An Introduction to Cognitive Behaviour Therapy, 2 edition


Basic Theory, Development and Current Status of CBT


In this chapter we want to introduce you to some of the essential background to cognitive behaviour therapy (CBT), including the basic theory and the development of the approach. We start here because CBT is sometimes criticised for being a rather simpleminded ‘cookbook’ approach to therapy: if the client has this problem then use that technique. However, the approach we take in this book is based not on the mechanical application of techniques but on understanding: understanding your patient, understanding CBT theory, and bringing the two together in a formulation (see Chapter 4). You should already have some ideas about understanding people, based on your clinical and personal experience. This chapter will start you on the road to understanding CBT theory.

One further clarification. Talking about CBT as if it were a single therapy is misleading. Modern CBT is not a monolithic structure, but a broad movement that is still developing, and full of controversies. The approach we take in this book is based on the ‘Beckian’ model, first formulated by A.T. Beck in the 1960s and 1970s (Beck, 1963, 1964; Beck, Rush, Shaw & Emery, 1979). This model has been dominant in the UK for the past 30 years, and we would therefore see ourselves as being in the mainstream of CBT in this country. However, other CBT theorists and clinicians might differ, in major or minor ways, with some of the approaches expounded here. We should also say that although we think that some of the newer ideas in CBT, such as the ‘Third Wave’ therapies (Hayes, 2004), are exciting developments that have the potential to enrich CBT greatly, our aim here is primarily to provide a foundation for ‘basic’ CBT. We therefore restrict our consideration of those developments to a separate chapter (Chapter 17).

A brief history of CBT

Just as some knowledge of a client’s background can be helpful in understanding his current state, an appreciation of how CBT developed can help us to understand its modern form. Modern CBT has two main influences: first, behaviour therapy as developed by Wolpe and others in the 1950s and 1960s (Wolpe, 1958); and second, the cognitive therapy approach developed by A.T. Beck, beginning in the 1960s but becoming far more influential with the ‘cognitive revolution’ of the 1970s.

Behaviour therapy (BT) arose as a reaction against the Freudian psychodynamic paradigm that had dominated psychotherapy from the nineteenth century onwards. In the 1950s, Freudian psychoanalysis was questioned by scientific psychology because of the lack of empirical evidence to support either its theory or its effectiveness (Eysenck, 1952). BT was strongly influenced by the behaviourist movement in academic psychology, which took the view that what went on inside a person’s mind was not directly observable and therefore not amenable to scientific study. Instead behaviourists looked for reproducible associations between observable events, particularly between stimuli (features or events in the environment) and responses (observable and measurable reactions from the people or animals being studied). Learning theory, a major model in psychology at that time, looked for general principles to explain how organisms learn new associations between stimuli and responses.

In this spirit, BT avoided speculations about unconscious processes, hidden motivations and unobservable structures of the mind, and instead used the principles of learning theory to modify unwanted behaviour and emotional reactions. For instance, instead of trying to probe the unconscious roots of an animal phobia, as Freud famously did with ‘Little Hans’ (a boy who had a fear of horses: Freud, 1909), behaviour therapists constructed procedures, based on learning theory, which they believed would help people learn new ways of responding. The BT view was that someone like Little Hans had learned an association between the stimulus of a horse and a fear response, and the task of therapy was therefore to establish a new, non-fearful, response to that stimulus. The resulting treatment for anxiety disorders, known as systematic desensitisation, asked clients to repeatedly imagine the feared stimulus whilst practising relaxation, so that the fearful response would be replaced by a relaxed response. Later developments often replaced imaginal exposure (e.g. thinking about a mental picture of the horse) with in vivo exposure (approaching a real horse).

BT rapidly became successful, especially with anxiety disorders such as phobias and obsessive-compulsive disorder (OCD), for two main reasons. First, in keeping with its roots in scientific psychology, BT had always taken an empirical approach, which soon allowed it to provide solid evidence that it was effective in relieving anxiety problems. Second, BT was a far more economical treatment than traditional psychotherapy, typically taking six to 12 sessions.

