AQA Clinical Psychology and Mental Health: Must Know, Should Know and Could Know Research Studies
The new Clinical Psychology and Mental Health topic is part of Paper 1 in the updated AQA A-level Psychology specification, with first A-level assessment in 2027. Students must understand definitions of mental health and the characteristics, explanations and treatments of phobias, depression and obsessive-compulsive disorder (OCD).
However, knowing the theories is only part of the challenge. To produce effective AO3, students also need research evidence that can support or challenge those theories and treatments.
This guide organises 13 useful pieces of research into three levels:
Must Know – the most useful evidence to learn first.
Should Know – additional research that can strengthen evaluation.
Could Know – evidence that can add depth, sophistication or a counterargument.
These categories are revision priorities rather than official AQA categories. AQA does not prescribe this list of studies. Students could use other relevant and accurately described research. The topic content itself is set out in the updated AQA specification.
It is also important to remember that Beck’s negative triad and Ellis’s ABC model are named cognitive explanations, not individual research studies. Students must learn both as AO1, but they have not been included in this research-study list.
Must Know studies
1. Watson and Rayner (1920): Little Albert
Watson and Rayner investigated whether an emotional response could be learned through classical conditioning. A young child known as Little Albert was initially unafraid of a white rat. The researchers repeatedly presented the rat alongside a frightening loud noise. Albert subsequently showed fear when the rat was presented without the noise, and his fear appeared to generalise to some other furry objects.
This study can be used to support the acquisition stage of the two-process model, because it demonstrates how a previously neutral stimulus may become associated with fear. However, it was a study of one child, the procedure was not standardised to modern research standards, and there are serious ethical concerns. It therefore provides vivid but limited evidence for explaining the development of phobias. The original study was published as Conditioned Emotional Reactions in 1920. (Watson and Rayner publication record)
Exam use: The study supports classical conditioning as a possible explanation for the acquisition of a phobia, but the case-study evidence cannot establish that all phobias develop in this way.
2. Gilroy et al. (2003): graded exposure for spider phobia
Gilroy and colleagues followed people who had received three 45-minute sessions of either live graded exposure, computer-aided vicarious exposure or progressive muscle relaxation for spider phobia. Forty-two of the original 45 participants were assessed at an average follow-up of 33 months. Improvements were found across several measures, although the interaction between treatment group and time was not statistically significant. (Gilroy et al., 2003)
The study is relevant to systematic desensitisation because live graded exposure is a central element of that treatment. However, it should not be described as a straightforward test of the complete systematic-desensitisation procedure: the live-exposure condition was compared with other interventions, and the long-term differences between groups were not clear.
Exam use: The findings suggest that exposure-based treatments can produce lasting improvements, but they do not isolate precisely which component caused the improvement.
3. March et al. (2007): CBT and fluoxetine for depression
The Treatment for Adolescents with Depression Study compared cognitive behavioural therapy (CBT), the antidepressant fluoxetine, and a combination of the two. At 18 weeks, the combined treatment had an 85% response rate, compared with 69% for fluoxetine and 65% for CBT. By 36 weeks, response rates were 86% for combined treatment and 81% for each treatment used alone. (March et al., 2007)
This provides evidence that CBT is an effective treatment for depression and that combining psychological and biological treatments may produce faster improvement. The study focused on adolescents, however, so its findings cannot automatically be generalised to all adults with depression.
Exam use: March et al. support the effectiveness of CBT, particularly when it is combined with antidepressant medication.
4. Nestadt et al. (2000): a family study of OCD
Nestadt and colleagues compared 80 people with OCD and their first-degree relatives with 73 community controls and their relatives. In total, 343 relatives of the OCD group and 300 relatives of the control group were assessed using structured procedures. The researchers found familial aggregation of OCD: the disorder was more common among the first-degree relatives of people with OCD. (Nestadt et al., 2000)
This supports a genetic contribution to OCD. Nevertheless, family members share environments as well as genes, so a family study cannot by itself show that the association is entirely genetic.
Exam use: The increased risk among first-degree relatives supports a genetic explanation, but shared family experiences provide an alternative explanation.
5. Soomro et al. (2008): SSRIs for OCD
Soomro and colleagues reviewed 17 studies involving 3,097 adults with OCD. Across the studies, selective serotonin reuptake inhibitors (SSRIs) reduced OCD symptoms more effectively than placebos over six to 13 weeks. In the studies measuring clinical response, people receiving SSRIs were almost twice as likely to respond as those receiving a placebo. Adverse effects were generally more common with SSRIs, and the review could not establish their longer-term effectiveness. (Soomro et al., 2008)
The findings support the effectiveness of drug therapy for OCD and are consistent with serotonin being involved in the disorder. However, successful treatment does not necessarily prove that low serotonin caused OCD in the first place.
Exam use: SSRIs appear more effective than placebo in the short term, although side effects and uncertainty about long-term outcomes must be considered.
Should Know studies
6. Öst (1989): one-session exposure treatment
Öst reported the outcomes of an intensive one-session treatment for 20 people with specific phobias. Treatment lasted an average of 2.1 hours and combined prolonged in-vivo exposure with modelling. At a follow-up averaging four years, 90% of the participants were described as much improved or completely recovered. (Öst, 1989)
This is often used as evidence for the effectiveness of intensive exposure or flooding. It is important to describe it accurately: the treatment combined exposure with modelling, so the results cannot be attributed to flooding alone. The series also had only 20 patients and did not use a randomised control group.
Exam use: The study suggests that intensive exposure can be rapid and long-lasting, but the mixed treatment and limited design make it difficult to establish cause and effect.
