> Design Projects > Speakat
A VR Serious Game against Public Speech Anxiety
About the Project
A meta-analysis on serious games for mental health shows effect favoring serious games over no intervention controls in reducing disorder-related symptoms[1], including the evaluation of computerized CBT (CCBT)[2]. By simulating the embarrassing scenarios in an immersive VR environment, the game intentionally and repetitively expose the user under the imaginary fears. Then Speakat will empower the user with corrsponding “magic power” and help user guide himeself/herself out of the dilemma so as to achieve a cognitive change. Unlike conventional thought training, a visualised instant feedback is provided to help user establish new cognition and evaluation more quickly and solidly. The times of use of magic power,however, are not infinitive and an anti-addiction and exit mechanism was thus adopted.
[1] Lau, H.M., Smit, J.H., Fleming, T.M. and Riper, H., 2017. Serious games for mental health: are they accessible, feasible, and effective? A systematic review and meta-analysis. Frontiers in psychiatry, 7, p.209.
[2] Fleming, T.M., Cheek, C., Merry, S.N., Thabrew, H., Bridgman, H., Stasiak, K., Shepherd, M., Perry, Y. and Hetrick, S., 2014. Serious games for the treatment or prevention of depression: a systematic review.
COURSE: Adaptive Serious Game
DURATION: Nov.2019 - Feb.2020
MY ROLE: UX/UI design, Animation
Demo Video
One of the magic powers: Cabbagizer.
By turning audience into cabbages visually, Speakat helps the user to get rid of distraction temporarily.
The evaluation framework
Considering the individual differences, we adopt Single Case Experimental Designs(SCED), a within-subject self evaluation framework inside the game, when the subject is its own control. Within each level of training, user will undergo three speech simulation sessions with similar challenges and the Speakat only assist in the second one(B), leaving the user to face the first and the last one alone(A). During each session, objective speech performance data such as filling words, pace, clearness, etc will be collected. After each speech, user will be asked to evaluate his/her own performance as well. By the end of the training, all data will be displayed as aggregated evaluation.
An aggregated overall evaluation is displayed after each course of training.
The expected experiment result is shown as following graph. The mean/median sores of latter non-intervention session(A1.2) shall be no lower than previous intervention session(A1.1) and could be accepted if not higher than the intervened session(B1).
