The aim of this paper is to summarize the first 18 months of work at the Cyberpsychology Lab of the University of Quebec in Outaouais on the applications of virtual reality (VR) in the treatment of anxiety disorders. This work is based on the use of modified 3D games to create new and useful immersive environments.

The high cost of VR equipments and environments is a significant deterrent to the widespread dissemination of this technology from the scientific research labs to the office of clinical psychologists. The prospective users of V R could either develop their own environments or buy the few already available, but these options are both very expensive. Developing environments often requires expertise in 3D programming, computer science, engineering, and on the available technology. It is rarely an option for the average clinical psychologist. The other option is to buy already available environments. But the cost is prohibitive (e.g., 10 000$ US for an environment to be used for heights phobias) and the environments cannot be modified to suit the needs of the user (e.g., the properties of the fear of heights program cannot be altered). An alternative solution is to adapt 3D games. For example, Unreal Tournament, a popular but violent 3D game is sold with a game editor. A user with low to moderate computer skills can adapt the environments for many clinical applications.

The clinical usefulness of VR environments in anxiety disorders is based on the fact that exposure to anxiety provoking stimuli is an essential ingredient of any effective treatment for anxiety disorders, including specific phobias. Virtual reality offers many advantages to psychologists who need to expose their patients to a variety of phobogenic environments, such as flexibility, safety, confidentiality and enticement for patients.

Our research team developed three VR environments to treat pathologic fears of heights, spiders and enclosed spaces. Four studies will be presented: a comparison between phobics and non-phobics immersed in the VR environments, two clinical trials on the effectiveness of modified 3D games to treat phobias in adults and one clinical study with children suffering from spider phobia

A total of 60 participants were immersed in VR environments. All participants were selected on the basis of a structured diagnostic interview using standard psychiatric criteria (DSM-IV) to assess the presence or absence of specific phobias. Participants in each clinical samples received a principal diagnosis of : acrophobia, arachnophobia, claustrophobia. Non-phobic participants were selected on the basis of not suffering from any anxiety disorder. Pre and post-immersion assessment tools were different for the participants in the four studies, but all consist of: (a) anxiety and depression questionnaires (Fear Survey Schedule, State-Trait Anxiety Inventory - Y, Beck Depression Inventory), (b) questionnaires specific to the target problem of each sample (e.g., Fear of Spider Questionnaire), (c) measures specific to VR such as the Presence Questionnaire, the Immersive Tendency Questionnaire and the Cybersickness Questionnaire. The treatments consisted of one introductory session and three 90-minute VR exposure sessions.

The virtual environment used for the virtual spiders was developed using the 3D game Half-Life and its editor (Worldcraft). This virtual world contains six rooms with spiders of different size, shape and number. The virtual environments used for the fear of heights and enclosed spaces were developed using the 3D game Unreal Tournament and its editor (Unreal Editor). More than 20 worlds were adapted or created for this study, each with different level of difficulty and various anxiety provoking stimuli (e.g., small or large bridges, buildings, cranes, tunnels, caverns, mazes, diving under water, etc.). The immersions were provided by a Windows based PC (Pentium 3), an ATI Radeon 64 graphics card, an Eye-Glass HMD by I-O Display and an Intertrax2 tracker by InterSence.

In all four studies, presence ratings were very high, as well as participants comments on the graphic qualities and the flexibility of the programs. The comparisons between phobics and non-phobics showed that anxiety plays a significant positive impact on presence and that fearful stimuli can generate a significantly higher level of anxiety among phobics. Results in all three clinical samples demonstrated a significant improvement in their phobias.

To sum up, it is possible to use and adapt easily 3D games to create virtual environments. These environments are relatively flexible and could be modified easily to suit therapists needs. Our results also add to the building body of data confirming the usefulness of virtual reality exposure for phobias for both adults and children.