The client plans and carries out real behavior that contradicts their schema or coping mechanism. The aim is to gain direct experience that refutes the catastrophic prediction of the schema. Before the experiment, the client formulates a concrete prediction; afterwards, the real outcome is analyzed. The method works where cognitive work is not enough: "I know it logically, but I still don't believe it".
Step-by-step guide
- Identify the target belief: "What stops you from acting differently?"
- Formulate the client's concrete prediction: "What will happen if you do this?"
- Rate the likelihood of a bad outcome on a 0–100 scale
- Design the experiment: what exactly, when, where, with whom
- Optional rehearsal in a role-play in session
- The client runs the experiment and records the outcome
- In the next session: what happened, did it match the prediction, what does it mean for the belief
When to use
- Middle phase of therapy — all types of schemas
- Overcoming avoidance, surrender, overcompensation
- When cognitive work is not enough to change the belief
- Clients who learn better through direct experience
Key phrases
What, in your view, will happen if you voice your opinion? Let's test it with an experiment.
Follow-up questions
On a 0–100 scale, how likely is the bad outcome?
What exactly will you do? When? How will you know the experiment is complete?
What happened? Did it match your prediction? What did you learn about your belief?
Alternative phrasings
Start with a small step — not the scariest experiment, but 30% of it.
Let's rehearse right here: I will be your partner / colleague / boss.
Warnings
- ⚠️ Do not plan experiments that can physically harm the client
- ⚠️ Too big a step causes paralysis — start small
- ⚠️ The client may sabotage the experiment with a "safety move" — a genuine attempt is needed
- ⚠️ Misinterpreting the success ("just lucky") — reinterpret together
Source: Young et al. (2003); adapted from CBT and behavioral psychology
Materials are informational and educational and summarize publicly available scientific sources. They are not medical or psychological advice, are not intended for self-diagnosis or self-treatment, and do not replace consultation with a qualified professional.