Behavioural experiment worksheet: test the prediction
A behavioural experiment is a CBT technique for testing an anxious prediction in real life, rather than arguing with it in your head. This guide explains how to write a prediction, plan the test and record the result, and includes a free printable worksheet.
Write your prediction, plan the test and record what actually happened, step by step.
Plan an experiment in EtappiWhat a behavioural experiment is
Anxious beliefs often feel like facts, such as "if I say no, people will be angry". A behavioural experiment tests the belief by doing something and seeing what happens, rather than debating it in your head.
You write down what you expect, try it, and compare the result with your prediction. The evidence you gather is your own, which can feel different from an argument.
Writing a testable prediction
A good prediction is specific and observable, so you can tell afterwards whether it came true. "Something bad will happen" cannot be tested. "If I say no, she'll be angry and stop talking to me" can.
Rate how strongly you believe the prediction, from 0 to 100 percent. You will rate it again after the experiment, and the change tells you what you learned.
Planning the experiment
Decide what you will do, when and where. Keep it small enough to be doable this week. Think about which safety behaviours you will drop, since using them can spoil the test.
Also decide what result would count. If your prediction is that she will be angry, what would you see or hear that would show it? Agreeing this beforehand keeps you honest.
Recording what happened
Straight afterwards, write down what actually happened. Start with facts: what was said, what was done, how long it lasted. Then add how you felt.
Keeping facts and feelings apart matters, because anxiety can colour memory quickly. If you wait a day, the experiment may feel worse than it was.
What did you learn?
Compare the result with your prediction. Then rate your belief again from 0 to 100 percent. Even a small drop is worth noticing, and a change of a few points is a sign that the belief can shift.
Finish by asking what you will do next. You might repeat the experiment in another setting, or try a bigger version. One result does not settle a belief, and a set of results gives you more to go on.
Example experiments
Here are five experiments people often try. Change them to match your own predictions.
- Send an email without rereading it
- Leave a small mistake in a document and see what happens
- Say no to a request
- Ask a question without rehearsing it first
- Arrive five minutes late to a casual meet-up
Free printable worksheet
Behavioural experiment worksheet
Test an anxious prediction by trying something and recording what really happens.
- Fill in the plan before the experiment.
- Fill in the results straight afterwards.
Plan
The belief I'm testing:
How strongly I believe it (0–100%):
My prediction (exactly what will happen):
The experiment (what, when, where):
Safety behaviours I'll drop:
Afterwards
What actually happened:
What I learned:
How strongly I believe it now (0–100%):
This page is general self-help information, not a diagnosis. It is not a substitute for professional care. Free worksheet from Etappi.
Write your prediction, plan the test and record what actually happened, step by step.
Plan an experiment in EtappiGet Etappi on the App StoreGet Etappi on Google Play
Frequently asked questions
What if my prediction comes true?
That's useful too. Look at how bad it was, how you coped, and whether it lasted as long as you expected.
How is an experiment different from exposure?
They overlap. Exposure focuses on staying with a feared situation; an experiment focuses on testing one specific prediction.
How strongly should I believe the prediction?
Any amount works. Rate it before and after so you can see whether it shifted.