Behavioural Experiment: test your predictions instead of believing them
Sometimes our thoughts or predictions feel absolutely true, even when we’ve never tested them. A behavioural experiment helps you challenge these beliefs in real life and discover what’s actually true for you. It’s about learning, not proving yourself right or wrong.
Free to try · Responses are private & encrypted
Related tools
What does the Behavioural Experiment Activity cover?
- The belief or prediction you want to test
- How strongly you believe it right now
- The experiment you will run and what you predict will happen
- What actually happened and what you learned
Who is this Behavioural Experiment Activity for?
Helpful if you:
🔮 Are sure something bad will happen if you try
🚫 Avoid situations because of a prediction
🧪 Want to test beliefs in real life rather than argue with them
For therapists:
Design behavioural experiments collaboratively in session and assign the worksheet for the client to complete in the real world. Review outcomes before the next session to consolidate belief change. Fully HIPAA and GDPR compliant.
Dr Harriet Mellotte
Clinical Psychologist & CBT Therapist, BABCP
“Offload is a fantastic resource for both clients and therapists. The world of CBT has been lacking an app which enables the sharing of CBT worksheets and homework tasks between clients and therapists, but Offload completely does this, making the service we offer more professional and the benefits to clients greater.”
Frequently asked questions
What is a behavioural experiment in CBT?
A behavioural experiment is a planned real-world test of a belief or prediction, such as "if I speak up in the meeting, people will think I am stupid". You design a small experiment, predict what will happen, try it, and record what actually happened. Direct experience changes beliefs more powerfully than argument alone.
How is this different from exposure?
Exposure focuses on staying in feared situations until anxiety reduces. A behavioural experiment focuses on testing whether a specific prediction comes true. They often overlap, but experiments are about gathering evidence, which makes them useful for beliefs about other people, yourself, or how events will unfold.
What makes a good behavioural experiment?
A specific prediction, a step that feels challenging but doable, and an honest record of the outcome. Vague experiments produce vague learning. This worksheet guides you through defining the belief, rating how strongly you hold it, designing the test, and reviewing what you found.
Do I need a therapist to do this?
No, though behavioural experiments are a staple of therapist-led CBT and your therapist can assign this worksheet through Offload and review your results. Working alone, start with low-stakes predictions and build up as your confidence in the method grows.
Related reading
Are you a therapist?
Assign this therapy tool and 170+ other interactive tools directly to your clients. Build personalised care plans, customise tools, and review client responses before sessions. HIPAA and GDPR compliant.
Explore Offload for therapistsReferences
Bennett-Levy, J., Butler, G., Fennell, M., Hackmann, A., Mueller, M., & Westbrook, D. (2004). Oxford guide to behavioural experiments in cognitive therapy. Oxford University Press. The foundational text on designing, conducting, and debriefing behavioural experiments within CBT. Bennett-Levy, J., & Padesky, C. A. (2014). Use of behavioural experiments to test beliefs. In J. Bennett-Levy et al. (Eds.), Oxford guide to behavioural experiments in cognitive therapy (2nd ed., pp. 1–25). Oxford University Press. Provides the theoretical and practical framework for belief testing through behavioural change. McManus, F., Van Doorn, K., & Yiend, J. (2012). Examining the effects of thought records and behavioural experiments in cognitive therapy for depression. Behaviour Research and Therapy, 50(3), 204–212. https://doi.org/10.1016/j.brat.2011.12.006 Empirical evidence showing behavioural experiments lead to stronger belief change than thought records alone.
