The main schools of therapy, and how they differ
CBT, person-centred, gestalt, psychodynamic, existential and transactional analysis explained plainly — what each believes the problem is, what a session looks like, and where the evidence sits.
Schools of therapy differ less in what happens in the room than the names suggest, and more in what each thinks the problem is. That is the useful axis, so it is the one used here.
Cognitive behavioural therapy
The problem, as CBT sees it: unhelpful patterns of thinking and behaviour that keep a difficulty going, whatever originally started it.
In the room: structured and collaborative. An agenda, a shared model of what maintains the problem, and tasks between sessions — testing a prediction, keeping a record, gradually approaching what is avoided. Usually time-limited, often twelve to twenty sessions.
Where the evidence sits: the strongest, by a distance, because CBT is manualised and therefore trial-friendly. NICE recommends it across depression and the anxiety disorders. Its practitioners are accredited through BABCP, which maintains the CBT Register.
The honest caveat: being the most-studied is not the same as being the most effective for everyone, and the structure that makes CBT researchable suits some people much less than others.
Person-centred therapy
The problem: a person cut off from their own experience by conditions others placed on their worth.
In the room: the practitioner offers empathy, genuineness and unconditional positive regard, and largely follows rather than directs. Few techniques, no homework, no agenda. Its founder, Carl Rogers, argued that those conditions were not preliminaries to the work — they were the work.
Where the evidence sits: counselling of this kind is offered within NHS services for depression, though the trial literature is thinner than CBT’s.
Gestalt therapy
The problem: awareness that has become blocked — feelings, needs and reactions kept out of view, so that a person acts on old patterns without noticing.
In the room: present-tense and experiential. Attention to what is happening now, including posture, breath and what is passing between client and practitioner. Gestalt is where the two-chair experiment comes from. It is a humanistic therapy, holding that people are not fixed entities stuck with their difficulties.
Where the evidence sits: a modest empirical literature; it is well established as a training tradition, and taught within UKCP’s humanistic and integrative colleges, but it has not been trialled at anything like the scale of CBT.
Psychodynamic therapy
The problem: patterns formed in early relationships, operating outside awareness and repeating in present ones.
In the room: less structured. What comes to mind, what recurs, dreams sometimes, and particular attention to how the relationship with the practitioner reproduces the pattern — transference. Can be brief and focused or open-ended.
Where the evidence sits: short-term psychodynamic therapy has been trialled for depression and appears in NHS provision as dynamic interpersonal therapy. Long-term open-ended work is much harder to study, and the evidence is correspondingly weaker — which is not the same as evidence of ineffectiveness, but should not be reported as if it were strength.
Existential therapy
The problem: not pathology but the conditions of being alive — mortality, freedom, isolation, the need to make meaning. Distress is treated as an intelligible response to those, not a fault.
In the room: philosophical and exploratory, less concerned with symptom reduction than with how a person is living.
Where the evidence sits: very little controlled research. It is a recognised UKCP modality with a substantial literature, but somebody choosing it should know that the case for it is philosophical rather than empirical.
Transactional analysis
The problem: habitual patterns of relating, described through accessible models — Parent, Adult and Child ego states, and the recurring “games” people play without meaning to.
In the room: contractual and explicit, with jointly agreed goals. The vocabulary is deliberately plain, which is part of why TA travelled so easily into training and organisational work.
Where the evidence sits: limited controlled research; a strong practitioner tradition.
Integrative and pluralistic practice
Many UK practitioners describe themselves as integrative: trained principally in one modality and drawing deliberately on others as the work requires. Done well this is responsiveness. Done badly it is a way of avoiding saying what is actually being offered.
The reasonable question is not “are you integrative?” but “what is your core training, and what are you drawing in from where?”
Which to choose
There is no clean answer, and anyone offering one should be treated with suspicion.
Where NICE recommends a specific therapy for a specific condition — CBT for the anxiety disorders, for instance — that recommendation is a good default and worth asking about. Beyond that, the consistent finding of the outcome literature is that differences between well-conducted therapies are modest, and that the fit between client and practitioner is not a minor variable. Choosing a person you can talk to is not a soft criterion.