7. David et al. (2008): psychological therapy and medication for depression
David and colleagues randomly allocated 170 outpatients with major depressive disorder to 14 weeks of rational emotive behaviour therapy (REBT), cognitive therapy or fluoxetine. Immediately after treatment, there were no significant differences between the three conditions. At the six-month follow-up, REBT showed a significantly larger effect than medication on one measure, while the advantage for cognitive therapy was not statistically significant. (David et al., 2008)
The study supports the view that psychological therapies can be at least as effective as medication and may have lasting benefits. However, it would be inaccurate to claim simply that CBT was definitively superior to medication in every analysis.
Exam use: Psychological and drug treatments produced similar immediate outcomes, with some evidence of a longer-term advantage for REBT.
8. Lewis (1936): family patterns in obsessional illness
Lewis’s early clinical work, Problems of Obsessional Illness, described patterns within families of people experiencing obsessional symptoms. It is often used as early support for the claim that OCD runs in families. (Lewis, 1936)
The evidence has historical value, but diagnostic criteria, research methods and standards of reporting have changed considerably since 1936. It should therefore be used cautiously and alongside more systematic evidence such as Nestadt et al.
Exam use: Lewis provides early support for familial transmission, but the age and limited methodological detail of the evidence reduce the strength of any genetic conclusion.
9. Menzies et al. (2007): neural evidence for OCD
Menzies and colleagues reviewed neuroimaging and neuropsychological research into OCD and also conducted a quantitative meta-analysis of functional MRI findings. They reported consistent abnormalities involving orbitofronto-striatal circuits, while noting that findings across studies were not entirely consistent and that other brain areas may also be involved. (Menzies et al., 2007)
This supports a neural explanation of OCD, particularly the involvement of the orbitofrontal cortex and connected circuits. However, brain-imaging evidence is usually correlational. Neural differences could contribute to OCD, result from the disorder, or reflect another variable.
Exam use: The review supports orbitofronto-striatal involvement but does not prove that neural abnormalities cause OCD.
Could Know studies
10. Choy et al. (2007): review of treatments for specific phobias
Choy, Fyer and Lipsitz reviewed research published between 1960 and 2005 on treatments for specific phobias, including in-vivo exposure, virtual-reality treatment and cognitive therapy. The review concluded that exposure-based approaches had the strongest evidence base. (Choy et al., 2007)
This strengthens the argument that behavioural treatments can be effective. Because it was a review of different studies, treatments and phobia types, variation between the included research makes a single simple conclusion more difficult.
Exam use: The wider evidence base supports exposure treatments, although effectiveness may vary according to the procedure and the type of phobia.
11. Cuijpers et al. (2013): meta-analysis of CBT for depression
Cuijpers and colleagues combined 115 studies of CBT for adult depression. CBT produced a substantial overall effect, but the estimated effect became smaller when publication bias was taken into account. Higher-quality studies also reported smaller effects than lower-quality studies. (Cuijpers et al., 2013)
This provides both support and a useful counterargument. CBT is effective, but its benefits may have been exaggerated if unsuccessful or less impressive studies were less likely to be published.
Exam use: The large evidence base supports CBT, while evidence of publication bias encourages a more cautious conclusion about the size of its effect.
12. Taylor (2013): candidate genes for OCD
Taylor conducted a meta-analysis of genetic-association research into OCD. The review identified 230 polymorphisms across 113 studies and found associations with several serotonin-related and other genetic variants. Most individual associations were small, supporting the view that OCD is polygenic rather than being caused by one particular gene. (Taylor, 2013)
The findings strengthen the genetic explanation while also showing why searches for a single “OCD gene” are unlikely to succeed. The modest and sometimes inconsistent associations suggest that genes probably interact with one another and with environmental influences.
Exam use: Taylor supports a polygenic explanation in which several genes each make a small contribution to vulnerability.
13. Cromer et al. (2007): traumatic life events and OCD
Cromer, Schmidt and Murphy studied 265 people diagnosed with OCD. Fifty-four per cent reported at least one traumatic life event, and experiencing one or more such events was associated with more severe OCD symptoms, even after several relevant variables were controlled. (Cromer et al., 2007)
This challenges a purely biological account by showing that environmental experiences are associated with the severity of OCD. The study was correlational, however, so it cannot establish that trauma caused the OCD symptoms. Recall of past events may also have been affected by participants’ current symptoms.
Exam use: The findings support a diathesis-stress account in which biological vulnerability may interact with traumatic or stressful experiences.
How many studies should an AQA Psychology student learn?
Students do not need to memorise every detail of all 13 studies. A more effective approach is to learn a small number very securely and then add further evidence when the core material is confident.
For each study, aim to remember:
Who conducted it and when.
What the researchers did.
The key finding.
Which explanation or treatment it supports or challenges.
One limitation or alternative interpretation.
A study should never be dropped into an essay without explanation. The student must make the link back to the question clear. For example:
Research support for drug therapy comes from Soomro et al. (2008), who found that SSRIs reduced OCD symptoms more effectively than placebos. This suggests that SSRIs are an effective treatment. However, treatment effectiveness does not demonstrate that low serotonin originally caused OCD, so the evidence supports the therapy more directly than the biological explanation.
That final sentence is what turns remembered research into developed AO3.
Final revision advice
Start with the five Must Know studies and practise applying each one to an exam question. Once those are secure, use the Should Know and Could Know evidence to introduce greater range, methodological analysis and counterarguments.
Above all, prioritise accurate explanation over the number of researchers named. One well-developed and correctly linked study is more valuable than several names listed without analysis.