Despite this early success, there was some dissatisfaction with the limitations of a purely behavioural approach. Mental processes such as thoughts, beliefs, interpretations, imagery and so on, are such an obvious part of life that it began to seem absurd for psychology not to deal with them. During the 1970s this dissatisfaction developed into what became known as the ‘cognitive revolution’, wherein ways were sought to bring cognitive phenomena into psychology and therapy, whilst still trying to maintain an empirical approach that would avoid ungrounded speculation. Beck and others had in fact begun to develop ideas about cognitive therapy (CT) during the 1950s and early 1960s, but their ideas became increasingly influential. The publication of Beck’s book on cognitive therapy for depression (Beck et al., 1979), and research trials showing that CT was as effective a treatment for depression as anti-depressant medication (e.g. Rush, Beck, Kovacs & Hollon, 1977), fuelled the revolution. Over the succeeding years, BT and CT grew together and influenced each other to such an extent that the resulting amalgam is now most commonly known as cognitive behaviour therapy – CBT.

Some basic principles

So, what elements of BT and CT have emerged to form the foundation of modern CBT? Here we set out what we see as the most basic principles and beliefs on which our model of CBT is based, so that you can decide for yourself whether you think they make sense – or at least enough sense to be worth giving CBT a try. Below are what we consider to be the fundamental beliefs about people, problems and therapy that are central to CBT. We are not suggesting that these beliefs are necessarily unique to CBT – many of them may be shared by other approaches – but the combination of these principles goes some way towards characterising CBT.

Figure 1.1    The basic cognitive principle

The cognitive principle

The core idea of any therapy calling itself ‘cognitive’ is that people’s emotional reactions and behaviour are strongly influenced by cognitions (in other words, their thoughts, beliefs and interpretations about themselves or the situations in which they find themselves – fundamentally the meaning they give to the events of their lives). What does this mean?

It may be easiest to start from a ‘non-cognitive’ perspective. In ordinary life, if we ask people what has made them sad (or happy, or angry, or whatever), they often give us accounts of events orsituations: for example, ‘I am fed up because I have just had a row with my girlfriend’. However, it cannot be quite that simple. If an event automatically gave rise to an emotion in such a straightforward way, then it would follow that the same event would have to result in the same emotion for anyone who experienced that event. What we actually see is that to a greater or lesser degree, people reactdifferently to similar events. Even events as obviously terrible as suffering a bereavement, or being diagnosed with a terminal illness, do not produce the same emotional state in everyone: some may be completely crushed by such events, whilst others cope reasonably well. So it is not just the event that determines emotion: there must be something else. CBT says that the ‘something else’ is cognition, i.e. the interpretations people make of the event. When two people react differently to an event it is because they are seeing it differently, and when one person reacts in what seems to be an unusual way, it is because he has unusual thoughts or beliefs about the event: it has an idiosyncratic meaning for him. Figure 1.1 illustrates this.

Let’s look at a simple example of this process. Suppose you are walking down the street and you see someone you know coming the other way, but she does not seem to notice you. Below are a number of possible thoughts about this event, and some possible emotional responses arising from those interpretations.


  • ‘I can’t think of anything to say to her, she’ll think I’m really boring and stupid.’ [Leading to anxiety]
  • ‘Nobody would ever want to talk to me anyway, no one seems to like me.’ [Depression]
  • ‘She’s got a nerve being so snooty, I’ve not done anything wrong.’ [Anger]
  • ‘She’s probably still hung over from that party last night!’ [Amusement]

This illustrates the fundamental cognitive principle, that different cognitions give rise to different emotions. It also shows the association between certain kinds of cognition and corresponding emotional states: for instance, that thoughts about others being unfair, or breaking rules that we hold dear, are likely to be associated with anger. We shall have more to say about this idea later.

There is, of course, nothing new about the idea that meaning is important. The ancient Greek Stoic philosopher Epictetus said over 1,800 years ago that ‘Men are disturbed, not by things, but by the principles and notions which they form concerning things.’ Yet as we shall see in the rest of this book, the ramifications and elaborations of this simple idea have led to the development of a powerful approach to helping people in distress. By helping people to change their cognitions, we may be able to help them change the way they feel.

The behavioural principle

Part of the inheritance from BT is that CBT considers behaviour (what we do) as crucial in maintaining – or in changing – psychological states. Consider the above example again. If you had either the first or second cognition, then your subsequent behaviour might have a significant effect on whether your anxiety or depression persisted. If you approached your acquaintance and chatted, you might discover that she was actually friendly towards you. As a result, you might be less inclined to think negatively in future. On the other hand, if you pretended not to see her, you would not have a chance to find out that your thoughts were inaccurate, and negative thoughts and associated emotions might persist. Thus, CBT believes that behaviour can have a strong impact on thought and emotion, and, in particular, that changing what you do is often a powerful way of changing thoughts and emotions.

