CBT records
Behavioural experiment worksheet for testing an anxious belief
A short plan for testing a belief by doing something, rather than weighing up the arguments for and against it.
A4, free, no sign-up. The PDF is tagged for screen readers.
When talking it through stops helping
Anxious beliefs rarely give way to argument. You can know that a pause in conversation is
normal and still be sure that yours will be noticed and judged. In a behavioural experiment
you write down what you expect, decide in advance what you would see if you were right and
if you were wrong, and then do the thing and look. The sheet is for adults and teenagers
with health anxiety, social anxiety, panic or generalised anxiety, and it follows the Oxford
guide edited by James Bennett-Levy and colleagues.
A worked example from social anxiety
This made-up example shows the first two sections for someone who dreads speaking in a team
meeting.
Example entries | On the sheet | What they wrote |
|---|
| What I believe will happen | If I stop mid-sentence, people will think I do not know my job, and someone will say something. |
|---|
| How much I believe it | 80 |
|---|
| If it were true | Someone comments or finishes my sentence, people exchange looks, my manager asks me about it afterwards. |
|---|
| If it were not | The meeting carries on and nobody mentions it. |
|---|
| What I am going to do | Pause for a slow count of three, twice, while giving my update. |
|---|
| What I am not going to do | Script my update, look at my notes during the pause, or apologise. |
|---|
If one colleague nods during the pause, nobody says anything and the belief drops to 45, the
next experiment might be a longer pause in a meeting with more senior people.
Getting the design right
A weak experiment tests a belief that cannot lose. “People will think badly of
me” cannot be checked, which is why the sheet asks what you would expect to see. The
other weakness is keeping something you rely on to get through, such as rehearsing the
update word for word, because you can then credit the good outcome to the rehearsal, and the
line about what you are not going to do rules that out.
Behavioural experiments overlap a great deal with exposure, meaning facing a feared
situation on purpose, and few studies have compared the two directly, so it is not settled
whether framing the exercise as a test of a belief adds much.