by Corey Mead
A few weeks afterward, Rizzo received a phone call from Russell Shilling of the Office of Naval Research. Shilling was a former member of Mike Zyda’s MOVES team at the Naval Postgraduate School; he had been both sound designer and principal investigator on America’s Army. Given his longstanding interest in advanced technologies, he needed no prodding from Rizzo to grasp Virtual Iraq’s potential. Though the Office of Naval Research did not traditionally fund clinical research, Shilling managed to get his hands on some money, and in March 2005 he delivered the funding Rizzo needed to continue his work.
Rizzo began conducting feedback sessions, in Iraq and at the Naval Medical Center in San Diego, with psychologists, military personnel, and veterans. Hundreds of soldiers and veterans also provided feedback on the USC campus. Following these sessions came a clinical trial featuring active-duty soldiers. At the end of the trial, 45 percent of subjects no longer had PTSD, while an additional 17 percent showed improvement in their symptoms—a better-than-average rate in treating post-traumatic stress disorder.
Rizzo envisioned Virtual Iraq as benefiting clinicians as well as patients. “To be a good exposure therapist, you’ve got to have some kind of imaginative skill,” he says. “You have to help guide and be comfortable with some of the hard things that people recall from the imaginal narratives. With virtual reality, you’ve got a very structured tool for pacing the exposure. So if you’re not such a great imaginative exposure therapist, you can become a very good one using this tool.”
The third benefit of the program, Rizzo felt, was that it could make mental health treatment less stigmatizing for soldiers. “That’s always been one of the pitches. [Maybe they’ll] say, ‘Oh, it’s kind of like a video game; I’ll give it a try.’ It’s still hard therapy, but it’s in that context. Soldiers that have gone through the treatment all say the same thing: ‘I’m glad they put it in this format. It’s better than having a doctor try to pull it out of you.’” A 2008 survey of more than three hundred active-duty service members showed that one in five people who categorically refused to try traditional therapy would be willing to try Virtual Iraq.
The Monster in a Box
When Jerry Della Salla began using Virtual Iraq at the Manhattan VA hospital, he and his therapist, Michael Kramer, were embarking on a relatively uncharted course. Still, the shape of their sessions matched what has become the program’s standard method for use. Based on his previous discussions with Della Salla, Kramer ordered up a convoy operation as the basic scenario for Della Salla to work through. Because the treatment involves gradually raising the patient’s stress level, over the course of several sessions Kramer added details that increased both the realism and the intensity of Della Salla’s experience. He added explosions and gridlocked roadways and bloody soldiers and civilians. Using the program’s scent machine, he mixed the smell of diesel from a Humvee with the odor that emanates from a fired weapon. He added details from the Battle of Abu Ghraib, particularly the deafening whistle of incoming mortars and RPGs, two of the sounds that continued to haunt Della Salla. As these details were added, Della Salla narrated his experiences over and over again, until the combination of sights, smells, sounds, and his own memories made him want to tear off his head-mounted display and run from the room.
When Della Salla reached these points of ultimate stress—usually about thirty minutes into a session—Kramer would feed him strategies for working through his ragged emotions. He would then coax Della Salla through yet another convoy scenario, urging him not to stop midway. (Because the patient controls the joystick, he controls the speed and length of a scenario.) According to Della Salla, “As long as you’re communicating in the process and allowing [the therapist] to bring you down from any sudden moments of panic, that allows you to see what’s going on with your body, with yourself, and then to have the strength and the ability to just slow it down.”
As the weeks went by, Della Salla found himself gaining confidence in his ability to make it to the end of a scenario. The difference was subtle at first, but over time he began to feel less as if his emotions were controlling him and more as if he could control his emotions.