The ‘continuum’ principle

In contrast to some more traditional medical approaches, CBT believes that it is usually more helpful to see mental health problems as arising from exaggerated or extreme versions of normal processes, rather than as pathological states that are qualitatively different from, and inexplicable by, normal states and processes. In other words, psychological problems are at one end of a continuum, not in a different dimension altogether. Related to this belief are the further ideas that (a) psychological problems can happen to anyone, rather than being some freakish oddity; and (b) that CBT theory applies to therapists as much as to clients.

The ‘here and now’ principle

Traditional psychodynamic therapy took the view that looking at the symptoms of a problem – for example, the anxiety of a phobic person – was superficial, and that successful treatment must uncover the developmental processes, hidden motivations and unconscious conflicts that were supposed to lie at the root of a problem. BT took the view that the main target of treatment was the symptoms themselves and that one could tackle the anxiety (or whatever) directly, by looking at what processes currently maintained it and then changing those processes. Psychoanalysis argued that treating symptoms rather than the supposed ‘root causes’ would result in symptom substitution, i.e. the unresolved unconscious conflict would result in the client’s developing new symptoms. In fact, a wealth of research in BT showed that such an outcome, although possible, was rare: more commonly, tackling symptoms directly actually resulted in more global improvement.

Modern CBT has inherited BT’s approach. The main focus of therapy, at least most of the time, is on what is happening in the present, and our main concerns are the processes currently maintaining the problem, rather than the processes that might have led to its development many years ago. Chapter 4 on assessment and formulation discusses this further.

The ‘interacting systems’ principle

This is the view that problems should be thought of as interactions between various ‘systems’ within the person and in their environment, and it is another legacy from BT (Lang, 1968). Modern CBT commonly identifies four such systems:


  • cognition
  • affect, or emotion
  • behaviour
  • physiology.

These systems interact with each other in complex feedback processes and also interact with the environment – where ‘environment’ is to be understood in the widest possible sense, including not just the obvious physical environment but also the social, family, cultural and economic environment. Figure 1.2, based on what is sometimes called the ‘hot cross bun’ model (Padesky & Mooney, 1990), illustrates these interactions.

Figure 1.2    Interacting systems

This kind of analysis helps us to describe problems in more detail, to target specific aspects of a problem and also to consider times when one or more systems are not correlated with the others. For example, ‘courage’ could be said to describe a state where a person’s behaviour is not correlated with her emotional state: although she is feeling fearful, her behaviour is not fearful.

The empirical principle

CBT believes we should evaluate theories and treatments as rigorously as possible, using scientific evidence rather than just clinical anecdote. This is important for several reasons:


  • Scientifically, so that our treatments can be founded on sound, well-established theories. One of the characteristic features of CBT is that, in contrast to some schools of therapy that have remained little changed since they were first devised, it has developed and made steady advances into new areas through the use of scientific research.
  • Ethically, so that we can have confidence in telling people who are receiving and/or purchasing our treatments that they are likely to be effective.
  • Economically, so that we can make sure that limited mental-health resources are used in the way that will bring most benefit.

Summary of CBT principles

These then are what we would take as the basic principles at the heart of CBT. To summarise:


  • The cognitive principle: it is interpretations of events, not events themselves, which are crucial.
  • The behavioural principle: what we do has a powerful influence on our thoughts and emotions.
  • The continuum principle: mental-health problems are best conceptualised as exaggerations of normal processes.
  • The here-and-now principle: it is usually more fruitful to focus on current processes rather than the past.
  • The interacting-systems principle: it is helpful to look at problems as interactions between thoughts, emotions, behaviour and physiology and the environment in which the person operates.
  • The empirical principle: it is important to evaluate both our theories and our therapy empirically.

Let us now turn to an elaboration of the fundamental cognitive principles.

‘Levels’ of cognition

So far we have talked about ‘cognition’ as if it were a single concept. In fact, CBT usually distinguishes between different kinds or ‘levels’ of cognition. The following account of levels of cognition is based on what has been found clinically useful; a later section will briefly consider the scientific evidence for some of these ideas. Note that different CBT practitioners might categorise cognitions differently, and although the following classification is commonly used, it is not the only one.

Negative automatic thoughts (NATs)

Negative automatic thoughts,1 as first described by Beck, are fundamental to CBT. This term is used to describe a stream of thoughts that almost all of us can notice if we try to pay attention to them. They are negatively tinged appraisals or interpretations – meanings we take from what happens around us or within us.

Think of a recent time when you became upset: anxious, annoyed, fed up or whatever. Put yourself back in that situation and remember what was going through your mind. Most people can fairly easily pick out NATs. For example, if you were anxious, you might have had thoughts about the threat of something bad happening to you or people you care about; if you were annoyed, you might have had thoughts about others being unfair, or not following rules you consider important; if you were fed up, there might have been thoughts about loss or defeat, or negative views of yourself.

