Full content for this article includes other and illustration.

Source: Computer Graphics World, Dec 1997 v20 n12 p52(6).

Title: Virtual therapy nets real results. (virtual exposure to fears helps to overcome phobias) (Technology Information)
Author: Diana Phillips Mahoney

Abstract: Mental health professionals are employing virtual reality (VR) technology to treat mental disorders, such as phobias. Georgia Tech's Graphics Visualization and Usability Center first used VR to treat acrophobia, the fear of heights. Researchers have also had success in using VR to treat people with a fear of heights, or spiders. The use of VR for phobia desensitization provides a practical and controllable means of therapy for these disorders.

Acrophobia treatment typically requires repeated exposure to incremental increases in height. This is a time-consuming and expensive process that is impractical in real-life situations. A virtual environment is much easier to deal with than reality, and more effective than merely asking the patient to visualize the objects or situations they fear. VR also show promise for teaching autistic children to cope with their environment.

Subjects: Virtual reality - Usage
Phobias - Innovations

Electronic Collection: A20101039
RN: A20101039

Full Text COPYRIGHT 1997 PennWell Publishing Company In the treatment of mental health conditions ranging from phobias to behavioral disorders to cognitive disabilities, virtual reality is opening therapeutic avenues not available in the real world Because exposure to virtual spiders ion a virtual world evokes strong emotional reactions in some patients, researchers at the Human Interface Technology Laboratory at the University of Washington are able to use VR-based graded exposure therapy--therapy in which subjects are incrementally exposed to the objects or situations they fear--to treat the patients' spider phobias.

The early promise that virtual reality (VR) would fundamentally change the way we perceive and interact with digital information has yet to be fully realized. The fact is, virtual reality is not very real at all, at least not visually. In all but the highest-end systems, true virtual worlds--those that allow real-time interaction with all of the environments' 3D objects--are simplistic representations of their real-world counterparts.

While this has hindered widespread acceptance and implementation of VR in many areas, there appears to be one area in which it is not prohibitive: mental health. In fact, the technology's growing popularity as a tool for treating mental-health disorders is not so much driven by the promise that VR can mimic the real world as it is by the belief that virtual environments offer therapeutic options not possible or practical in the real world.

One well known mental-health application of VR is a project by researchers in Georgia Tech's Graphics Visualization and Usability Center. In 1995, professor Larry Hodges and behavioral sciences specialist Barbara Rothbaum conducted the first controlled study of virtual-reality exposure therapy in the treatment of acrophobia (fear of heights). In that study, the VR system the researchers developed, which comprised various, height-related environments (a bridge and a glass elevator, for example), was proven an effective means of reducing the height-related anxieties of acrophobic subjects. Since that time, similar results have been seen in the treatment of other types of phobias, such as the fear of flying and the fear of spiders.

Phobia desensitization, it seems, is a perfect match for virtual reality because it provides a more practical, controllable means for facilitating conventional therapy for this type of disorder. The normal treatment for most  phobias is graded exposure therapy, which involves systematically exposing, in measured amounts, phobic patients to the objects or situations they fear. Ideally, this involves real-life exposure to the feared environment, but this can be impractical. For example, for someone with a fear of heights, treatment might involve riding elevators, climbing ladders, crossing bridges, or any number of other situations that would evoke fear in the patient. Each of these tasks would have to be repeated many times as the patient got comfortable with each level of the challenge--obviously a time-consuming process and an expensive one in terms of the therapist's fees.

When real-life exposure is impossible or impractical, therapists turn to visualization therapy, through which patients are instructed to visualize the fear-inducing objects or situations. This approach is fully dependent on the patient's ability to imagine such things well enough to establish a real fear response.

Virtual environments seem to address the disadvantages of both these methods. An example of this can be seen in a recent project by researchers at the University of Washington's Human Interface Technology Lab in which immersive VR was used to treat a patient with a severe spider phobia. For this application, 3D virtual spiders were modeled and placed in Division Inc.'s (San Mateo, CA) Kitchen-World environment. As part of the VR therapy, the patient was first exposed to the virtual spider and eventually encouraged to pick it up and place it in certain anxiety-provoking positions. The experimenters could control the spider as well, either by moving a position sensor, by entering position coordinates into the keyboard, or by implementing pre-programmed spider behaviors. In the environment, the spider's movements were sometimes unpredictable (random jumps, etc.) as is the case in real life. To enhance the sense of presence in some sessions, a furry toy spider was attached to a Polhemus (Burlington, VT) position tracker and registered with the virtual spider.