This shift points to a key issue with PTSD, as well as with its treatment. Patients like Della Salla often feel at the mercy of their own bodies and minds, and the fear generated by this inability to regulate their responses can sometimes be as crippling as their traumatic memories. For Della Salla, learning that he was no longer helpless in the face of his disorder started him on the path to recovery. Recovery is different from a cure, of course. The traumatic memories didn’t go away; they simply lost some of their power over him. “PTSD is this monster in a box,” Della Salla says. “And you have to respect it enough to know how to control it, because at some point it has control over you. And it’s not your fault. Once you make the decision to accept that, you can maybe, hopefully, learn how to put it in its place.”
Della Salla says that his discussions with Kramer over the first year of treatment prepared him for what Virtual Iraq offered: a more intense, immersive level of therapy in which he could more easily make progress. The program enabled him to work on issues that he would not have confronted otherwise. Without it, Della Salla says, he wouldn’t be where he is now. “What I like to believe is that you do get stronger. You get more confident. That’s probably the best word. But you shouldn’t let your confidence be the thing that puts your guard down.”
“I would be the first person to say that technology doesn’t fix anybody,” Skip Rizzo admits. “Technology’s just a cold tool that can help the therapists do their job better. And that’s what I think Virtual Iraq/Afghanistan is. And I think that’s what the power of simulation technology is. The human brain has this great propensity to suspend disbelief when we’re watching a TV show, or a movie or a play or whatever. Well, simulation technology can do it in a more comprehensive and relevant way, and in a very controlled fashion.” When PTSD sufferers begin to gain control over their memories in therapy, they embark on a process that psychologists refer to as “extinction.” “They’re reliving the memories,” Rizzo says, “but there’s no real bad thing objectively happening. Yes, there are bad memories, but memories can’t objectively hurt you in your current life. They can haunt you a little bit, but that’s where the emotional processing, by telling the story, occurs.”
According to the Journal of CyberTherapy and Rehabilitation, more than fifteen studies entailing diverse populations have shown that virtual-reality exposure therapy enhances traditional cognitive behavioral treatment regimens for PTSD. The journal reports that most studies reveal a treatment success rate of 66 to 90 percent. An issue of Studies in Health Technology and Informatics, meanwhile, summarized case studies from a navy-funded project comparing the effects of VRET with the effects of traditional treatment on active-duty navy corpsmen, Seabees, and navy and Marine Corps support personnel. The results showed that VRET led to measurable reductions in reported symptoms of depression, anxiety, and PTSD.
Most recently, Military Medicine reported on a treatment project aimed at developing and testing a method for applying VRET to active-duty service members diagnosed with PTSD. Forty-two service members were enrolled; twenty of them completed treatment. Of those twenty, 75 percent experienced at least a 50 percent reduction in PTSD symptoms and no longer met accepted medical criteria for PTSD after treatment. On average, PTSD scores decreased by 50.4 percent, depression scores by 46.6 percent, and anxiety scores by 36 percent. Analyses showed that statistically significant improvements in PTSD, depression, and anxiety occurred over the course of treatment and were maintained at follow-up.
Crucial to the growth of VRET, of course, is the willingness of clinicians to use it. A recent issue of Psychiatric Services presented the results of a study gauging Veterans Health Administration mental health clinicians’ perceptions of virtual reality as an assessment tool or a component of exposure therapy. Although the study demonstrated that the use of virtual reality as a therapy wa
s feasible and acceptable to clinicians, it also showed that successful implementation of the technology as an assessment and treatment tool will depend on consideration of the helps and hindrances, bureaucratic and otherwise, in each given setting.
Given the positive track record of the initial studies on VRET and PTSD, the military would do well to remove any barriers to continued implementation and expansion of the therapy. For soldiers like Della Salla, VRET may be key to managing recovery. Marine Master Sergeant Robert Butler, who spent a year in Iraq and returned home with the classic symptoms of PTSD, acknowledges that VRET “is tough. It’s tough. Because you’ve spent so much time trying to avoid thinking about your deployment, and they’re dredging up these memories that you tried to avoid at all costs. It’s difficult.” Still, Sergeant Butler says, “I think it was a great idea for them to put treatment in that [virtual-reality] format. Better probably than just sitting there and having some doctor try to pull the events out of you. You’re right there. Boom. Smack. Face-to-face with your worst demons.
