Moving beyond cognitive behaviour therapy

Paul Gilbert with the second contribution to the opinion special
Cognitive behaviour therapy (CBT) has recently emerged as a dominant paradigm in psychotherapy, and is influencing training and even our professional organisation (e.g. via graduate workers). It is useful therefore to stand back and consider how CBT sits in the far larger landscape of clinical psychology. Cognitive therapy (CT) developed from the ego-analytic traditions of the 1940s and 1950s along with the emergent interest in how humans ‘construct’ reality (Kelly, 1955).

Cognitive behaviour therapy (CBT) has recently emerged as a dominant paradigm in psychotherapy, and is influencing training and even our professional organisation (e.g. via graduate workers). It is useful therefore to stand back and consider how CBT sits in the far larger landscape of clinical psychology. Cognitive therapy (CT) developed from the ego-analytic traditions of the 1940s and 1950s along with the emergent interest in how humans ‘construct’ reality (Kelly, 1955).

Cognitive behaviour therapy emerged from a not entirely happy marriage in the 1970s of CT and behaviour therapy (BT). Today CBT uses Socratic dialogues, guided discovery, behavioural experiments, exposures to the feared and avoided, psycho-education and teaching the skills of self-monitoring, self-reflection and self-change. CBT is not about correcting faulty thinking as is often simplistically suggested in the press; rather it is about helping people understand how they have become trapped by their attention, reasoning and safety-seeking strategies and how to find ways to develop out of those traps. The focus is on helping people find and distinguish helpful ways of thinking and behaving – not just accuracy. Thinking one will die if one falls whilst trying to escape from the top floor of a burning house might be ‘true’, but is not a helpful focus.

Whilst BT was derived from the physiological studies of learning going back to Pavlov 120 years ago, and the later operant work of Thorndike and others, CT was never based on a sophisticated theory of mind but on a set of useful heuristics, such as automatic thoughts, core beliefs and schemas.

Its therapy was a straightforward focus on eliciting and testing out one’s interpretations and trying new things.
It had two key strengths. In part because of opposition from psychodynamic and other approaches that were contemptuous of its apparent simplicity and ‘conscious mind focus’, it needed to prove its value with outcome research, and so strove to do so. Second, it was adopted by many clinical psychologists who brought their ‘science’ to the study of psychological processes in psychopathology, and over the years welded psychological science into CBT. However, clinical psychologists’ commitment to CBT has been both a strength and a problem – a problem because, as I will note, CBT is a model for therapy, not a science of mind or a service model.

Since the rise of CBT has stirred debate let us clarify a couple of things. The National Institute for Health and Clinical Excellence (NICE) endorsed CBT because of its research base developed over the last 30 years, although of course there are many disputes about the values of methodologies used in this endeavour, especially randomised controlled trials. However, with the exception of some anxiety disorders, it is not true, as some critics suggest, that NICE guidelines support the superiority of CBT over all other interventions. In the depression guidelines, for example, most comparisons have the statement ‘insufficient evidence’ next to them. Most acceptable comparisons on CBT for depression are with drugs or waiting list controls, not other therapies. NICE calls for more research on all the therapies. It is also untrue that CBT has little interest in the therapeutic relationship, or engaging with peoples’ backgrounds. CBT has constantly stressed the value of a comprehensive assessment and attending to the needs and complexities of the therapeutic relationship (Gilbert & Leahy, 2007). One cannot treat trauma or sexual abuse, for example, without careful understanding and working with backgrounds and memories, and a safe relationship.

However, some supporters of CBT play down the huge variation and complexity in process models that clinical psychologists have developed over the last 30 years and introduced into CBT. For example dialectical behaviour therapy, schema therapy, acceptance commitment therapy, mindfulness, radical behaviour therapy (just some of many), although quite different, all fall under the general umbrella of CBT. Moreover, all over the world, for the full range of psychological disorders (e.g. psychotic, mood and behavioural disorders), psychologists are studying complex psychological processes including: different types of memory systems, affect processing, the importance of non-conscious processing, difference between associative and non-associative learning, the influence of attachment relations on brain development and gene expression, and mentalising, to name but a few. Clinical psychologists have spring-boarded from CBT and are developing sophisticated models of mind, combining various aspects of psychodynamic theory with a better understanding of non-conscious processing, attachment theory, somatic memory (i.e. conditioning), in addition to the Eastern psychologies of mindfulness, compassion and much more besides. We are some way from an integrated science of mind, psychopathology and psychotherapy, but clinical psychology will be fundamental to moving us in that direction – provided we stay evidence-based.