The patient showed signs of improvement after just a couple of VR therapy sessions, according to HIT Lab researcher Suzanne Weghorst, who designed and implemented the study along with Albert Carlin, a clinical psychologist at the University of Washington Medical Center, and Hunter Hoffman, also an HIT Lab researcher. "Early on, the patient started to report changes in the way she felt about spiders and in the amount of fear she felt. She was no longer having nightmares about them and, after the therapy was finished, she was able to go camping for the first time in years." This is a huge leap for someone who previously was so debilitated by her fear that she would duct tape her doors at night and could only work certain shifts because she didn't want to be around when spiders were awake.

One of the most significant benefits of VR in this type of application is the ability to vary many of the parameters about the phobic stimulus, says Weghorst. "[With VR] you create a hierarchy of fearful stimuli and, based on your interviews with the patients about what makes them scared, you can keep a certain amount of anxiety with each exposure, so you're upping the ante each time."

Another advantage to virtual therapy is that patients are more willing to enter a VR exposure than a real one, says Weghorst, because they know it's VR. "But even though they know it's not real, we still get all the fear response: the sweating, shaking--the kinds of things we get in the real world."

Unlike in other applications, the graphical limitations of VR are not an impediment to success with this type of virtual therapy. For the spider project, says Weghorst, "we were using an evironment that was pretty cartoony. But even in a cartoon world a sense of presence can be achieved, and it seems to be linked more to whether or not the scene updates appropriately as you turn your head around than whether or not it looks real."

Most researchers in this area agree that for virtual therapy to be effective, a multidisciplinary approach is critical. "You want to have a therapist, technical support people, and design people when you're creating environments. If it's handled wrong, VR has the potential to do a lot of damage," says Weghorst. She remembers a particular incident in the spider project that could have had serious consequences. "The first time we moved the spider object into this environment, we had it scaled wrong. We couldn't find it in this kitchen. We looked out the window and there was this football-field-sized spider out there. It was really frightening. It was lucky that we did it before putting the patient in there, or we could have made the fear a lot worse."

Helping Autistic Kids Learn

Another promising application of VR in mental health is its use in teaching environmental coping strategies to autistic children. In 1995, Dorothy Strickland, a computer engineer and then-doctoral candidate at North Carolina State University, undertook a research project to assess whether virtual reality could help autistic children learn to function in the real world. Based on her knowledge of some of the characteristic behavioral disabilities of autistic children, Strickland saw some areas in which VR would have more to offer them than the real world. "Part of the problem for autistic individuals is an abundance of input stimuli. They're flooded in sights and sounds. One thing that a VR headset allows us to do is isolate the environment, simplify it, and bring it back in very slow, controlled pieces that the individual might be able to handle and respond to," says Strickland, who is now at Stetson University in Florida.

To test her hypothesis, Strickland designed a simple environment in which children could learn to recognize an object in a scene--a car moving down a street--and eventually move to and stop at another object---a stop sign. The goal was to teach the children in the study to cross a street alone. While teaching such a task to these kids in a real street environment could be dangerous, a virtual world provides a safe learning environment. Strickland worked in conjunction with therapists and staff from the Treatment and Education of Autistic and Related Communication Handicapped Children Center, a division of the Department of Psychiatry in the School of Medicine at UNC Chapel Hill. She also included the children's parents, siblings, and teachers in the test environment as a way to minimize any fear of the new surroundings. She was particularly concerned that the kids would resist wearing the heavy VR headset, since autistic children typically do not like to wear anything on their heads. However, by watching their siblings use and enjoy the headset, the children were enticed to try it.

The results of Strickland's early study were encouraging. The test subjects were able to track the virtual car, move toward it, and identify objects and colors when prompted. One of the participants moved to the stop sign and stopped. Considering that it often takes years of training for autistic individuals to learn new tasks, the fact that such results were obtained in an hour were significant.

Last summer, Strickland and her colleagues received a National Institutes of Health grant to study whether the initial test would hold up through more controlled research. "We basically went back and did something a little less glamorous than street crossing. We did object identification. There's been a lot of work done with children of autism in this area, and it's still a real problem," says Strickland. While the first study was done on a high-end VR system from Division, the second one was created using an off-the-shelf PC, free software downloaded from the Internet, and a rented headset. Because the nature of the application required only the simplest environments--a black-and-white environment with, say, a bright yellow pencil or some other object--this solution was sufficient, says Strickland.

The second phase of the study ended in September and the final results were still being analyzed as of press time, but initial results indicate the technology was an effective tool for helping the children learn. "It's exhilarating to see the effects when we actually take a technology out there and try it with real people," says Strickland.