NATs are thought to exert a direct influence over mood from moment to moment, and they are therefore of central importance to any CBT therapy. They have several common characteristics:


  • As the name suggests, one does not have to tryto think NATs – they just happen, automatically and without effort (although it may take effort to pay attention to them and notice them).
  • They are specific thoughts about specific events or situations. Although they may become stereotyped, particularly in chronic problems, they may also vary a great deal from time to time and situation to situation.
  • They are, or can easily become, conscious. Most people are either aware of this kind of thought, or can soon learn to be aware of them with some practice in monitoring them.
  • They may be so brief and frequent, and so habitual, that they are not ‘heard’. They are so much a part of our ordinary mental environment that unless we focus on them we may not notice them, any more than we notice breathing most of the time.
  • They are often plausible and taken as obviously true, especially when emotions are strong. Most of the time we do not question them, but simply swallow them whole. If I think ‘I am useless’ when I am feeling fed up about something’s having gone wrong, it seems a simple statement of the truth. One of the crucial steps in therapy is to help clients stop swallowing their NATs in this way, so that they can step back and consider their accuracy. As a common CBT motto has it, ‘Thoughts are opinions not facts’ – and like all opinions they may or may not be accurate.
  • Although we usually talk about NATs as if they were verbal constructs – e.g. ‘I am useless’ – it is important to be aware that they may also take the form of images. For example, in social phobia, rather than thinking in words, ‘Other people think I’m peculiar’, a person may get a mental image of himself looking red-faced, sweaty and incoherent.
  • Because of their immediate effect on emotional states, and their accessibility, NATs are usually tackled early on in therapy.

Core beliefs

At the other end of the scale from NATs, core beliefs represent a person’s ‘bottom line’, their fundamental beliefs about themselves, other people, or the world in general. Characteristics of core beliefs are:


  • Most of the time they are not immediately accessible to consciousness. They may have to be inferred by observation of one’s characteristic thoughts and behaviours in many different situations.
  • They manifest as general and absolute statements, e.g. ‘I am bad’, or ‘Others are not to be trusted’. Unlike NATs, they do not typically vary much across times or situations but are seen by the person as fundamental truths that apply in all situations.
  • They are usually learned early on in life as a result of childhood experiences, but they may sometimes develop or change later in life, e.g. as a result of severe trauma.
  • They are generally not tackled directly in short-term therapy for focal problems such as anxiety disorders or major depression (although they may change anyway). Tackling them directly may be more important in therapy for chronic problems like personality disorders (see Chapter 17).

Dysfunctional assumptions

Dysfunctional assumptions (DAs) can be considered as bridging the gap between core beliefs and NATs. They provide the ‘soil’ from which NATs sprout. DAs can be thought of as ‘rules for living’, more specific in their applicability than core beliefs, but more general than NATs. They often take the form of conditional ‘If … then …’ propositions, or are framed as ‘should’ or ‘must’ statements. They often represent attempts to live with negative core beliefs. For example, if I believe that I am fundamentally unlovable, I may develop the assumption, ‘If I always try to please other people then they will tolerate me, but if I stand up for my own needs I will be rejected’ or ‘I must always put other’s needs first, otherwise they will reject me’. Such a DA offers me a guide to how to live my life so as to overcome some of the effects of the core belief, but it is always a fragile truce: if I fail to please someone, then I am in trouble. When one of my DAs is violated, then NATs and strong emotions are likely to be triggered. Characteristics of DAs are:


  • Like core beliefs, they are not as obvious as NATs and may not be easily verbalised. They often have to be inferred from actions or from patterns of common NATs.
  • They are usually conditional statements, taking the form of ‘If … then …’, or ‘should/must … otherwise …’ statements.
  • Some may be culturally reinforced: for example, beliefs about putting others first, or the importance of success, may be approved of in some cultures.
  • What makes them dysfunctional is that they are too rigid and over-generalised, not flexible enough to cope with the inevitable complications and setbacks of life.
  • They are usually tackled later on in therapy, after the client has developed some ability to work with challenging NATs. It is thought that modifying DAs may be helpful in making clients more resistant to future relapse (Beck et al., 1979).

Figure 1.3    Illustration of levels of cognition

Figure 1.3 illustrates these levels of cognitions for one kind of belief and also shows some of the dimensions along which the levels vary.