This is important because in the understandable rush to bring CBT to as many people as possible, there are concerns. For example, one of the major studies in the NICE depression guidelines that gave CBT its good effect size was of the cognitive behavioural analysis system of psychotherapy (CBASP) (Keller et al., 2000). This is different to standard CBT. It focuses on social cognition, sees a major role for the analysis of transference, and trains people in understanding and engaging in social behaviour that facilitates supportive and helpful relationships. With our keenness on evidence-based therapies we would have anticipated training in CBASP being wheeled out across the country. That didn’t happen.

CBT is not the unitary school its critics or supporters like to make out – there are major differences of view, with major advances in psychological research feeding into CBT all the time. Simplistic versions of CBT do not make this visible or acknowledge this exciting, rich and fundamental research development process – nor the problems looming, such as a potential blurring of the boundaries between CBT and clinical psychology as a science and profession. So let’s explore why we must clearly indicate that as useful as CBT is, it is not a science of mind in the way that clinical psychology is. Below are five themes for thought.

What is cognition?
Cognitive therapists are fond of using the terms ‘cognition’ and ‘information processing’ interchangeably as if they are the same thing. They are not. Your computer, your DNA and your amygdala are information processing systems but they don’t have cognitions. By not defining ‘cognition’ clearly it can mean anything you want it to mean and apply to processing at any point in any organic system. The distinction between a classically conditioned response, an emotional bias, a cognitive bias, and a complex meta-cognition (self-aware reflection) is blurred. The roles of, say, emotional and body memory are not easily located only as cognitive systems. The new trend to try to integrate CBT concepts with research in neuroscience and genetics is laudable and is clearly a challenge for the future. However, it is now clear that the way the brain lays down key pathways, under the influence of genes, developmental and social contexts, and the biases that result, cannot be subsumed only in cognitive terms – unless of course one sees all information processing routines as cognitive.

Motivational systems and attachment
Comparative and evolutionary psychology have revealed that humans, like other animals, are motivated to achieve certain goals, such as attachments to caregivers, group belonging, status and sexual relationships. These motivational systems are guided by emotional, cognitive and behavioural competencies. Forty years ago John Bowlby disputed behavioural claims that development occurred purely on the basis of reward and punishment, and also disputed Klein’s claims of innate destructive impulses in the infant. He suggested instead that infants are orientated to develop (and need) attachments  to primary caregivers for protection and care and have special systems in their brain to help them do that. Research is now showing that the quality of inputs (caregiver accessibility, facial expressions, voice tone, holding, validation, empathy and general emotional tone) significantly influence infant brain maturation (Cozolino, 2007) and subsequent psychology and psychopathology (Mikulincer & Shaver, 2007).

For the most part standard CBT has not engaged with motivational psychology and can struggle with issues of developmental needs and stages. Clinical psychologists, especially developmental psychologists, study development and motivational systems in young children, and should be pushing psychological therapies to take account of these. The idea that only psychodynamic therapies know anything or bother about early childhood should cause ruffles in clinical psychology. We can say the same about many other motivational systems, such as belonging and status.

Intersubjectivity, theory of mind and empathy
There is a link between evolutionary psychology, the science of relationships and their impact on physiological states (sometimes called social neuroscience), and new research looking at intersubjectivity, theory of mind and empathy. We now know that these interpersonal communication processes, by which the mind of one person impacts on the mind of another, play out in complex, dynamic reciprocal interactions. They operate at both conscious and non-conscious levels, and play key roles in self-experience, social behaviour and the regulation of emotion (Decety & Jackson, 2004). For example, recent research has revealed that the brain has mechanisms for emotional attunement called ‘mirror neurons’. If you watch somebody displaying an emotion or experiencing pain, cells in your brain will also fire – as if it were happening to you. There is no ‘cognition’ here; it is a direct activation of brain systems. People with psychopathic disorders and people with autistic disorders may have problems in these systems. Facial expressions and voice tones are stimuli that may well directly impact on mirror neurons and amygdala processes that code for threat and safeness. These are key avenues of research on the therapeutic relationship. So neuroscience is confronting us with some very major processing systems that cannot be classified as (just) cognitive, but clearly have reciprocal relationships with cognitive processes.