The next phase of the research will include determining whether the children's success in identifying objects in the virtual environment can generalize to objects in the real world. She also hopes to be able to combine the object identification with teaching the use of the object. "For example, if they learn what a pan is, then we might have a clip of a pan being used in a kitchen," she says, noting that most systems now allow live feeds from video clips. She will also be studying options for implementing such technology in a classroom setting. "We actually used a classroom-type setting in the earlier studies, but I ran all the tests and did all the programming, so I didn't have it user-friendly. It would be nice to provide some of the advantages of the technology to the teachers in a way they could use it."

In addition to autism, Strickland can imagine VR being a useful tool for other psychological/behavioral applications, such as hyperactivity and attention deficit disorder, in which there's a need to control outside stimuli.

VR in Cognitive Testing and Rehab

Another area in which a virtual world may have more to offer than the real one is in the assessment and rehabilitation of cognitive abilities in patients with brain injuries or other neurological disorders. Albert Rizzo and colleagues at the University of Southern California's Alzheimer's Disease Research Center have developed a virtual environment to test, measure, and possibly rehabilitate such patients' cognitive skills. The value of VR for such tasks, he says, "is that in VR, you have total control of complex stimulus presentation, and you can record just about everything that goes on in the environment."

The current phase of Rizzo's study involves measuring and possibly training mental-rotation abilities. The conventional method for testing mental-rotation ability is via a "paper-and-pencil" test in which patients are presented with two images of block configurations--either the same images or mirror images--and are asked to judge whether the configurations are the same or different. "It's basically a mental imagery kind of thing. Just like you would physically rotate blocks in your hand to match them up, you're actually doing that mentally in your mind," says Rizzo.

Rizzo and his colleagues have developed an OpenGL-based virtual version of this paper-and-pencil test in which the user is able to manipulate one set of blocks and superimpose them onto the other set. "They have to rotate them with their hands and superimpose them within the target stimuli. When they successfully do that within a certain tolerance, a little tone goes off, and they go on to the next one." Such a scenario couldn't easily be duplicated with physical blocks because, for one, the blocks couldn't be superimposed, and setting up the multiple configurations (which in the virtual world get progressively more difficult) would be prohibitively lime-consuming. Another advantage of the virtual system is that it enables the researchers to measure the amount of time it takes a person to correctly superimpose the blocks, which is an indication of the patient's mental-rotation capability.

In addition, the system is programmed to compare the quality of the block rotations. "For each one of these block rotations, we've plotted out a course of perfect movements--the most efficient way for the blocks to be moved and coordinated together," says Rizzo. "Then when the actual behavior occurs, we can compare the perfect trajectory with the actual one, so we can get a better handle on measuring what's going on." Also, he notes, the system is encoded with all the precise angles of rotation. "This lets us measure things like whether the rotation is a depth rotation vs. a picture plane rotation, which is one that rotates to the side. We can see differences in the complexity of the blocks vs. the angle of rotation needed. We can look at visual-field differences, whether the stimuli is presented in the left or right visual field, which might be useful for looking at brain functioning in terms of right/left hemisphere considerations." Such assessments, he notes, can't be made in testing with physical blocks or via the paper-and-pencil tests.

To monitor the relationship between test subjects' virtual performance and their performance in conventional tests, they are first given a paper-and-pencil test, then undergo the virtual exercises, and then return to the same paper-and-pencil test. Initial results have been promising, says Rizzo. "One pilot subject started with a score of 8 out of 40 or the paper and pencil test. After about 25 minutes of using the virtual stimuli, he got a score of 20 out of 40. That's a 150% increase. Mind you, that's only one person, so we can't yet generalize that result."

The mental-rotation test is a perfect fit with the current state of VR technology, notes Rizzo. "This is a primary environment. We're focusing on very basic structures. We're not navigating through buildings or anything like that." Because of the researchers' affiliation with a group at USC called the Information Sciences Institute, the mental-rotation system was developed on an SGI Onyx and is displayed via a Pyramid Systems' Immersadesk, which is a high-end drafting table-like projection system that utilizes 3D shutter glasses. The images appear almost like holograms floating on the surface of the table, says Rizzo. "This is technological overkill," he notes. "This could be developed and run on an under-$10,000 PC system using a headmounted display. We're using the high-end system because these resources were practically handed to us."

Eventually Rizzo plans to expand his research to include other cognitive skills. In addition, he envisions incorporating more elaborate settings. "Maybe we'd have the patient manipulate items in a real-world setting--stacking boxes in a closet, for instance, or any kind of situation where he or she would have to visualize an object in one position and predict or envision its ultimate location in 3D space. That's where the real hard-core useful rehabilitation applications would be."