It is easy to assume that core beliefs are ‘at the root’ of the problem, or are the ‘underlying’ cause, and that therefore they must be tackled directly for therapy to be effective. We would question this assumption. Core beliefs are certainly more general than NATs, but that does not necessarily mean they are more important. Most successful CBT research to date targets NATs, but that does not make the therapy ineffective or short-lived. This is probably because people with common mental-health problems such as anxiety or depression have a range of core beliefs, not just negative and unhelpful ones. Through the process of therapy they can bring their more positive beliefs back into operation. Although there is not yet much research evidence, working with core beliefs may be more important in lifelong problems such as personality disorders, where clients may never have formed much in the way of positive beliefs.

Characteristic cognitions in different problems

We mentioned earlier that modern CBT theories see characteristic forms of cognition associated with particular kinds of problem. These characteristic patterns involve both the content of cognitions and theprocess of cognition. If we take depression as an example, then the thoughts of depressed people are likely to contain characteristic contents, e.g. negative thoughts about themselves or others. Depressed people are also likely to show characteristic general biases in the way that they think, e.g. towards perceiving and remembering negative events more than positive ones; or tending to see anything that goes wrong as being their fault; or over-generalising from one small negative event to a broad negative conclusion. Here, we briefly consider some examples. (See also later chapters on specific problems.)


As first described by Beck, the characteristic cognitions in depression are the negative cognitive triad, namely negatively biased views of oneself, of the world in general and of the future. In other words, the typical depressed view is that I am bad (useless, unlovable, incompetent, worthless, a failure, etc.); the world is bad (nothing good happens, life is just a series of trials); and the future is also bad (not only are myself and the world bad, but it will always be like this and nothing I can do will make any difference).


The general process here is a bias towards the over estimation of threat, i.e. perceiving a high risk of some unwanted outcome. The exact nature of the threat, and therefore the content of cognitions, is different in different disorders. For example:


  • In panic, there is catastrophic misinterpretation of harmless anxiety symptoms as indicating some imminent disaster, e.g. dying or ‘going mad’.
  • In health anxiety, there is a similar misinterpretation of harmless symptoms as indicating illness, but on a longer time scale: e.g. I might have a disease that will make me die sometime in the future.
  • In social anxiety, thoughts are about being negatively evaluated by others, e.g. ‘They will think I am stupid (or boring, or peculiar, or …)’.
  • In OCD, thoughts are about being responsible for, and/or needing to prevent, some harm to oneself or others.


In anger, the thoughts are usually about others’ behaviour being unfair, breaking some implicit or explicit rule, or having hostile intent: ‘They ought not to do that, it’s not fair, they’re trying to put me down’.

Figure 1.4    Generic problem development model

Generic CBT model of problem development

We shall finally put together the ideas introduced so far to develop a broad picture of how CBT sees the development of problems (see Figure 1.4). It proposes that through experience (most commonly childhood experience, but sometimes later experience), we develop core beliefs and assumptions which are to a greater or lesser extent functional and which allow us to make sense of our world and find a way through it. Most of us have a mixture of functional and dysfunctional beliefs, with the functional ones allowing us to cope reasonably well most of the time. Even quite dysfunctional beliefs may not cause any particular problems for many years. However, if we encounter an event or series of events that violates a core belief or assumption and cannot be handled by our more positive beliefs (sometimes called a critical incident), then dysfunctional assumptions become more active, negative thoughts are evoked, and unpleasant emotional states such as anxiety or depression result. Interactions between negative thoughts, emotions, behaviour and physiological changes may then result in persisting dysfunctional patterns, and we get locked into vicious cycles or feedback loops that serve to maintain the problem.

The current status of CBT

Having earlier reviewed the history of CBT, in this section we review its current status and important contexts.

Improving access to psychological therapies (IAPT)

Perhaps the most significant development in the wider landscape of CBT since the first edition of this book, at least in the UK, is the explosive growth of the UK government’s Improving Access to Psychological Therapies (IAPT) programme. This programme resulted from planning and lobbying led by a prominent economist and adviser to the government, Lord Layard, who became convinced that (a) mental health problems are a major source of unhappiness and loss of economic activity; and (b) that CBT could make a difference to many of the most common mental health problems (Centre for Economic Performance, 2006). He and others, especially Professors David Clark and David Richards, persuaded the government that a major investment in psychological therapy would make an impact on health, and also that such an investment would be largely self-funding, since improvements in mental health would allow a proportion of patients to return to work, thus saving on unemployment benefits.

