15 Behavioral Experiment Examples for Anxiety and Worry

August 31, 2026 · 7 min read

A behavioral experiment is the CBT technique of testing an anxious belief against reality instead of arguing with it. The idea is simple; the hard part is coming up with a test for your worry. So here is a list of concrete examples, grouped by the kind of anxious belief they target.

Every example follows the same recipe: the belief, a prediction specific enough to be wrong, and a check-back moment when reality gets to answer. Steal the closest one and adjust the details to your life.

Social anxiety and what people think of you

1. Saying no. Belief: "If I decline, they will be annoyed with me." Prediction: "When I turn down the extra shift tomorrow, my manager will respond coldly or bring it up again this week." Check back: end of the week.
2. Being visibly nervous. Belief: "If my hands shake, everyone will notice and judge me." Prediction: "If I hold my coffee in the meeting without hiding my hands, at least one person will comment or stare." Check back: right after the meeting.
3. Asking a "stupid" question. Belief: "People will think less of me." Prediction: "If I ask the basic question in the call, someone will laugh, sigh or answer dismissively." Check back: after the call.
4. Leaving the awkward silence. Belief: "I have to fill every pause or people will think I'm boring." Prediction: "If I let one pause sit for three seconds at lunch, the other person will look uncomfortable or end the conversation early." Check back: after lunch.

Catastrophizing and everyday worry

5. The unanswered message. Belief: "They have not replied because something is wrong between us." Prediction: "When my friend finally responds, it will be cold or distant." Check back: when the reply arrives.
6. The presentation. Belief: "This talk is going to be a disaster." Prediction: "I will lose my place badly enough that someone mentions it afterwards." Check back: same evening.
7. The scary email. Belief: "This is bad news, I can feel it." Prediction: "The email from my landlord contains something that costs me money or my apartment." Check back: the moment you open it, which the experiment forces you to do today instead of Friday.
8. The health worry. Belief: "This symptom means something serious." Prediction: "By next Sunday the headache will have gotten worse, not better." Check back: next Sunday. (For persistent symptoms, see a doctor. The experiment targets the hour-by-hour dread, not the checkup.)

Perfectionism and performance

9. Shipping at 90%. Belief: "If I send this report without another polish pass, the flaws will get flagged." Prediction: "Someone will point out an error or sloppy section within two days." Check back: two days after sending.
10. The unpolished message. Belief: "If I write this without rereading it five times, I will embarrass myself." Prediction: "My quickly-written Slack update will get a confused or negative reaction." Check back: end of day.
11. Deliberate small mistake. Belief: "Small mistakes destroy how competent I look." Prediction: "If I leave one harmless typo in an internal doc, someone will judge me for it out loud." Check back: one week.

Safety behaviors: testing what is really carrying you

Safety behaviors are the little defenses that steal credit from every good outcome: over-preparing, rehearsing lines, keeping your camera off. These experiments drop the defense to find out whether it was ever needed.

12. Half the rehearsal. Belief: "It only ever goes okay because I over-prepare." Prediction: "If I prepare for 30 minutes instead of my usual three hours, the meeting will go noticeably worse and someone will notice." Check back: after the meeting.
13. Phone away at the party. Belief: "Without my phone as an escape hatch, I will not cope with the awkwardness." Prediction: "If I keep my phone in my pocket all evening, I will have zero real conversations and leave feeling terrible." Check back: on the way home.

Beliefs about worry itself

14. Worry as insurance. Belief: "My worrying is what keeps bad things from happening." Prediction: "If I postpone all worrying about the trip to a single 15-minute slot, something will go wrong that worrying more would have caught." Check back: after the trip. This one pairs well with the worry time technique.
15. The accuracy audit. Belief: "My worries are usually right." Prediction: log every worry for two weeks as a dated prediction, then count. "At least half of my feared outcomes will actually happen." Check back: day 14. Almost nobody's hit rate survives this audit; across Worrycast users, 73% of tracked fears never happened.

How to actually run these

Three rules keep an experiment honest. Write the prediction down before the event, in wording specific enough that a stranger could score it. Do the test without safety behaviors, or you will hand them the credit. And review the result in writing, because anxious memory quietly deletes the outcomes that disconfirm it. If you want the full method, start with our guide to behavioral experiments.

Worrycast exists because steps one and three are where people fall off. The app stores the prediction the moment the worry strikes, reminds you at the exact check-back moment, and keeps your personal accuracy score across weeks so the pattern becomes impossible to ignore.

Pick one example above and log it as your first prediction.

Worrycast is a self-help tool based on CBT techniques and is not a substitute for professional care. If anxiety is seriously affecting your life, a licensed therapist can help, and behavioral experiments are even better with one.