Wednesday, 29 September 2010

Reality-Testing the Beliefs we Live by: Using Behavioural Experiments

Behavioural Experiments are crucial to working at Level 3 of the Integrative CBT model, and build on Socratic questioning in an ongoing process of guided discovery. Guided discovery (which can also be called collaborative empiricism, trial & review, corrective experience, etc) is the heart of the cognitive-behavioural process of experiential relearning.

Here is one description of the concept of behavioural experiments:

“Behavioural experiments are planned experiential activities, based on experimentation or observation ... Their design is derived directly from a cognitive formulation of the problem, and their primary purpose is to obtain new information which may help to test the validity of ... existing beliefs about themselves, others, and the world...” (Bennett-Levy, J. et al, eds, 2004, Oxford Guide to Behavioural Experiments in Cognitive Therapy. Oxford: OUP – p.8).

Behavioural Experimentation is therefore crucially different from the Behavioural Intervention/Change process which is central to Level 2. Behavioural intervention/change work always focuses on identifying, learning and practising “good”, helpful, useful behaviours in areas of the client’s life where this is necessary (e.g. asserting oneself, relaxing, meditating, eating more healthily, cutting down on drinking, treating oneself well, etc).

Behavioural experimentation on the other hand might equally focus on “bad” behaviours (e.g. leaving a task unfinished, staying up all night, not being “nice” to everyone, etc), since just as much (or more) can be learned experimentally/experientially from the results of such experiments.

An important related point is that behavioural intervention/change work always focuses directly on central areas in the client’s life that need improvement (e.g. socialising, travelling etc), because the main goal is positive change in the client’s daily behaviour and environment. Behavioural experimentation, on the other hand, often works better if it focuses on areas of the client's life which are less central, but where fruitful experiential relearning is more likely to occur. This is because the stakes are less high in these situations, and new learning is less blocked by negative cognitions and high negative emotions. A behavioural experiment sometimes focuses on something quite small, and it may therefore seem trivial to a client, but it needs to be explained that the main goal is some bit of adaptive change in the client’s cognitions.

Some common examples of behavioural experiments are:

• Asking a depressed client to try drinking a cup of tea more slowly than usual, and with a little more mindfulness. As with all behavioural experiments, it is vital that they first of all make a relevant prediction about this. Here the prediction would be a rating between 1 and 10 as to how much they think they will enjoy drinking the tea. Because of the nature of depressive cognitions, they will nearly always underestimate this, and be pleasantly surprised as a result. This serves both to briefly increase their sense of the potential “rewarding-ness” of their environment, but more crucially it also slightly loosens the grip of their depressive beliefs.

• Suggesting to a socially anxious client that they deliberately make themselves blush (perhaps by agreeing to meet and chat with a colleague of yours for a few minutes). Some socially anxious clients have a range of beliefs about blushing, e.g. “I go as red as a fire engine”, “Everybody notices”, “They think I’m pathetic”. Because they believe these things are true, the last thing they want to do is actually face them and check them out, but this is exactly what is needed. The “fire-engine-red” theory can nowadays easily be tested (and falsified) by the use of video recording and replay.

• Suggesting to a client who suffers from panic attacks, and who believes that they are going to have a heart attack when they experience the tightness in the chest which anxiety brings, that they run up and down the stairs rather than sit down and try to calm themselves. This tests the prediction that they would have had a heart attack if they hadn’t minded themselves carefully, and kept their arousal level low, by deliberately increasing their level of physiological arousal (it is generally advised that they have had a check-up from their GP first, just to rule out genuine heart problems – some dangers are real, occasionally!)

All of this is designed to break down self-perpetuating emotional-cognitive-behavioural Vicious Cycles, which can block out experiential relearning for a lifetime if they are not properly addressed. Behavioural experiments are one great way of doing this, and they can be wonderfully immediate, vibrant and exciting (not to mention scary) for both client and therapist.

More on vicious cycles in next week’s blog...


  1. Very informative. Great read.

  2. Thanks for the comment. Glad you find it useful...