After some early pilot work in two sites starting in 2006, the government announced in October 2007 that the programme, known as Improving Access to Psychological Therapies (IAPT:, would receive a large amount of funding over at least three years, amounting to over £170 million annually by the third year. This funding is going into a massive increase in the provision of evidence-based psychological therapies, mainly aimed at treating anxiety and depression in primary care: the goal is to train and put into NHS services 3,600 new therapists over this period. The first of the new services and training courses were up and running by the autumn of 2008.

IAPT’s first wave involved two types of CBT, because it was thought that although CBT has a strong evidence base (see later in this chapter), there was a particular shortage of qualified CBT therapists. (In the latest IAPT plans another evidence-based therapy, interpersonal therapy [IPT], is also to be provided for depression, as well as counselling and couples therapy for mild/moderate depression.) The first category of IAPT CBT, making up about 60% of the new therapists, is known as ‘High Intensity’ therapy (HI), by contrast with the second group, making up about 40% of therapists, who will be ‘Low Intensity’ (LI) workers (since re-titled ‘Psychological Wellbeing Practitioners’, PWPs). Both groups’ training is initially funded by the government, and consists of a one-year in-service training course. The HI course is a total of around 65 days over the year, whilst LI training is around 25 days. HI workers are supposed to be professionally qualified – nurses, psychologists, etc, – but LI workers do not have to have any professional qualification and are expected to match their local communities more closely in terms of education, class and so on. Preliminary data from the pilot sites support the effectiveness of the programme (Clark, Layard, Smithies, Richards, Suckling & Wright, 2009) but of course more evidence is still needed. Both the pilot sites are the subjects of formal external evaluation research, but the results have not yet been published.

The main focus of this book is on traditional CBT, or HI therapy in IAPT terms, but in many ways the LI services are the most radical part of the programme, as they deliver CBT in ways quite different to traditional therapy. We briefly consider some of the features of LI CBT in Chapter 16, but for more detail see Richards (2010) and Bennett-Levy et al. (2010).

CBT competences

Another important development since this book’s first edition is the publication of the ‘CBT competences framework’. This initiative, funded by the UK Department of Health, was linked to the IAPT programme and had the goal of identifying what skills a therapist needs in order to provide good-quality CBT for anxiety and depression: if we are going to train the many more CBT therapists that IAPT proposed, what exactly should we be training them to do? Roth and Pilling (2007), in consultation with an expert reference group, produced a useful mapping of competences for both LI and HI interventions. Their approach was to identify important competences through a close examination of the treatment manuals for CBT interventions that have been shown to be effective for different disorders. It seemed reasonable to suppose that since treatment based on these manuals works, then if other therapists follow the same strategies they will be providing effective treatment. Roth and Pilling produced a ‘map’ of competences divided into five domains:


  • Generic competences in psychological therapy.These are the basic competences needed by a therapist from any school of therapy: e.g. knowledge of mental health, ability to relate to clients, and so on.
  • Basic CBT competences.Skills related to the basic structure of CBT therapies, such as agenda-setting or use of homework.
  • Specific CBT techniques.The core treatment strategies, such as using thought records and identifying and testing thoughts and beliefs.
  • Problem-specific competences.Approaches used in treatment programmes for particular disorders, such as Beckian cognitive therapy for depression, or exposure and response prevention for obsessive-compulsive disorder.
  • Meta-competences.The ‘higher level’ skills that allow a therapist to make effective judgements about when to use which specific treatment strategy. Includes using the formulation to adapt treatment to an individual, dealing with difficulties during treatment, etc.

This framework is too detailed to reproduce here, but see Roth and Pilling (2007), and the CORE website cited in ‘Further Reading’, for more detailed information about the competence framework.

In this book we aim to introduce you to CBT skills in all these domains, with some chapters mapping particularly closely onto specific domains as follows:


  • generic competences in psychological therapy – Chapters 3and 19
  • basic CBT competences – Chapters 1256and 11
  • specific CBT techniques – Chapters 710
  • problem-specific competences – Chapters 1215
  • meta-competences – Chapters 4and 11.

The empirical evidence about CBT

Finally, since we have talked about CBT’s commitment to empiricism, we should consider the empirical status of CBT. What is the evidence that CBT is effective? And what is the evidence that CBT theory is an accurate model of human functioning?

Evidence regarding CBT treatment

Roth and Fonagy (2005), in the second edition of What works for whom? (their landmark summary of psychotherapy efficacy), report evidence showing that CBT is strongly supported as a therapy for most of the psychological disorders in adults that they studied, and has more support in more kinds of problem than any other therapy. Figure 1.5 summarises this.