The importance of empathy, theory of mind abilities, and how these differ from simpler attributions has been explored with a new therapy called mentalising (Bateman & Fonagy, 2003). Helping people to stand back from their immediate reactions, to think about the intentions and processes in the minds of others, and in their own minds, is a very important focus for therapy (Choi-Kain & Gunderson, 2008). This is an example of how clinical psychologists are building interventions from the science (in this case of empathy and theory of mind).

Social relationships and community
The fundamental approach of clinical psychology is biopsychosocial, placing central importance on social relationships and communities in the regulation of brain states and mental well-being. There is now good evidence that relationships are powerful regulators of a whole range of physiological states and even gene expression. This opens us up to study how relationships, from intimate ones to community and social support, impact on brain states and physical and mental health.

Given that poor mental health is linked to poverty (Melzer et al., 2004) poor community integrations, and the competitiveness of groups (Arrindell et al., 2003), key questions arise as to whether we should be working with individuals at all or with communities (Orford, 2008). The Beacon Project (Stuteley, 2002) was pioneered by health workers who set out to help those with depression and other health problems by working with their whole community – addressing their basic social needs and developing mutual social support systems. There were significant changes in physical and mental health for the whole community, thus showing the benefits of fostering a psychology of mutual support and building social capital. In fact clinical psychologists areat the forefront of such community work (e.g.; see Orford, 2008).

But is it easy to get ‘psychology time’ for such community projects? Not if CBT – or indeed any ‘therapy school’ – gets used as a service plan to stand in for clinical psychology and biopsychosocial models. This is not a critique of CBT, because even if it were the best therapy in the world it is not a service delivery model. But it may be seen in that light and primary care trusts become concerned only to increase access to one-to-one CBT. Businesses like Slimming World, that recognise the huge importance of bringing people together for mutual support, would have gone out of business years ago with such a one-to-one model. Time and again members refer to the support kindness, care and sense of family these groups can provide, in helping them come through difficult times.

There are many ways to deliver living-skills and self-help information, or to de-shame and support people, other than by one-to-one contacts or with computers. We know that depression is often a relapsing condition, therefore we should be talking much more about the delivery of self-help skills within socially supportive systems that people can return to time and again. This is not helped by the restrictive barriers between primary and secondary care, which make developing integrated services very difficult.

Neuroplasticity and compassion-focused therapy
My fifth theme is neuroplasticity and how this is developing a science of brain training. Nearly 3000 years ago the Buddha argued that our minds are chaotic and require training. This training should be regular and guided – and community based (shared and practised in a community). Since the 1960s psychologists have explored various forms of meditation and their impact on physiological and mental states. More recently, mindfulness has emerged under the cognitive umbrella, but has expressly clarified that it is not changing the contents of our cognitions/thoughts but our relationship to them. It involves attention training and metacognition focusing. Recent attention has also been directed to other traditions within Buddhism that focus on compassion training, with research showing that compassion-focused meditations change brain processes (Lutz et al., 2008). Compassion focused therapy (Gilbert, 2000, 2009; Gilbert & Proctor, 2006) was developed as an integration of attachment theory, affective neuroscience, CBT and Mayahana Buddhist practice to help people who are highly shame-prone and struggle to experience feelings of inner warmth and self-soothing. It uses specific exercises to try to stimulate and develop specific abilities. It is not only for psychological disorders but also for physical ones, such as strokes, where psychologists and medical colleagues are exploring how to develop exercises for specific brain areas and patterns. Psychological interventions in the future are going to be much more sensitive to individual variation in physiology and genes, much more orientated to tailoring specific inputs, and brain-training exercises for particular people, and much more socially contextualised.

So much more to offer
CBT has done a good job in getting us to a position where psychological therapies are seen as major players in health care, but there are concerns that CBT (as the name implies) only addresses a specified range of psychological processes. Cognitive processes cannot stand equivalent to all psychological processes. It is vital to clarify why a research focus on motivation, human needs, developmental processes, social relationships and contexts is important to any comprehensive model of mental health and care systems. If you thwart
an individual’s basic emotional and motivational needs, it has major consequences on how the brain deals with those crises of absence.