The Pros and Cons of Virtual Therapy

As can be gleaned from the examples noted above, virtual reality has a lot to offer the mental health community in terms of treatment options for certain conditions. "The neat thing about VR is that the whole therapeutic experience can be more easily modulated by a patient in terms of his or her discomfort as they progress through the therapy," says Richard Bloom, associate professor of political and clinical psychology in Embry Riddle University's Science, Technology, and Globalization program. "Also, it's easier to standardize therapeutic procedures through VR. Hopefully, what's going to end up happening is that you'll have VR therapeutic packages for all kinds of different psychiatric problems."

Before this can happen, however, mental health professionals studying the therapeutic value of the technology need to broaden their perspective to take into account certain negative effects, says Bloom. "When you read the mental health literature on VR technology, whet you see being addressed is a referral question. That is, someone has a problem, like a phobia, [researchers] devise some kind of therapeutic program using VR, and they report what kind of effect if any it has on the problem. They're not really entertaining experimentally the possibility that VR could have a negative effect either on the referral questions or, more importantly, on other areas of the person's psychological functioning, which aren't even being looked at."

Because of these considerations, Bloom urges mental health professionals to look at VR in areas outside their profession. "There's evidence from VR research outside of mental health that a certain proportion of people report negative consequences of their interaction with VR. Some of the more common ones are feelings of malaise, nausea, and physical discomfort. And there are a small number of people who actually have unpredictable effects on their perceptions of reality, how they deal with other people, and their emotional functioning." Bloom stresses that such reports come from only a small number of people, and he does not mt an to suggest that VR is a bad thing. "My objective is to influence researchers to start looking at some of the unintended or unexpected negative consequences of what they're up to."

Most of the VR applications in mental health are still R&D projects. Before virtual therapy can realistically be implemented in clinical settings, there are a number of obstacles that must be overcome. For example, says Bloom, there are many unanswered questions when it comes to the duration of any positive therapeutic effect. "Are you talking about curing people? Do they need booster sessions every once in a while? Does generalization break down in certain kinds of situations? All of that really has to be addressed."

The technology also needs to become more user-friendly, says Dorothy Strickland. "Part of the problem in any kind of commitment to do VR is that it's still not anything you can get off the shelf. The gaming industry hasn't entered the market in this area much, so there isn't the volume, and therefore there hasn't been the kind of product support to allow it to move into a general application." As a result, she says, "in order to do a VR system, particularly something immersive or something that involves anything new or complex, such as tactile feedback, you have to have someone who knows computers well enough to do a customized system. And that's a real limitation."

One answer, according to Suzanne Weghorst, "might be to build a library of environments for specific conditions and provide usable tools for modifying them for people who are Mac and PC literate."

Another impediment to full implementation of the technology has to do with the available interface devices. According to Weghorst, while computing power is not as big an issue as it was maybe two years ago because of faster graphics boards, interface devices have yet to catch up with the computing advances. "Tracking is one of the big issues and another is display technology. For example, while there are some low-end HMDs that arc acceptable for what we're doing, they're skill not ideal in terms of resolution and field of view."

In addition to these considerations, there's the ever-present hurdle of general acceptance in the field. The topic has definitely been a focus of discussion at psychology and psychiatry conferences, and the interest of professionals in these fields is certainly piqued, but will they buy into it? "I think psychologists are more receptive than psychiatrists," says Weghorst. "Psychologists tend to have more of a behavioral model of treatment, so for them it's a natural, whereas psychiatrists still have this medical model of mental health--how could you treat the disease without going into the depths of psychoanalysis and all that." However, Weghorst has encountered a number of psychiatrists "on the cutting edge" and some who are receptive to technology in general. "They're the ones who jump right on this."

Richard Bloom of Embry Riddle University thinks psychiatrists will give the technology whatever attention it deserves. Psychiatrists have a mind-set that says, `Show me results that make me want to learn about your technique.' The fact that there's been a lot published already about the positive consequences of psychiatric applications of VR makes it e aster for mental health professionals to take this seriously" Albert Rizzo of USC believes the technology will sell itself, as long as no unreasonable promises are made. "This is a psychologist's dream, to be able to control an environment and record all responses within that environment. If we don't go off with wild eyes and say that we're going to design all these exotic environments and then disappoint people with poor graphics and flow response, there's no question the field will grow."

Diana Phillips Mahoney is managing editor of CGW.

[Back to Team #2 homepage]