Figure 1.5    Summary by the current authors, adapted from Roth and Fonagy (2005), Chapter 17

In addition to this evidence of CBT’s efficacy (i.e. that it works in tightly controlled research trials), there is also some useful evidence demonstrating its effectiveness (i.e. that it can also work well in ordinary clinical practice, outside specialist research centres). See, for example, Merrill, Tolbert and Wade (2003), Stuart, Treat and Wade (2000) and Westbrook and Kirk (2005).

A second useful source of evidence is the UK National Institute for Health and Clinical Excellence (NICE). This is an agency charged by the government with the task of surveying the evidence for the effectiveness of different treatments and making recommendations about which treatments ought therefore to be made available in the National Health Service (NHS). In the past ten years, NICE has produced guidelines on several major mental-health problems, which include the following recommendations:


  • Schizophrenia (NICE, 2009a): ‘Offer cognitive behavioural therapy (CBT) to all people with schizophrenia.…’ (p. 9).
  • Depression (NICE, 2009b):

o    For people with persistent subthreshold depressive symptoms or mild to moderate depression, consider offering one or more of … individual guided self-help based on the principles of cognitive behavioural therapy (CBT); computerised cognitive behavioural therapy (CCBT) …’ (p. 9);

o    For people with moderate or severe depression, provide a combination of antidepressant medication and a high-intensity psychological intervention (CBT or IPT) …’ (p. 9);

o    People with depression who are considered to be at significant risk of relapse … or who have residual symptoms, should be offered one of the following … individual CBT; … mindfulness-based cognitive therapy for people who are currently well but have experienced three or more previous episodes of depression …’ (p. 10).

  • Eating disorders (NICE, 2004b): ‘Cognitive behaviour therapy for bulimia nervosa … should be offered to adults with bulimia nervosa …’ (p. 4); ‘Cognitive behaviour therapy for binge eating disorder … should be offered to adults with binge eating disorder …’ (p. 5).
  • Generalised anxiety and panic (NICE, 2004c): ‘The interventions that have evidence for the longest duration of effect, in descending order, are: [first] cognitive behavioural therapy; …’ (p. 6).
  • Post-traumatic stress disorder (PTSD) (NICE, 2005): ‘All people with PTSD should be offered a course of trauma-focused psychological treatment (trauma-focused cognitive behavioural therapy [CBT] or eye movement desensitisation and reprocessing [EMDR]) …’ (p. 4).
  • Obsessive compulsive disorder (OCD) and body dysmorphic disorder (BDD) (NICE, 2005): people with OCD or BDD should be offered CBT (including exposure and response prevention), either in group or individual format and, depending on severity and preference, also consider SSRI medication; also says that ‘… when adults with OCD request forms of psychological therapy other than cognitive and/or behavioural therapies as a specific treatment for OCD … they should be informed that there is as yet no convincing evidence for a clinically important effect of these treatments …’ (pp. 18–21).

In summary then, at the time of writing, CBT is the psychological therapy with the most solid and wide evidence base for efficacy and effectiveness.

Evidence regarding CBT theory

It is a fallacy to think that demonstrating the efficacy of a treatment proves the truth of the theory on which that treatment is based. The treatment’s efficacy could be due to some combination of factors not imagined in the theory. Thus, for most of us, even a randomised controlled trial (RCT) showing that a treatment based on traditional witchcraft was effective for depression would not necessarily convince us that depression was in fact caused by evil spirits; instead we might investigate whether there was a powerful placebo effect, or perhaps whether the herbal potions used in the treatment contained a psychoactive substance. In the same way, the efficacy of CBT as a treatment does not show that CBT theory is true. In fact, the evidence for some of the fundamental theoretical ideas of CBT is more patchy than the evidence for the treatment’s efficacy. Clark, Beck and Alford (1999) present a detailed consideration of the balance of scientific evidence in the case of the cognitive theory of depression. In summary, they conclude that regarding the supposed patterns of negative thinking in depression, there is evidence that:


  • there is an increase in negative thinking about oneself, the future and (less clearly) the world;
  • there is a reduction in positive thinking about the self, but this change is less marked and may be less specific to depression (in other words, the same thing also happens in other problems);
  • there is a specific increase in thoughts and beliefs about loss and failure (more so than people who suffer from anxiety problems).

Regarding the proposed causal role of negative thoughts, i.e. the suggestion that negative thinking can provoke low mood, Clark et al. conclude that there is some experimental evidence that negative self-referent thinking can indeed induce subjective, behavioural, motivational and physiological features similar to mild to moderate depression. If we experimentally provoke negative thoughts about themselves in non-depressed people, we can produce temporary states quite similar to depression.

