Paranoia, exaggerated fears about threats, is much more common among the general population than was previously thought and just as commonplace as anxiety and depression, according to a study.

Until now, researchers have been unable to study paranoia in lab settings, relying instead on questionnaires.

Now, a team led by Daniel Freeman of King's College, London, has developed a computer simulation program to study paranoid thoughts.

Wearing virtual reality headsets, 200 volunteers walked around a virtual London underground carriage in a four-minute journey.

The carriage contained neutral computer people that breathed, looked around, and sometimes met the gaze of the participants. One read a newspaper, another would occasionally smile if looked at.

Participants came out with interesting responses.

"There was a guy spooking me out - tried to get away from him. Didn't like his face. I'm sure he looked at me more than a couple of times, though I might be imagining it," said one.

"A girl kept moving her hand. Looked like she was a pickpocket and would pass it to the person standing opposite her," said another.

"Felt trapped between two men in the doorway. As a woman I'm a lot more suspicious of men. Didn't like the close proximity of the men. The guy opposite may have had sexual intent, manipulation or whatever," said yet another.

The results of the study have been published in the latest issue of the British Journal of Psychiatry.

"Paranoid thoughts are often triggered by ambiguous events such as people looking in one's direction or hearing laughter in a room, but it is very difficult to recreate such social interactions," said Freeman.

"Virtual reality allows us to do just that, to look at how different people interpret exactly the same social situation. It is a uniquely powerful method to detect those liable to misinterpret other people," he added.

Freeman believes that paranoid thoughts are more likely to develop in settings such as on public transport, where people can feel trapped and observed, and can't hear what others are saying.