Background: The use of online role-play conversations with virtual humans in game-based environments are being increasingly recognized as a research-based and cost-effective method for training large geographically dispersed populations in how to effectively change health and behavioral health behaviors. These simulations are being used to train: 1) Pediatricians in how to build intrinsic motivation within children and their parents to reduce childhood obesity, 2) Parents in how to talk with their children to reduce teenage drinking, 3) Primary care practitioners in how to better identify and manage the care of patients where underlying psychological trauma is driving physical symptoms, 4) Educators in identifying and talking to students in psychological distress including depression and suicidal ideation and if necessary, effectively referring them to support services, and 5) Family members in how to support their veteran's transition home, and if necessary, make a referral for post-deployment stress. In all these applications users learn how to manage challenging conversations by engaging in role-play conversations with emotionally responsive virtual humans that have memory and personality and will react like real people. It is by practicing these role-plays, and receiving ongoing feedback from a virtual coach, that learners gain the skill and self confidence to effectively manage similar conversations in real life.
Program background: This presentation will overview why virtual humans are preferred over real humans in training simulations, the neuroscience underlying the effectiveness of virtual role-plays and an overview of the development process – from needs analysis to how evidence-based practices such as motivational interviewing are woven into the fabric of these programs. The role-play simulation technology will also be demonstrated.
Evaluation Methods and Results: Longitudinal study results that examine training efficacy will be presented including meta-analytic data from over 8,700 participants that completed one of five separate Kognito online role-play simulations to train college educators, students and K-12 school personnel to identify and help students in psychological distress. Composite effect size for pre- and post-training attitudinal measures of preparedness, likelihood and self-efficacy to engage in target behaviors was large at 0.72. The effect sizes comparing pre-to the follow-up for preparedness was 0.70, likelihood 0.35 and self-efficacy 0.42. Changes in actual target behaviors, including increases in the number of distressed students participants were concerned about, approached and referred to support services was 0.21.
Conclusions: Data from this meta-analytic study makes a strong argument that the use of online game-based role-play simulations that employ virtual humans that react like real people can potentially have a significant impact on changing a wide-range of health behaviors.
Implications for research and/or practice: This new and innovative role-play technology has tremendous potential to engage large numbers of people in addressing significant public health issues. Whether it is parents talking to their child about sex, drugs or alcohol use; health care professionals motivating patients to comply with treatment guidelines, or high school teachers detecting and handling bullying - role-play games offer significant advantages over traditional messaging media because they engage users in hands-on, best-practice experiences to bring about health-driven behaviors.