Skip to main content
cognitive behavior therapy (CBT)

How To Use Behavioral Experiments To Test What You Believe

Human beings don’t see the world as it is – we all ‘filter’ what we see based on our prior experiences and beliefs. Sometimes our ‘filters’ are so entrenched that they distort how we see the world in ways that are harmful for us. For example, Jennifer believed so strongly that she was unlovable that despite having a devoted partner she was never able to feel safe or happy. In therapy it is often necessary and helpful to examine our (sometimes long-held) beliefs and assumptions. Behavioral Experiments are an excellent way of finding out the truth about our beliefs concerning ourselves, other people, and the world around us.

Our thoughts, beliefs, and assumptions

We all have ‘models’ in our minds of how the world works. Our models are a unique result of our history, experiences, beliefs, and assumptions. The models are a sort of ‘mental short-cut’: one advantage of having them is that they help us to go about our lives without having to think in detail about every single thing that happens to us. One disadvantage is that once our models have formed they are reluctant to change, so they will:

  • pay more attention to information that fits with them,
  • pay less attention to information that contradicts them,
  • ‘squish’ or ‘twist’ information to make it fit.

The result is that we do not perceive the world as it really is. This can cause pain and suffering.

Emma was often criticized and told off as a child, and her sister was their parent’s favourite. As an adult she lives with the belief that she is not good enough, despite being a loving mother and conscientious employee. She is quick to spot her flaws and mistakes, and assumes others do too. She often puts herself down thinking “it is better to get there first before someone else does”. Emma has blind spots for her achievements and positive qualities. She is so focused on potential failure that she can not celebrate her successes. She finds it hard to feel happy for long – and her default state is to feel anxious.

One practice in cognitive behavioral therapy is to examine our beliefs and assumptions, and put them to the test. Many people hold beliefs that are not supported by evidence, and allow these beliefs to dictate how they live their lives. It can be extremely liberating to let go of beliefs and assumptions that have been holding us back.

By the end of therapy David’s opinion of himself was completely different: “I can’t believe that I thought that about myself, you know? I can see it now, but I was blinkered before and it was making me miserable”.

Types of behavioral experiments in CBT

There are many different kinds of behavioral experiments. All have the same purpose – they all allow us to explore or test the truthfulness of beliefs about ourselves, other people, or the world around us. These may be old beliefs that you have had for a long time, or new beliefs that you don’t feel sure about.

Discovery experiments

Sometimes we have ideas about ourselves, other people, or the world, but we only have a tentative grasp of why we think that way. Discovery experiments can be a helpful way of learning more about a situation, even when we don’t have a clear hypothesis about what the result will be.

Alice had been abused by her stepfather when she was a little girl. She often felt ‘dirty’ and believed that she was ‘contaminated’ by what her stepfather had done to her. Her therapist invited her to conduct a discovery experiment using the computer in her office. She asked her to spend 20 minutes searching for information about how the body repairs and replaces its cells. Alice was interested to discover that the body replaces all of its cells every 7 to 10 years. She was reassured by the idea that she was not the same girl who had been abused and found that her anxiety lessened.

Hypothesis testing experiments

Sometimes we have a very clear idea about what we think will happen if we behave a particular way in certain situations. One way of way of testing belief like this to conduct an experiment. We do this the same way in therapy that scientists do in a lab, by:

  1. Making a prediction about what we think will happen (forming a hypothesis)
  2. Carrying out an experiment that will test that hypothesis (more than once if necessary)
  3. Examining the results, and going back to step 1

One way of testing a hypothesis is to do a survey:

Dana believed “other people will think I’m disgusting if they see my surgery scar” very strongly (95%). She and her therapist took a picture of the scar and decided on a quick survey question “What would you think about a person if you found out they had a scar like this?”. Together, Dana and her therapist asked for people’s answers to the survey questions. Dana was surprised to discover that most people would not think anything bad about the person who had the scar. Her belief that “other people will think I’m disgusting if they see my surgery scar” dropped significantly (40%).

Another way of testing a belief is to conduct an experiment:

Dana’s therapist encouraged her to test her beliefs in the real world. Dana still believed that if she showed her scar to people that most would react with disgust. Dana realised that the only way to find out if her belief was true would be to show her scar to people and to gauge their reaction. She decided that in order for it to be a fair test she would need show her scar to 10 people. She predicted that at least 5 would react with disgust, and that she would be able to tell this by looking at the expression on their faces.

How can I do a behavioral experiment?

We can approach our beliefs like a scientist approaching a new phenomenon: they are curious and methodical.

Step 1: Identify the belief to be tested

What is the belief that you have identified that you want to test? Write it down in a single sentence. Examples might include:

  • “I can’t eat in-front of people – if I do they’ll think I’m disgusting”
  • “If I make eye contact with people I’ll be attacked”
  • “If I don’t check the door locks ten times before I go out I’ll leave the house unlocked”

Step 2: Rate the strength of the belief

How strongly do you believe this statement? Rate it from 0% (not at all) to 100% (completely, with all my heart).

Sometimes it can be helpful to give separate ratings for how much you believe it with you head (logically), and how much you believe it with your heart (emotionally).

Step 3: Plan an experiment that could test the belief

Common methods of gathering information to test beliefs are:

  • Surveys – asking “do other people believe the same thing that I believe?”
  • Hypothesis-testing experiments – consider the following conversation in therapy:
    Client: I’m sure that if I do X then Y will happen.
    Therapist: How do you know if you never do X?
    Client: I suppose I don’t.
    Therapist: How could we find out?
    Client: I suppose I could do X and see what happens?

Step 4: Identify any obstacles that could make it difficult to carry out the experiment

Is there anything that could get in the way of doing the experiment?

  • If you need people to be around to help you, who could you ask?
  • If it can only be done in a certain place, when can you go there?
  • Are there any safety issues? How could you minimize them?

Step 5: Carry out the experiment

This is the part that will require courage. You may want to have someone with you who can encourage you, and who can remind you why you’re doing this.

Step 6: Record the result

Every good scientist records what happened.

Step 7: Reflect on your results and re-rate how strongly you believe in the original belief now

Once you have done the experiment go back to your original belief. Read it to yourself and then re-rate how strongly you believe in it now (0-100%).