Clinical psychologists must continue to present biopsychosocial models as a foundation for our science and systems of service delivery. We must do a much better job in working with communities and the voluntary sector in setting up long-term supportive systems for people that will also offer life-skills learning and help. For the more severe cases we all know that more complex therapies are needed, but these are developing fast, often bringing together different schools and with new research on psychological processes (e.g. attachment and emotional memory). Clinical psychologists must also push forward on preventative agendas, because our science has shown so clearly that the early years matter enormously to how our brains mature and that poverty disadvantage, neglect and abuse are breeding grounds for mental distress. CBT is good at what it attempts to do, but (clinical) psychology as a science and service model goes way beyond it.

Paul Gilbert is Professor of Psychology at the
Mental Health Research Unit, Kingsway Hospital,
Derby [email protected]


Arrindell, W.A., Steptoe, A. & Wardle, J. (2003). Higher levels of depression in masculine than in feminine nations. Behaviour Research and Therapy, 41, 809–817.
Bateman, A. & Fonagy, P. (2003). Psychotherapy for borderline personality disorder: Mentalization based treatment. Oxford: Oxford University Press.
Choi-Kain, L.W &. Gunderson J.G. (2008). Mentalization: Ontogeny, assessment, and application in the treatment of borderline personality disorder. American Journal Psychiatry, 165: 1127–1135.
Cozolino, L. (2007). The neuroscience of human relationships: Attachment and the developing brain. New York: Norton.
Decety, J. & Jackson, P.L. (2004). The functional architecture of human empathy. Behavioral and Cognitive Neuroscience Reviews, 3, 71–100.
Gilbert, P. (2000). Social mentalities: Internal ‘social’ conflicts and the role of inner warmth and compassion in cognitive therapy. In P. Gilbert & K.G. Bailey (Eds.) Genes on the couch: Explorations in evolutionary psychotherapy (pp.118–150). Hove: Psychology Press.
Gilbert, P. (2009). The compassionate mind. A new approach to life’s challenges. London: Constable Robinson
Gilbert, P. & Leahy, R. (Eds.) (2007). The therapeutic relationship in the cognitive behavioural psychotherapies. London: Routledge.
Gilbert. P. & Procter, S. (2006). Compassionate mind training for people with high shame and self-criticism: A pilot study of a group therapy approach. Clinical Psychology and Psychotherapy, 13, 353–379.
Keller, M.B., McCullough, J.P., Klein, D.N. et al. (2000). A comparison of nefazodone, the cognitive behavioral analysis system of psychotherapy and their combination for the treatment of chronic depression. New England Journal of Medicine, 32, 342, 1462–1470.
Kelly, G. (1955). The psychology of personal constructs. New York: Norton.
Lutz, A., Brefczynski-Lewis, J., Johnstone, T. & Davidson, R.J. (2008). Regulation of the neural circuitry of emotion by compassion meditation: Effects of the meditative expertise. Public Library of Science, 3, 1–5.
Melzer, D., Fryers, T. & Jenkins, R. (2004). Social inequalities and the distribution of the common mental disorders. Hove: Psychology Press.
Mikulincer, M. & Shaver, P.R. (2007). Attachment in adulthood: Structure, dynamics, and change. New York: Guilford.
Orford, J. (2008). Community psychology: Challenges, controversies and emerging consensus. Chichester: Wiley.
 Stuteley, H. (2002). The Beacon Project – A community-based health improvement project. British Journal of General Practice, 52, Supplement 1, 44–45.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber


While i agree with so much of Paul's work there are issues in what he says:
1) he refers to the importance of the therapeutic relationship - i cant understand why this is seen as a special relationship. everyone creates relationships when working together irrespective of profession or activity. these are built upon trust hope and collaboration. they are not built by the "expert" nor can they be defined by the expert. the only way that the expert can help a relationship be successful is be listening with a constructive ear to the words spoken by the client and responding with words that the client can understand. furthermore, there are therapies in which 25% of all successful interventions are one session - no time for relationship building. Furthermore how can you train therapists in all the nuances of TOM and empathy, for example, and be sure that they are using these hypotheses correctly?

