Patients suffering from depression could benefit from a virtual reality therapy in which they give advice to a crying child, research has found.

The avatars are then switched and the patient is portrayed as the child and hears their own advice repeated back to them.

The study, by researchers from University College London and ICREA-University of Barcelona, is published in the British Journal of Psychiatry Open and was funded by the UK's Medical Research Council.

A trial was said to have helped nine of 15 NHS patients aged between 23-61, four of whom reported “a clinically significant drop in depression severity” after three 45-minute sessions.

Prof Chris Brewin, lead author of the study, said the results were promising and patients had described the experience as "very powerful".

He said he believed the effects of the treatment could last for up to a month.

“People who struggle with anxiety and depression can be excessively self-critical when things go wrong in their lives,” he said.

“In this study, by comforting the child and then hearing their own words back, patients are indirectly giving themselves compassion. The aim was to teach patients to be more compassionate towards themselves and less self-critical, and we saw promising results.”

The treatment first requires the patient to wear a headset that projects a life-sized image of themselves through a virtual reality mirror.

They are asked to identify with the avatar, which replicates their movements in a process known as “embodiment” before the image of a small crying child appears alongside them.

When told to try and comfort and console the child, patients reportedly asked them to think of a time when it was happy and to think of someone who loved them.

The roles were then reversed and the patient was embodied into the avatar of the child before hearing the same advice repeated back to them in their own voice.

Prof Mel Slater, co-author of the study said: "We now hope to develop the technique further to conduct a larger controlled trial, so that we can confidently determine any clinical benefit."

The researchers have said the trial so far is too small to prove conclusively that the therapy will be useful. They now hope to develop the technique further and carry out a larger trial to establish the benefits.