“I mean, am I a hundred percent better? No, I wouldn’t say I’m a hundred percent better. But I do have my life back. I’m able to do a lot of the things that I did before . . . I’m not running around angry all the time. This treatment was ... it saved my life, probably. It saved my marriage, for sure. So if you asked me if it works, I would say, ‘Yeah, it works.’”
Wars of Innovation: Medical Virtual Reality in the Twenty-First Century
“War sucks . . . but it does drive innovation.” So reads the PowerPoint slide with which Rizzo often begins his public presentations. As he points out, these innovations have had a major impact on civilian health care and mental health rehabilitation. Indeed, as much as I have emphasized the military’s influence on education, it has had an equal or greater impact on American health care, including the development and growth of the field of psychology. “It’s a sad commentary on humanity, but it’s a reality,” Rizzo says. “The military is out front.”
When Rizzo received his initial funding for Virtual Iraq from the Office of Naval Research, he realized that the project had the potential to branch out into areas of health care beyond the treatment of PTSD. He approached Randall Hill, the Institute for Creative Technologies’ executive director. “Virtual Iraq could be a good start-off,” he told Hill. “I think I’ll be able to grow a research program, because the military’s going to want more than just PTSD treatment.” Hill agreed, and Rizzo officially joined the institute on October 1, 2004. In the intervening years, true to his word, he has developed not only Virtual Iraq/Afghanistan but a series of other projects that use video game technologies in the service of veterans’ health care.
Today Rizzo oversees four affiliated labs, known collectively as the ICT MedVR Lab, all of which grew out of Virtual Iraq. He conceptualized and orchestrated these labs, then brought other people in to run them. (As a manager, Rizzo now spends most of his time writing grants and papers and networking.) The Virtual-Reality Psychology Lab focuses on PTSD, stress resilience, and pain distraction work; the Motor Rehabilitation Lab addresses physical rehabilitation; the NeuroSim Lab focuses on neuropsychology; and the Virtual Patient Simulation Lab leverages the ICT’s decade of work in developing virtual humans, which Rizzo’s team translates into clinical tools.
The most developed of these clinical tools is a Web-based application called SimCoach, in which virtual humans act as advisers and sounding boards for members of the military community who may be suffering from such issues as depression, stress, substance abuse, suicidal ideation, brain injuries, and relationship difficulties. It is a preintake tool, as opposed to a form of online therapy. One of the primary goals of SimCoach, which cost $10 million to develop, is to break down the barriers to care that have traditionally dogged the military—most prominently, the belief among soldiers that they will be judged negatively if they admit to needing help. A recent Mental Health Advisory Team study of soldiers in Afghanistan found that over 50 percent believed that they would be seen as weak if they sought behavioral health care, while 34 percent believed that seeking help would harm their careers. Because users interact with SimCoach anonymously, Rizzo and his team are hoping that soldiers will be inclined to use it. Though it features the same basic content as WebMD, the use of virtual characters to relay that content adds a social dynamic that is intended to engage patients and keep them involved as they gather information.
There are other barriers to health care for military personnel, including accessibility and availability—whether there are enough health-care providers in a given location and whether that location can be easily reached by service members in need of help. The American Psychological Association’s Presidential Task Force on Military Deployment Services recently declared itself unable “to find any evidence of a well-coordinated or well-disseminated approach to providing behavioral health care to service members and their families.” Because SimCoach is a Web-based application, it is available anywhere at any time, which will address at least some issues of accessibility and availability.
One bright October morning, I sat in a sunlit conference room at the Institute for Creative Technologies as Rizzo presented SimCoach to a group of faculty members and doctors from USC’s Keck School of Medicine and the Los Angeles County Hospital. The meeting was part of a push by Rizzo and his team to promote and expand SimCoach and its related applications—by having them taken up first by the broader USC community and then by the nationwide medical community. Rizzo had even dressed up for the occasion: he was wearing black jeans and a black button-down shirt and had pulled his hair back in a ponytail.