Paul makes the observation that, "One cannot treat trauma or sexual abuse, for example, without careful understanding and working with backgrounds and memories, and a safe relationship." This is an assumption - as it seems to be a belief rather than tried and tested. i have worked with trauma victims for 20 years - i no longer re-traumatise them with endless discussions of the event or events or exposing them to unwanted images etc., i no longer take histories, i focus on what is wanted. I don't build a "relationship" - i ask them what is wanted, as do many of my colleagues (paper due out later this year)

2) i love his comments on cognition! but even if we could define all the different aspects what difference would it make in the room?

3) Attachment - interesting but there also studies that suggest attachment is not significant, e.g. Elizabeth Meins, the Psychologist 2017, “The fact is that there’s no strong evidence for parent–child attachment in infancy predicting anything much about children’s later development. Indeed, Booth-LaForce and Roisman’s definitive 2014 study showed that early attachment doesn’t even predict attachment later in development, let alone all of these other things.”

Sadly, Paul continues to allude to concepts that are not, nor can ever be anything other than hypotheses. There are no axioms in psychology. Indeed, there are very few theorems in the correct and scientific sense of the word, there are many hypotheses that lead to a vast range of different interventions that then create confusion and dilute the direction of progress by sending folk off in a multiplicity of directions. The main difficulty psychology has is that it is shooting at a swiftly moving target with cumbersome, slow moving and unwieldy tools. I have no particular beef with anything he has said, just that all he describes is such complexity that can never be replicated in the therapy room or community. Great for research and maybe for the most highly skilled experts to be aware of - but could never be translated with any real certainty or safety into the world of the everyday.

There are two tools that humans have that can help them with their lives: the first is language (which Paul alludes to but does not pay any real attention to) - as psychologists we should be listening to our communities and clients with real attention and not spend time trying to map their utterances on to hypotheses, as this changes the interaction immediately by defining, in the head of the psychologist, their view of the situation and not that of the community or individual. Then we must ask questions based upon what was said and not what we wanted to hear that fits with our hypotheses.

Second is our imagination. Imagination helps us to build models of the world in our heads. But each person has a different model that we can never be privy to. Furthermore, that model, created by the individual is in constant flux and as Alan Francis rightly points out does not equate to the normal bell curve so loved by psychologists.

If we are to be effective in our work, then it is necessary to engage the imagination of each person or community that we work with in developing a view of the future that is desired and can be moved toward – something to be achieved – usually in small steps. It is entirely illogical to attempt to remove from imagination what is not wanted as most classic psychological therapy does currently. Nature abhors a vacuum and in order to help a person move on from trauma, distress, loss or any other psychological damage (that we all incur as part of our journey through life) it is necessary to help a person or community find a vision that helps them move forward, not simply to try to remove unwanted images or even normalise them through exposure therapy.

Sometimes it is helpful to think about our thoughts in a different way. On this occasion I want you to think about your thoughts as water in a bucket (the bucket is our brain). The bucket is full of water, when we are feeling good the water is sweet and clear. However, when we are feeling down it is like the bucket has been polluted with a bottle of ink, the water is cloudy and dark and unpleasant.
The big question is how do we clear the ink out of the water in the bucket? We can’t empty the fluids out of the bucket, we can’t drill a hole in the bottom and drain it (this our brain remember), we can’t bail it out. So how do we change the water?
The easiest and best way is to pour more clean water in to dilute the dirty water. The more clean water we pour in the less foul water remains in the bucket and eventually the water is good enough to drink again. This is all about attentional bias.

If we are going to be effective as helpers we have to consider first what being helpful would be like and what change it would bring about - not hypothesise our own ideas of what is needed (though research, hypothesing and dissemination is also vital in a general sense) and impose those hypotheses upon those we are working with.

there are two different paradigms here - the one of research and academia whose job it is to explore, disseminate and teach. the other is the practical application of that knowledge - which has to be at a level where everyone can comprehend and make changes based upon their own understanding of the world - not have a model or way of being or thinking imposed upon them because this is the latest psychological fad.
the best we can hope for are simple straightforward interventions that are based upon the lay person's understanding of the world not the understanding of the expert!

After all in evolutionary processes our ancestors did not work out the trajectory, force, weight and dynamics of their spear before creating it. they tried it, experimented and then sat down afterwards with a full belly and worked out how it happened!

We need to think about what works, then use it explore it and develop it. As Paul points out imposing models isn't the last word. Because of the dynamic nature of human development - there never will be a last word!

Steve Flatt