Oxford psychologists believe mental images of past and future events may be a powerful route to treating psychological disorders. Georgina Ferry reports

I am visualising an imaginary lemon, 'seeing' the cut surface held close to my face. Dr Emily Holmes, who runs the Experimental Psychopathology and Cognitive Therapy (EPACT) research group at Oxford's Department of Psychiatry, asks me to focus on its colour, texture, even its smell. Then she tells me to squeeze it so that the juice goes in my eye. It's not as bad as the real thing, but there's a definite hint of the distressing sensation real lemon juice would induce.

Holmes is demonstrating part of the training she uses with her human experimental subjects to get them to focus on visual rather than verbal aspects of their mental furniture, and the power of images to invoke emotional responses. With her colleagues in EPACT, she is dedicated to developing a greater understanding of this phenomenon, and to harnessing it in the treatment of patients with a variety of debilitating psychological disorders.

A researcher in psychology - she began her career as an undergraduate in PPP at LMH - Holmes is also a clinical psychologist trained in psychological therapies. While working at the Traumatic Stress Clinic in London, she began to think about the questions posed by disorders in which images play an important part. Most of her clients suffered from post-traumatic stress disorder (PTSD), a condition common among survivors of rape, torture, lifethreatening accidents or armed combat. Their main symptom is intrusive memories of the trauma, or flashbacks. 'Flashbacks come out of the blue', says Holmes. 'You may be getting on with your everyday life and suddenly have a flashback that brings with it an overwhelming sense of current threat. It can be extremely debilitating: people are highly alert to threat in their environment, and tend to avoid things that trigger the flashbacks, such as loud noises or even other people. A woman who has been in a car accident with a red car might then avoid all red cars and loud noises, which could mean that she finds it very difficult to travel to work, and can lead to social withdrawal'.

The therapy employed by Holmes and her colleagues in these cases has helped patients to diminish flashbacks not by avoiding them but by deliberately calling to mind the most distressing part of their memory of the trauma and, where appropriate, 'editing' it. 'For the woman who had been in the car crash, the "hotspot" might be that she thought she was about to die and would never see her son again', says Holmes. 'In a flashback, the same feelings - of being about to die and never seeing her son - would come with the image. So one technique is systematically to establish the opposite: that she survived and will be able to see her child grow up. Restructured, more positive imagery is a useful tool to capture this alternative meaning'.

This technique is just one part of the treatment for PTSD known as cognitive behaviour therapy (CBT), which shows patients how the way they think affects the way they feel, and teaches them that they can gain control over their thinking styles. There is now a large body of evidence that CBT works; it is recommended for PTSD as well as for depression and anxiety by the National Institute for Clinical Excellence, and received a huge boost in funding from the Department of Health in October 2007. Traditionally, it has tended to focus on what patients say about themselves when they make statements such as 'I'm hopeless, I might as well be dead'.

Since she came to Oxford in 2005 as a Royal Society Dorothy Hodgkin Research Fellow, Holmes has been working to develop CBT techniques that focus on images rather than just words. She thinks these techniques might be useful in depressed and suicidal patients as well as those who have suffered from trauma. 'More people in the UK die by suicide than as a result of road traffic accidents', says Holmes. 'If you are suicidal, is it possible that you are mentally simulating harming yourself in the future?' She is beginning to gather evidence that people at risk of suicide do have involuntary 'flash forwards' of this type, though it is difficult to research.

Much of the current work of the lab, set up with grants from the Economic and Social Research Council and the John Fell OUP Research Fund, does not involve patients, but that stalwart of experimental psychology, the healthy student volunteer. One study is trying to understand more about how traumatic memories are laid down and how that influences the likelihood of flashbacks. Participants watched a traumatic film while simultaneously carrying out a cognitive task. 'If you give them a task - such as counting backwards in threes - that interferes with the ability to understand what's happening, it seems to make flashbacks worse', says Holmes. 'Whereas if they have to tap a pattern and hold its image in mind, you consistently see a reduction in the number of intrusions.' While it would take a special kind of coolness to remember to tap out a pattern while involved in a head-on collision, Holmes is curious to know whether such tasks performed after traumatic events might help to dampen down involuntary memory.

At the same time, in what she calls the 'happy lab', her colleagues train volunteers to modify their thinking style or 'cognitive bias' as they interpret ambiguous situations. We all tend to have either a positive or a negative bias in the way we see the world: our glass is either half full or half empty. 'We find that if we get people to focus on the happy resolutions in a verbal way, their bias doesn't change', says Holmes. 'But if we get them to simulate them visually, we get quite a powerful effect.' Her colleagues are now developing computer-based systems that display ambiguous images and encourage subjects to interpret them positively. In one of those happy juxtapositions that Oxford often makes possible, many of these images of ordinary, everyday scenes and objects were contributed by fine art students at the Ruskin School.

'It's really nice that we can use lab paradigms to test out our theories and work out what might be the most effective tasks before we take them back to the clinic', says Holmes. Her own glass, it seems, is not just half full, but brimming over.

Georgina Ferry, former editor of Oxford Today, is a science writer and biographer, whose book Max Perutz and the Secret of Life was published last year.