As the presentation began, we were introduced to Bill Ford, one of SimCoach’s virtual characters. Bill is a gray-haired Vietnam veteran, a white man with a southern accent who talks in a cloyingly folksy manner. Wearing a long-sleeved gray shirt and blue jeans, Bill sits at a wooden table on the back porch of a house, fields and trees spreading into the distance. (SimCoach looks like a video game.) With a cup of coffee near at hand, Bill starts things off by saying, “The suits want me to explain to you that I’m not real.” Once this detail is out of the way, Bill talks about a soldier friend of his, Jared, who has recently returned from Afghanistan and Iraq and who is having a number of problems now that he’s back home. As Bill discusses Jared’s various issues—he’s depressed, he’s fighting with his family, he’s having flashbacks—questions appear at the bottom of the screen. “Are you experiencing similar issues?” the first question asks. (Soldiers using SimCoach answer this question and similar ones, all of which are designed to provide an outline of what they might be struggling with.) Eventually Bill stops talking about Jared. “All right,” he says to the user, “from what you’re telling me, it looks like you’re
having flashbacks
upset at memories
avoiding things that might trigger memories
Is that right?”
Depending on how the user responds, Bill suggests various resources that might be helpful: articles, websites, video testimonials, support groups, lists of local providers. “Here’s a link to people in your area,” he might say, and follow that up with “Do you have any idea of what to look for in a counselor?” If the user responds in the negative, then Bill provides additional guidance.
Though Bill Ford is somewhat cheesy, he is also undeniably helpful. The question for Rizzo and his team is whether soldiers will suspend their disbelief when they interact with Bill and the other SimCoach characters. (Rizzo is currently running a study to determine the answer to this question.) Previous studies have shown, surprisingly, that graphics are not the most important element when it comes to virtual humans. Instead, character movements (hand gestures, head nods, and so on) are the key to establishing rapport. Studies have also shown that people respond to virtual humans in the same way that they respond to real humans, simply because we don’t know any other way to react.
Rizzo’s team has also developed virtual humans that will be use
d to train therapists and social workers in dealing with military populations. USC’s School of Social Work features a pioneering concentration in military social work, intended to prepare students for the special needs of the nation’s military personnel. At the same meeting at which Rizzo presented SimCoach, Patrick Kenny, the director of the ICT’s Virtual Patient Simulation Lab, showed off his lab’s latest creations, which the School of Social Work will soon employ. Kenny began by describing the limitations of using live actors, who have traditionally been used to train budding social workers. To begin with, he said, live actors require a great deal of training before they can inhabit their roles in a believable fashion. The high rate of turnover among live actors is also a negative, as is their lack of availability on a 24/7 basis. Child actors in particular, Kenny said, can be quite difficult to find.
By way of contrast, Kenny is developing a series of standardized virtual patients who don’t suffer from any of these real-world faults. He shows off Alamar Castilla, the lab’s most developed virtual patient. Castilla, the story goes, has just returned from his third deployment overseas and is suffering from PTSD. Fidgety, angry, and clearly uncomfortable in a therapeutic setting, Castilla requires a subtle approach on the part of a mental-health-care worker. Kenny shows us a worst-case scenario—the student clinician asks a number of inappropriate questions, including, most glaringly, “So, did you ever kill anyone?” “Fuck this, I don’t need this shit,” Castilla responds before storming off in a huff. The simulation is hardly perfect: Castilla’s mouth movements don’t match his words, and his voice is disconcertingly robotic. Because voice recognition software is less than ideal, Castilla and the other virtual humans sometimes offer incorrect responses. At this point, too, the questions that Kenny has designed are exclusively closed-ended, meaning that they are directed to very specific responses; open-ended questions are much more difficult to program. These are all issues that Kenny and his team are still working on.