There is also some evidence that the proposed cognitive processing biases can be identified in experiments, with evidence that in depressed people there is:


  • a bias towards processing negative information relevant to themselves (but no such bias for neutral or impersonal information);
  • enhanced recall of negative events, and increased negative beliefs.

Furthermore, there is evidence that these changes in processing can occur at an automatic, pre-conscious level.

The least well-supported part of the theory is the suggestion that people are vulnerable to depression because of negative beliefs that are still present in ‘latent’ form even when they are not depressed. Clark et al. suggest that there is a little supportive evidence for this idea, but that it has proved difficult to get clear evidence (perhaps not surprisingly, when one considers the difficulties of identifying such ‘latent’ beliefs experimentally).

A similar picture is found for specific CBT models for other disorders: in some areas there is good solid research support and in others the evidence is equivocal. Overall, then, the evidence is:

a    that CBT is undoubtedly an effective treatment for many problems; and

b    that there is support for CBT theory but that there is still room for exploring and developing this approach further in some areas.



  • Modern CBT is derived from the legacy of behaviour therapy (with its emphasis on the importance of behaviour changein overcoming mental health problems), and cognitive therapy (with its emphasis on understanding and changing the meaning of events).
  • Problems can usefully be described in terms of the interactions between four ‘systems’:

o    The cognitive system – what a person thinks, imagines, believes.

o    The behavioural system – what they do or say that can be directly observed by others.

o    The affective system – their emotions.

o    The physiological system – what happens to their body, such as autonomic arousal or changes in appetite.

  • We distinguish three ‘levels’ of cognition:

o    Negative automatic thoughts – specific thoughts that arise spontaneously in various situations, which have a negative effect on mood, and which are relatively accessible to consciousness.

o    Dysfunctional assumptions – ‘rules for living’ that guide behaviour and expectations in a variety of situations, and which are often in conditional (if … then …) form.

o    Core beliefs – very general beliefs about oneself, other people or the world in general, which operate across a wide range of situation but which are often not immediately conscious.

  • Different kinds of psychological problem have different characteristic cognitions, in content, style or both – e.g. in anxiety there is a preoccupation with threat, and associated biases towards perceiving threat.
  • There is considerable evidence that CBT can be an effective way of helping various mental health problems; and less clear, but still significant, evidence for the theories lying behind the treatment.

Learning exercises

Review and reflection:


  • Consider what you think about the basic principles of CBT outlined in this chapter. Do they make sense to you? Are there any principles which do not fit, or which do not make sense to you?
  • What do you think of the cognitive theory underpinning CBT? Does it make sense to you? Does it fit with your clinical experience?
  • Does it matter that the evidence for the theory of CBT is less solid than the evidence for its efficacy as a treatment?

Taking it forward:


  • Observe your own experience of negative thoughts, dysfunctional assumptions and core beliefs. Try to tune into your thoughts, especially when you are upset or emotionally aroused in some way. Do your thoughts follow any of the patterns described here? What are the similarities or differences between your experience and our descriptions of different kinds of cognition?
  • Does this observation of your own thoughts have any implications for your clinical practice?
  • If so, how will you adapt your clinical practice?

Further reading

Beck, A.T., Rush, A.J., Shaw, B.F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press.

Although now over 30 years old, the book that started the cognitive revolution is still a classic, with a real feel for the clinical realities of working with depressed patients.

Greenberger, D., & Padesky, C. (1995). Mind over mood. New York: Guilford Press.

A self-help book designed to assist clients, but also a clear and simple introduction to CBT that many new therapists find very useful, whether or not they plan to use it with clients!

House, R., & Loewenthal, D. (Eds.). (2009). Against and for CBT: towards a constructive dialogue? Ross-on-Wye: PCCS Books.

A mixture of (mostly highly critical) views on the philosophy, science, ethics and politics of CBT (and by association, the IAPT programme). Often coming from a post-modernist stance, and containing some of the jargon so often associated with that approach, many of the chapters are not easy reading – but interesting if you want a different perspective.

The Centre for Outcomes Research & Effectiveness (CORE) website at University College, London, under whose auspices the CBT Competences framework was developed

The CORE site contains more detailed descriptions of the CBT competences for anxiety and depression, and a self-assessment tool to allow clinicians to evaluate how well their own skills match the competences.

1Note that there can also be positive automatic thoughts, or indeed neutral ones; but clients do not tend to want help with those, so we will not consider them further here.