalation risk along the way.

The theory behind it

This lab draws on several overlapping traditions: interpersonal circumplex thinking around agency and communion, working alliance, psychological safety, conflict escalation, and coach-style reframing. It does not claim to measure those constructs clinically. It uses them as a practical vocabulary for noticing conversational dynamics that are often felt but not named.

The rehearsal logic also reflects a broader coaching and communication principle: people usually need more than abstract advice. They need a place to try language, observe consequences, and compare alternatives. This lab provides that place in a structured, reflective form.

Limits of the model

This is a simulated conversation environment, not a prediction engine. Real people bring history, power, culture, mood, trauma, timing, and unpredictability that no branching lab can fully capture. Treat the outputs as prompts for reflection and experimentation, not as guarantees.

If you want to go further

After using the lab, compare two branches that both felt plausible and ask what each one was optimising for. Often the real insight is not which line is “best,” but what each line protects, risks, or invites.

References

Interpersonal circumplex (Agency / Communion): Wiggins, J. S. (1991). Agency and communion as conceptual coordinates for the understanding and measurement of interpersonal behavior. In D. Cicchetti & W. M. Grove (Eds.), Thinking clearly about psychology: Essays in honor of Paul E. Meehl, Vol. 1. Matters of public interest; Vol. 2. Personality and psychopathology (pp. 89–113). University of Minnesota Press.

Working alliance: Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research & Practice, 16(3), 252–260.

Psychological safety: Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative science quarterly, 44(2), 350-383.

Conflict escalation (risk): Glasl, F. (1982). The process of conflict escalation and roles of third parties. In Conflict management and industrial relations (pp. 119-140). Dordrecht: Springer Netherlands.

Nonviolent Communication (coach-style reframes): Rosenberg, M. B. (2015). Nonviolent communication: A language of life (3rd ed.). Encinitas, CA: PuddleDancer.