Behavioral experiments are planned experiential activities to test the validity of a belief. They are one of the most powerful techniques available to CBT therapists. Behavioral experiments are an information gathering exercise, the purpose of which is to test the accuracy of an individual’s beliefs (about themselves, others, and the world) or to test new, more adaptive beliefs. The use of behavioral experiments in cognitive behavioral therapy mirrors the role that experiments play in other branches of science: experiments are used to gather evidence with which to test a theory.
Behavioural experimentation is widely regarded as the single most powerful way of changing cognitions. (Waller, 2009)
The value of behavioural experiments transcends mere exposure; such experiments allow patient and therapist to collaborate in the gathering of new information assessing the validity of non-threatening explanations of anxiety and associated symptoms. (Salkovskis, 1991)
Beliefs rarely change as a result of intellectual challenge, but only through engaging emotions and behaving in new ways that produce evidence that confirms new beliefs. (Chadwick, Birchwood, Trower, 1996)
Behavioral experiments are planned experiential activities to test the validity of a belief. They are an information gathering exercise, the purpose of which is to test the accuracy of an individual’s beliefs (about themselves, others, and the world) or to test new, more adaptive beliefs (Bennett-Levy et al., 2004). The use of behavioral experiments in cognitive behavioral therapy mirrors the role that experiments play in other branches of science: experiments are used to gather evidence with which to test a theory.
There are strong theoretical grounds for believing that behavioral experiments are capable of promoting greater cognitive, affective, and behavioral change than purely verbal cognitive techniques.
- Teasdale’s Interacting Cognitive Subsystems (ICS) model proposes that people process information using multiple systems: a propositional system which is rational, verbal, & logical; and an implicational system which is holistic, non-linguistic, and which has strong links to emotional systems. Some have argued that behavioral experiments are more likely than purely cognitive tasks to result in changes at the level of the non-linguistic ‘felt sense’ (Bennet-Levy, 2003).
- Well’s metacognitive theory (2000) distinguishes between declarative memory (factual information) and procedural memory (often implicit and automatic). One argument given as an advantage for behavioral experiments is that they promote the development of procedural knowledge as well as declarative memory.
- Theories explaining how people learn have emphasised the importance of personal experience and reflection, both of which are core components of behavioral experiments (Kolb, 1984).
- Behavioural experiments frequently involve some form of exposure to a feared stimulus which typically makes any subsequent learning emotionally ‘hot’. Some authors have proposed that the particular effectiveness of behavioral experimnents is due to the combination of physiological arousal and inhibitory learning (Herbert & Dugas, 2018; Craske et al, 2014).
Behavioral experiments take different forms, often broken down in to ‘hypothesis testing’ and ‘observational’ forms.
- Testing hypothesis A: testing an existing (unhelpful) belief. For example, testing a belief about the catastrophic nature of particular body sensations by practising interoceptive exposure exercises. This is often done in the treatment of panic disorder.
- Testing hypothesis B: testing a new belief. For example, a client with low self-esteem might test a new belief “I deserve to be treated in the same way as other people” by being assertive and saying ‘no’.
- Testing hypothesis A vs. hypothesis B: testing whether the original (threatening) belief or a newly constructed (less threatening) belief better accounts for the evidence. For example, resisting the urge to perform my compulsions to test whether “my family will die” (hypothesis A) or “my intrusive thoughts don’t affect whether people die” (hypothesis B). This is often done after formulating a client’s difficulties using the Theory A vs. Theory B technique.
- Discovery experiments: where the individual does not have a clear hypothesis about what might happen. For example, a client who has successfully avoided their feared situation for a long time might be unclear about specifically what might happen except that it will be ‘bad’.
- Surveys: used where an individual has a belief about what other people think. For example, doing an anonymous survey of ten people to find out the worst intrusive thought they have ever had, and asking them to rate how disgusting they would think a person was for having ‘my’ intrusive thought.
- Direct observation: where an individual has a hypothesis about what might happen, but does not feel capable of testing it directly for themselves. For example, a person who holds the belief that nobody would help them if they were in trouble might watch their therapist pretend to collapse in the street to test whether people offer help or not.
- Information gathering from other sources: such as gathering information from the internet. For example, a person who holds the belief that they are permanently contaminated after experiencing abuse might be encouraged to research how quickly the body replaces and renews all of its cells.
Step 1: Identify the target cognition
The first step in carrying out a behavioral experiment is to identify the target cognition. It is essential to identify these as precisely as possible, and to assess how strongly the individual believes in this prediction or outcome at the outset.
- Beliefs might take the form of an “if… then…” statement, such as “If I make eye contact with people they will attack me”.
- It can be helpful to explore what safety behaviours clients use to prevent negative outcomes. These can then be used to explore underlying beliefs. It can be helpful to ask “What would happen if you were in that situation and didn’t use that safety behavior?”.
- An essential step is to rate the client’s degree of conviction in the belief. This allows for later assessment of change in belief. Conviction ratings can be taken on a 0–10 or 0–100 scale.
Step 2: Design an experiment
Once a target cognition has been established the next step is to design an experiment which will allow the belief to be tested.
- Consider what type of experiment might best test the belief. For example, a direct hypothesis testing experiment, a survey, or a an observational experiment.
- Consider whether the experiment can be conducted in the therapy office, or outside. Quick in-office experiments can help to generate momentum for more substantial out-of-office experiments.
- Consider where the experiment can be conducted, when it will take place, how it is to be conducted (consider what data will need to be recorded: own thoughts, feelings, body sensations and behavior; other’s behavior; the environment), and who will need to be present.
- Therapists should consider: safety, client readiness, and additional practicalities.
- Some thought should be given to preparing for problems. Helpful questions can include “What problems might arise?” and “How would you deal with that?”.
- Have you identified client safety behaviors and agreed to forego them for the experiment? (Or agreed to minimise or monitor their use).
Step 3: Outcome & learning
Take time to understand the meaning of the experiment and the data. What sense has the individual made of it? What does the result say about you? About other people? Encourage reflection on what has been achieved, and what has been learned. Helpful questions might include:
- What happened?
- What did you learn?
- How much do you believe the original belief now?
- How does the outcome of the experiment affect the beliefs you identified?
- What does the result say about ?
- What is a more helpful way of looking at ?
- How does the outcome relate to your original belief? Does it fully support it? Or does it offer any contradictions?
- What are the implications of what you have just done? How could it affect your daily life now?
Step 4: What next?
Reflect upon what needs to be done next to build upon what has been learned. Helpful questions to ask at this stage include:
- What have you learned in this experiment that could be tried again in new situations?
- How can we consolidate what you have learned?
- What other experiments could you do?
- What might you need to do to maintain what you have learned?
- What other therapy tasks could build upon the learning?
- Have you developed any new (perhaps tentative) perspective, and how could they be tested?
- How could you put what you have learned into practice?
- What else needs to be explored or tested?
- Beck, A. T., Rush, A. J., Shaw, B. F., Emery, G. (1979). Cognitive therapy of depression. Guilford press.
- Bennett-Levy, J., Butler, G., Fennell, M. J. V., Hackmann, A., Mueller, M., & Westbrook, D. (Eds.) (2004). The Oxford handbook of behavioural experiments. Oxford: Oxford University Press.
- Bennett–Levy, J. (2003). Mechanisms of change in cognitive therapy: the case of automatic thought records and behavioural experiments. Behavioural and Cognitive
- Psychotherapy, 31, 261–77.
- Chadwick, P. D. J., Birchwood, M. J., & Trower, P. (1996). Cognitive therapy for delusions, voices and paranoia. Chichester: Wiley.
- Craske, M. G., Treanor, M., Conway, C. C., Zbozinek, T., & Vervliet, B. (2014). Maximizing exposure therapy: an inhibitory learning approach. Behaviour Research and Therapy, 58, 10-23.
- Herbert, E. A., Dugas, M. J. (2018). Behavioral expeirments for intolerance of uncertainty: challenging the unknown in the treatment of generalized anxiety disorder. Cognitive and Behavioral Practice, 26(2), 421-436.
- Kolb, D. (1984). Experiential learning: experience as the source of learning and development. Prentice Hall, Englewood Cliffs NJ.
- Salkovskis, P.M. (1991). The importance of behaviour in the maintenance of anxiety and panic: a cognitive account. Behavioural Psychotherapy, 19, 6–19.
- Waller, G. (2009). Evidence-based treatment and therapist drift. Behavior Research and Therapy, 47, 119e127.